A dendritic nano-sized hexanuclear ruthenium(II) complex as a one- and two-photon luminescent tracking non-viral gene vector

Fluorescent tracking gene delivery could provide us with a better understanding of the critical steps in the transfection process. However, for in vivo tracking applications, a small diameter (<10 nm) is one of the rigorous requirements for tracking vectors. Herein, we have demonstrated a new paradigm for two-photon tracking gene delivery based on a dendritic nano-sized hexanuclear ruthenium(II) polypyridyl complex. Because this metallodendrimer has a multivalent periphery, the complex, which is 6.1 nm, showed high stability and excellent dispersibility and could stepwise condense DNA in vitro. With the outstanding photochemical properties of Ru(II) polypyridyl, this complex could track gene delivery in vivo using one- and two-photon imaging.

Herein, a dendritic hexanuclear Ru(II) polypyridyl complex based on polybenzimidazole ligands, Ru 6 L ( Fig. 1), was synthesized and developed as a one-and two-photon luminescent tracking non-viral gene vector. The photophysical properties and the size of Ru 6 L were characterized. After studying the interactions of Ru 6 L with DNA and the DNA condensation mechanism in vitro, the cellular uptake of both the complex and the Ru 6 L-DNA particles was investigated. Then, the complex was used as a one-and two-photon luminescent tracking non-viral gene vector. Finally, the transfection efficiency and cytotoxicity of the DNA condensates were evaluated by luciferase assays and MTT assays.
The photophysical data of Ru 6 L are summarized in Table S1. The absorption band at approximately 400-550 nm was assigned to metal-ligand charge transfer (MLCT) absorption ( Figure S4). The maximum emission wavelength (λ em ) was 612 nm, with a lifetime of 110 ns. The emission quantum yield was 0.027, and the largest two-photon absorption (TPA) cross-section was 175 GM at 830 nm with reference to rhodamine B ( Figure S5). The complex was a perfectly monodisperse nanoparticle with 12 positive charges derived from Ru 2+ ions ( Figure S6). The diameter of Ru 6 L was determined to be 6.1 nm by TEM (Fig. 2), which was similar to the hydrodynamic diameter 6.7 ± 0.3 nm determined by DLS ( Figure S6).
We investigated the complex's ability to condense DNA. The first study examined the migration of Ru 6 L-pBR 322 DNA by gel retardation assays with a load of 45 μ M (bp) DNA in each well. As shown in Fig. 3, with increasing the concentrations of Ru 6 L (0-6 μ M), pBR 322 DNA gradually diminished and was retained in the well at a + /− ratio (total positive charge/total negative charge) of 1.1. The primary DNA condensation-driving forces were considered electrostatic interactions by the multivalent periphery of Ru 6 L. The DNA binding ability was determined by a DNA titration approach with CT-DNA (calf thymus DNA). The binding constant was calculated to be 1.75 × 10 5 M −1 ( Figure S7).
The average hydrodynamic diameter of Ru 6 L-pBR 322 DNA particles at various + /− ratios in aqueous solution is shown in Fig. 3. The particle diameter was stable at approximately 140 nm when the + /− ratio was 4 or higher, while the zeta potential increased as the + /− ratio increased. To visually investigate the condensation process, we performed AFM of the Ru 6 L-2kb DNA particles at various + /− ratios (Fig. 4). When the + /− ratio was 1, local DNA bending, long-range cross-links and micro-loops were observed. Upon increasing the + /− ratio to 2, small particles were observed at the middle or the tail of the DNA duplex. As the + /− ratio increased, highly condensed DNA molecules were observed, and they condensed into compact hemispherical structures at a + /− ratio of 20. Ru 6 L-pBR 322 DNA particles at a + /− ratio of 20 were also observed as well-distributed DNA particles with a diameter from 50 to 90 nm ( Figure S8). In addition, we performed a DNA nuclease-catalyzed biodegradation assay ( Figure S9a), a photocleavage assay ( Figure S9b) and a continuous irradiation assay ( Figure S10) with Ru 6 L. The results showed that Ru 6 L displayed no photocleavage or photoreactivity toward DNA; in contrast, it protected DNA against nuclease-catalyzed biodegradation.
