Radiolabeling and Preliminary Evaluation of 99mTc-Labeled DNA Cube Nanoparticles as Potential Tracers for SPECT Imaging.

PURPOSE: DNA nanostructures, with the advantages of structural designability and spatial addressability, have shown a great potential in the field of drug delivery and bio-medicine. Herein, we aimed to prepare technetium-99m radiolabeled DNA cube nanoparticles (99mTc-DCN) and expect to build a DCN-based drug carrier and nuclear medicine imaging platform.
METHODS: DCN could be readily assembled with 6 designed DNA oligonucleotides at an equal mole ratio in a single annealing procedure. 99mTc-MAG3-ssDNA (A20) was obtained by labeling MAG3-ssDNA (A20) with technetium-99m by using a stannous chloride reduction method. 99mTc-DCN was prepared by hybridize DCN with side chains (T20-DCN) with 99mTc-MAG3-ssDNA (A20). The biodistribution study and SPECT/CT imaging were conducted on KM mice.
RESULTS: DCN was successfully assembled, and as-prepared DCN was characterized by native polyacrylamide gel electrophoresis, atomic force microscope and fluorescence resonance energy transfer. The size of DCN was about 5 nm. The radiolabeling yield of 99mTc-MAG3-ssDNA (A20) was approximately 90% by radio thin-layer chromatography. T20-DCN mixed with 99mTc-MAG3-ssDNA (A20) in PBS could generate 99mTc-DCN upon hybridization. The retention time (RT) of 99mTc-MAG3-ssDNA (A20) was at ~22 min, and the RT of as-prepared 99mTc-DCN was at ~12 min by radio-HPLC. The results from biodistribution study and SPECT/CT imaging showed that a significant proportion of DCNs were metabolized through the liver and kidney. Intestine exhibited a relatively indicative signal as well, which might be explained by the enterohepatic circulation of DCN via the liver and gallbladder.
CONCLUSION: We have successfully prepared 99mTc-DCN as a SPECT/CT imaging probe via the side-chain hybridization strategy. The probe was metabolized mainly by the liver and excreted primarily to the bladder. Due to the superior properties of DNA cube nanoparticles, we believe DCN may potentially be translated into a preclinical setting for diagnosis and treatment of cancer-related diseases.

Introduction

Nuclear medicine imaging is to introduce a small amount of imaging drugs called “probes” into the body and use emission computed tomography equipment to explore the dynamic and/or static distribution of “probes” in the human body or target organs. These probes have the same or similar physiological and biochemical characteristics as the natural metabolites in the human body, so as to understand the changes in the functions, physiology and biochemistry, metabolism and gene expression of human organs. Single photon emission computed tomography (SPECT), one of the molecular imaging methods in nuclear medicine, involves the use of radioisotope-labelling probes. It has the advantages over other imaging modalities (such as optical imaging, computed tomography, magnetic resonance imaging, etc.) in high sensitivity, deep penetration depth, and the possibility to perform functional imaging.1,2 Extensive efforts have been made in SPECT imaging to collect more detailed in-vivo information by using different radioisotope-labelling probes.3–5 Nanotechnologies in the development of smart molecular imaging platforms are supplying potential imaging probes for multimodal imaging, which combine the advantages of various imaging modalities in order to improve specificity and sensitivity to the target.6–8

DNA nanotechnology is a technology for artificially designing DNA strand sequences, following the base-pairing rules to combine DNA strands into a double helix structure, finally forming a defined and robust three-dimensional structure. This technology can be used to predict and design complex DNA nanostructures. In this field, DNAs are used as non-biological nanomaterials, rather than as genetic information carriers for living cells. DNA nanostructures, which can be functionalized with different targeting ligands (such as organic molecules, peptides, antibody, aptamer and so on),9 display excellent delivery capacity for various drugs (such as CpG, doxorubicin, or siRNA), as well as signal moieties (fluorescent dyes or radionuclides). These signal molecules can be utilized to explore the in-vivo information of DNA nanostructures including biodistribution, metabolism and stability properties. DNA nanostructures, with the advantages of structural designability and spatial addressability, have shown a great potential in the field of bio-sensing,10−12 drug delivery,13–15 and bio-medicine.16–19

DNA cube nanoparticles (DCN),20 a classic example of DNA nanostructures, could be readily assembled with 6 designed DNA oligonucleotides. To successfully prepare DCN, we designed a set of six short oligonucleotides for self-assembly, and the results supported our establishment of DCN. MAG3 was chosen for Tc-99m chelation since we preferred a smaller and more hydrophilic ligand for radiolabeling. Herein, we aimed to prepare technetium-99m radiolabeled DCN (99mTc-DCN), apply the as-formed 99mTc-DCN for SPECT imaging to investigate its in-vivo biodistribution properties and pharmacokinetics profiles, and expect to build a DCN-based drug carrier and nuclear medicine imaging platform.