Before performing cellular uptake studies of Ru 6 L-DNA particles, we investigated the mechanism of cellular uptake of Ru 6 L using confocal microscopy ( Figure S11). Generally speaking, the cellular uptake of small molecules can occur through energy-independent (facilitated diffusion, passive diffusion) and energy-dependent (endocytosis, active transport) pathways. To determine whether Ru 6 L entered the cell via an energy-independent or energy-dependent transport pathway, HeLa cells were either incubated with Ru 6 L at 4 °C or pretreated with the metabolic inhibitors 2-deoxy-D-glucose and oligomycin. Figure  S11b and S11c show that cellular luminescence was significantly suppressed in cases when the cells were incubated with the complex at 4 °C or pretreated with the metabolic inhibitors, indicating that Ru 6 L uptake followed an energy-dependent pathway. Endocytosis is well known as the most common energy-dependent pathway by which eukaryotic cells uptake extracellular materials. Endocytosis is also affected by temperature or adenosine triphosphate (ATP). We used the endocytic inhibitors chloroquine and NH 4 Cl to examine the role of this pathway in Ru 6 L uptake. As observed from the relative intensities and location of Ru 6 L in the cells after treatment with these inhibitors ( Figure S11d and S11e), they had no effect on the cellular uptake of Ru 6 L, which indicates that Ru 6 L is not taken up by living cells via an endocytosis pathway. This finding is not surprising because cellular uptake mechanisms are generally complex and diverse.
Flow cytometry was used to quantify the intracellular uptake of the DNA particles at the + /− ratio of 20 ( Figure S12). The result indicated that most of cells took up the DNA particles after a 4 h incubation in serum-free DMEM. We examined the localization of DNA particles in HeLa cells by TEM. TEM micrographs confirmed the uptake of the particles by endocytosis and showed that the DNA particles made HeLa cell form an endosome after they entered the HeLa cells (Fig. 5). Particles were also found in the cytoplasm, meaning that the particles successfully escaped from endosomes, which was important for transfection. The particle size was approximately 50 nm, as measured by AFM. The mechanism of cellular uptake of the DNA particles was also studied ( Figure S13). The cellular luminescence of Figure S13b which incubated with the particles at 4 °C was significantly suppressed, it means the particles entered the cell via an energy-independent transport pathway. After treatment with the metabolic inhibitors ( Figure  S13c) and the endocytic inhibitors ( Figure S13d and S13e), no effect was found in the cells which treated with the metabolic inhibitors, but the cellular luminescence which treated with the endocytic inhibitors was suppressed, the results indicated that Ru 6 L-pEGFP DNA particles were taken up by living cells via an endocytosis pathway.
To investigate the intracellular behaviors of Ru 6 L-pEGFP DNA particles, one-and two-photon fluorescence microscopy was used to monitor the time-dependent transport and transfection of pEGFP DNA plasmids that were condensed by Ru 6 L (Fig. 6). We stained the nuclei of HeLa cells with Hoechst 33258; then, the DNA particles were added to the cells. In 20 min, the Ru 6 L-DNA particles attached to the cell membrane quickly due to their high positive zeta potential. After 4 h, particles were found in the cytoplasm. Then, the culture medium was replaced with fresh DMEM containing 10% fetal bovine serum (FBS). After another 12 h, EGFP expression was detected, and higher EGFP expression was found after 24 h. Similar behaviors were observed under one-photon and two-photon excitation. In addition, due to the two-photon absorption, the background signal was strongly suppressed, and a higher resolution image was obtained using two-photon confocal laser scanning microscopy.
The most commonly used method for intracellular plasmid trafficking is the fluorescent labeling of non-viral vectors with organic dyes 14,15 . It should be pointed out that to track the dynamic changes during a specific period of time, the dye must possess improved photostability and must be photostable under continual irradiation with light from fluorescent microscopes. However, most organic dyes have notable shortcomings, including poor water solubility, high toxicity to living cells and poor photostability. Organic dyes may also cause extensive cellular damage and unwanted background signal due the ultraviolet (UV) radiation required for their excitation and small Stokes shifts 26,27 . The short excitation wavelengths (< 650 nm) also inhibit the use of these materials in thick tissues or live animals due to the resultant short penetration depth 47 . In addition, the introduction of dyes may alter the delivery mechanisms and have increased side-effects 29 . The use of Ru 6 L with intrinsic two-photon luminescence as DNA carriers is an attractive solution to these problems because this complex exhibits near infrared (NIR) excitation wavelengths 48 .
We determined the relative transfection efficiency of this nonviral system by luciferase assays, and plasmid pGL3 was used a control vector (Fig. 7). As a control, applying DNA only resulted in a low level of luciferase expression. The luciferase expression was increased when Ru 6 L was used. The luciferase expression was stable when the + /− ratios of the particles ranged from 12 to 24, and the highest luciferase expression was in response to the + /− ratio of 20. Cytotoxicity of non-viral gene vectors is one of the major concerns of gene delivery. Therefore, we examined the viability of HeLa cells treated with the complex and Ru 6 L-pEGFP DNA particles (Fig. 8).
With increasing concentrations, the viability of the HeLa cells decreased slowly for both the complex and the particles. Although Ru 6 L exhibited the related cytotoxicity as the concentration increased, the viability of HeLa cells was more than 80% at a concentration of 5 μ M. The results indicated that this transfection system had relatively low cytotoxicity.

Conclusion
In the present study, we have designed and synthesized a dendritic nano-sized hexanuclear Ru(II) polypyridyl complex (Ru 6 L) based on polybenzimidazole ligands. This complex has a multivalent periphery and a nanosize of 6.1 nm. Gel retardation assay, TEM, AFM and dynamic light scattering  studies show that Ru 6 L exhibits high stability and excellent dispersibility and could stepwise condense DNA in vitro. More interestingly, Ru 6 L was applied as a non-viral gene vector for tracking DNA delivery in live cells using one-and two-photon fluorescence microscopes. With two-photon microscopy, a high signal-to-noise contrast was achieved by irradiation with an 830 nm laser. Our work provides new insights into improving real-time tracking during gene delivery and transfection as well as important information for the design of multifunctional non-viral vectors.

Methods
All reagents were purchased from commercial sources and used without further purification unless otherwise specified. The plasmid pBR 322 DNA was obtained from MBI Fermentas, the plasmid pEGFP DNA was purchased from Clontech, and the plasmid pGL3 control vector and luciferase kid were obtained from Promega. Unless otherwise stated, the DNA concentrations are expressed in base pairs. All samples were prepared using distilled water that had been passed through a Millipore-Q ultra-purification system. The compounds 1,10-phenanthroline-5,6-dione 49 and [Ru(bpy) 2 ]Cl 2 .2H 2 O 50 were prepared according to literature methods. The complex was dissolved in DMSO prior to the experiments. Then, the calculated quantities of the complex solutions were added to the appropriate medium to yield a final DMSO concentration of less than 1% (v/v).
Microanalyses (C, H and N) were performed with a vario EL cube elemental analyzer. Infrared spectra were obtained with a Nicolet 170SX-FTIR spectrophotometer and KBr discs. Electrospray ionization mass spectra (ESI-MS) were recorded on an LCQ system (Finnigan MAT, USA).Fast atom bombardment mass spectrometry (FAB-MS) was recorded using a VG ZAB-HS. 1 H NMR spectra were recorded on a Varian-500 spectrometer at 25 °C. All chemical shifts are given relative to tetramethylsilane (TMS). The UV-Vis spectra were recorded on a Perkin-Elmer Lambda 850 spectrophotometer. Emission spectra were recorded on a Perkin-Elmer LS 55 spectrophotometer at room temperature (25 °C). Time-resolved emission measurements were conducted on an FLS 920 combined fluorescence-lifetime and steady-state spectrometer. Quantum yields of luminescence at room temperature (25 °C) were calculated according to literature procedures using [Ru(bpy) 3 ] 2+ (ϕ = 0.028 in aerated aqueous solution) as the reference emitter 51 . All date were processed using the Origin 8 software package.
Atomic force microscopy (AFM) images were obtained in air at room temperature with an SPA400 atomic force microscope unit and an SPI3800N control station (Seiko Instruments) operated in the tapping mode. Probes were made of a single silicon crystal with a cantilever length of 129 mm and a spring constant of 33-62N/m (OMCLAC160TS-W2, Olympus). Dynamic light scattering and zeta potential experiments were performed using dynamic laser light scattering equipment (DLS, Brooken Haven BI-200SM). Transmission electron micrographs were obtained with a JEM100CX electron microscope.

Synthesis of complex {[Ru
Dynamic laser light scattering equipment was used to determine the average hydrodynamic diameter and the zeta potential of Ru 6 L at a concentration of 5 μ M in distilled water that had been passed through a Millipore-Q ultra-purification system. Typically, 5 runs were measured for the solution, and the average of all runs is reported.
Determination of two-photon absorption cross-sections. The two-photon absorption (TPA) spectra of complex were determined over a broad spectral region by a two-photon induced luminescence (TPL) method relative to Rhodamine B in methanol as the standard. The two-photon luminescence data were acquired using an Opolette TM 355II (pulse width ≤ 100 fs, 80 MHz repetition rate, tuning range 730-890 nm, Spectra Physics, Inc., USA). Two-photon luminescence measurements were performed in fluorometric quartz cuvettes. The experimental luminescence excitation and detection conditions were conducted with negligible reabsorption processes, which can affect TPA measurements. The quadratic dependence of the two-photon induced luminescence intensity on the excitation power was verified at an excitation wavelength of 830 nm. The two-photon absorption cross-section of the complex was calculated at each wavelength according to Equation (1)  where I is the integrated luminescence intensity, C is the concentration, n is the refractive index, and ϕ is the quantum yield. The subscript '1' refers to the reference samples, and '2' for the experimental samples.
Preparation of DNA particles. The DNA particles were prepared by incubating the mixtures containing DNA and Ru 6 L at the given + /− ratios in 50 mM Tris-HCl (Tris = Tris(hydroxymethyl) aminomethane) solution (pH = 7.4) or in cell culture, followed by vortexing for 30 min to allow equilibration at room temperature. Dynamic light scattering and zeta potential assay. Dynamic laser light scattering equipment was used to determine the average hydrodynamic diameter and the zeta potential of Ru 6 L-pBR 322 DNA particles at various + /− ratios in 50 mM Tris-HCl solution (pH = 7.4). Typically, 5 runs were measured for each solution, with the average of all the runs reported.

Gel retardation assay.
AFM imaging. The 2kb DNA fragment was generated by PCR using pBR 322 plasmid DNA as a template 53 . The condensation process of Ru 6 L-2kb DNA and the morphologies Ru 6 L-pBR 322 DNA were examined with AFM. Samples were dropwise-added (10 μ L) onto a mica substrate, which was freshly cleaved by pulling off the top sheets with tape. One min later, the substrate was spin coated (1400 rpm, 30 s) and rinsed with 20 μ L of distilled water. AFM images were obtained in air at room temperature. The images were captured in a 256 × 256 pixels format and analyzed with the software accompanying the imaging module. Cellular uptake of Ru6L-DNA particles. The cells were trypsinized, counted, and adjusted to 1 × 10 4 cells mL −1 , and 1 mL of the cell solution was added to each plate. After 24 h, the cell culture medium was replaced with 800 μ L serum-free DMEM. Ru 6 L-pEGFP DNA particles at the + /− ratio of 20, containing 1 μ g pEGFP DNA in 200 μ L serum-free DMEM, was added to the cells, and the cells were incubated at 37 °C for 4 h.
For flow cytometry, the cells were washed with PBS three times, trypsinized and centrifuged in PBS. Cells were harvested, and single cell suspensions in 0.5 mL PBS were prepared and subjected to flow cytometric analysis. A flow cytometer (Coulter Co. USA) was used to measure the fluorescent intensity, with excitation at 488 nm.
For TEM imaging analysis, cell processing was performed in situ, without displacement from the culture dish. Cells were fixed in 0.1 M PBS containing 2.5% glutaraldehyde and 4% paraformaldehyde for 1 h, rinsed with distilled water, stained with 0.5% uranyl acetate for 1 h, dehydrated in a graded series of ethanol (30, 60, 70, 90 and 100%), and embedded in epoxy resin. The resin was polymerized at 60 °C for 48 h. Ultrathin sections (50-75 nm) obtained with an LKB ultramicrotome were stained with 2% aqueous uranyl acetate and 2% aqueous lead citrate.
The mechanism of cellular uptake of Ru6L-DNA particles. The cells were trypsinized, counted, and adjusted to 1 × 10 5 cells mL −1 and 1 mL was added to five 35-mm 2 Petri dishes (MatTek, USA) for laser confocal microscopy. After 24 h, the first dish was incubated with Ru 6 L-pEGFP DNA particles at the + /− ratio of 20 at 37 °C for 4 h. The second dish was incubated with Ru 6 L-pEGFP DNA particles at the + /− ratio of 20 at 4 °C for 4 h. The third dish was pretreated with 50 mM 2-deoxy-D-glucose and 5 μ M oligomycin in PBS for 1 h at 37 °C and then incubated with Ru 6 L-pEGFP DNA particles at the + /− ratio of 20 at 37 °C for 4 h. The fourth and the fifth dishes were pretreated with endocytic inhibitors NH 4 Cl (50 mM), and chloroquine (50 μ M) for 30 min respectively, and then incubated with Ru 6 L-pEGFP DNA particles at the + /− ratio of 20 at 37 °C for 4 h.
After being washed with fresh PBS (pH = 7.0) three times, the cells were imaged on a Zeiss LSM 710 NLO confocal microscope (63 × /NA 1.4 oil immersion objective). For two-photon images, the excitation wavelength of the laser was 830 nm for Ru 6 L and the emission spectra were integrated over 580-630 nm (single channel).
Scientific RepoRts | 5:10707 | DOi: 10.1038/srep10707 One-and two-photon luminescent imaging. The cells were trypsinized, counted, and adjusted to 1 × 10 5 cells mL −1 , and 1 mL of the cell suspension was added to a 35-mm 2 Petri dish (MatTek, USA) for laser confocal microscopy. After 24 h, the cell culture medium was replaced with 800 μ L serum-free DMEM. Ru 6 L-pEGFP DNA particles at the + /− ratio of 20, containing 1 μ g pEGFP DNA in 200 μ L serum-fee DMEM, was added to the cells, and the cells were incubated at 37 °C for 4 h. Then, the medium was replaced with fresh DMEM containing 10% FBS, and the cells were incubated for various timed (20 min and 4, 12, and 24 h).
After being washed with fresh PBS (pH = 7.0) three times, the cells were imaged on a Zeiss LSM 710 NLO confocal microscope (63 × /NA 1.4 oil immersion objective). The excitation wavelength of the laser was 488 nm, and the emission spectra were integrated over 580-630 nm (single channel). For two-photon images, the excitation wavelength of the laser was 830 nm for Ru 6 L.
Luciferase assay. HeLa cells were seeded onto a 96-well cell-culture plate at a cell density of 1 × 10 4 cells per well and then incubated for 24 h. Cells were washed with PBS three times and then cultured with serum-free DMEM. Ru 6 L-pGL3 DNA control vector particles with increasing concentrations of the tested complex containing 0.2 μ g plasmid were added to the cells, and the cells were incubated at 37 °C for 4 h. The medium was then replaced with fresh DMEM with 10% FBS, and the cells were incubated for an additional 24 h. Cells were washed with PBS, harvested and treated for 30 min at 4 °C with end-over-end rotation in lysis buffer (50 mM Tris-HCl, pH = 7.5, 150 mM NaCl, 2% Triton X-100, 2% NP40). The luciferase assay was performed according to the manufacturer's protocol (Promega). Relative light units (RLUs) were measured with Varioskan Flash (Thermo Scientific, USA) and a GloMax TM 96 microplate luminometer (Promega, USA).