Materials and Methods

General

Sodium Pertechnetate [99mTc] was supplied by HTA Co., Ltd. (Beijing, China), with radionuclide purity (99Mo% <5*10−2%) and radiochemical purity (>98%). Tris(hydroxymethyl)aminomethane (Tris base, ≥99.8%) was purchased from Merck Limited (Beijing, China). All other analytical reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. (SCRC, Shanghai, China) and used as received unless stated otherwise.

The interested tissues and organs from the biodistribution study were measured with a γ-counter (SN-697, Shanghai Rihuan Photoelectronic Instrument Co., Ltd.). NanoScan SPECT/CT (Mediso Ltd., Hungary) was used to conduct SPECT/CT imaging.

Preparation of DNA Cube Nanoparticles

DCNs were prepared via a single annealing procedure20,21 similar to the method which was used to prepare DNA bipyramid nanostructures (DBNs).3 In brief, equal molar quantities (1 μM) of six single-strand DNA (a, b, c, d, e, and f) with designated sequences, which were purchased from TaKaRa Biotechnology Co., Ltd. (Dalian, China), were mixed in TM buffer (10 mM Tris base and 50 mM MgCl2, pH 8.0), heated to 95°C for 10 min, and then cooled to 4°C for 20 min. The oligonucleotide sequences of each single-strand DNA (ssDNA) used in the study are listed in Table 1.

Table 1

The Oligonucleotide Sequences of Each ssDNA Used in the Study

ssDNA	Sequence
a	GGCAACTTTGATCCCTCAGGTTTAGCGCCGGTCCTTTTCTCCCACTACTTTCACTG
b	GGGAAACTTTCGTGTGTAGGTTTTGTTGCCCAGTGTTTCTACGATGTACTTTGGTC
c	GGACATTTTCGAGATCAGCATTTGTTTCCCGACCTTTGCGTGATTGTATTTTAGAG
d	GGCGCTTTTGACACTTCTGCTTTATGTCCCTCTATTTCTTAGATGACTTTTGGACC
e	GGGAGATTTAGTCATCTAAGTTTTACAATCACGCTTTGTACATCGTAGTTTGTAGT
f	GGGATCTTTACCTACACACGTTTTGCTGATCTCGTTTGCAGAAGTGTCTTTCCTGA
a-T20	TTTTTTTTTTTTTTTTTTTTGGCAACTTTGATCCCTCAGGTTTAGCGCCGGTCCTTTTCTCCCACTACTTTCACTG
A20	AAAAAAAAAAAAAAAAAAAA
Cy3-b	Cy3-GGGAAACTTTCGTGTGTAGGTTTTGTTGCCCAGTGTTTCTACGATGTACTTTGGTC
Cy5-c	GGACATTTTC(Cy5)GAGATCAGCATTTGTTTCCCGACCTTTGCGTGATTGTATTTTAGAG

PAGE Characterization of DCN

The single-step annealing process of DCN was analyzed using native polyacrylamide gel electrophoresis (PAGE). The as-formed DNA samples were run in 8% PAGE gel in TAE-Mg buffer at 4°C for about 2 h. The gel was stained with GelRed DNA gel stain solution (Biotium, Inc., Hayward, CA, USA) for 15 min and analyzed by the chemiluminescence imaging system (G:Box Chemi-XL, Syngene, A Division of Synoptics Ltd., Cambridge, United Kingdom). The yields of DCNs were semi-quantitatively determined by using ImageJ analysis software (NIH, Bethesda, MD, USA).

FRET for Verificating Correct Assembly of DCN

Fluorescence resonance energy transfer (FRET) technique was used to detect the correct assembly of DCN. Cy3 and Cy5 were chosen as FRET donors and acceptors, respectively. FRET-DCN was assembled with another set of six ssDNA (a, Cy3-b, Cy5-c, d, e and f). Oligonucleotide sequences of Cy3-b and Cy5-c are listed in Table 1. Cy3-b and Cy5-c were obtained by connecting Cy3 with b strand at 5ʹ end and connecting Cy5 with c strand at the tenth base from 5ʹ end. FRET-DCN was excited using the 535 nm laser line, and the fluorescent signal was measured using a fluorescence spectrophotometer (Hitachi F-4500 Fluorescence Spectrophotometer).

HPLC for Purifying DCN

DCNs were purified by high performance liquid chromatography (HPLC, Agilent 1100 chromatography system, Agilent Technologies) with a size exclusive chromatography column (Phenomenex® BioSec-SEC-4000, 300*7.8 mm). The mobile phase is 150 mM NaCl with a flow rate of 0.3 mL/min. The solution was concentrated with 30 kDa Amicon Ultra-0.5 mL centrifugal filters (EMD Millipore Corporation) at 3000 g for 20 min. The concentration of purified DCN was measured by their absorbance at 260 nm with a UV spectrophotometer (HITACHI U-3010).

Atomic Force Microscope Imaging of DCN

DCN (2 μL, 20 nM) was deposited onto a freshly cleaved mica cell and left to adsorb for 2–3 min. 1×TAE-Mg2+ buffer (400 μL) was added to the liquid cell, and the sample was scanned in a tapping mode under fluid on a Pico-Plus AFM (Agilent Technologies, Santa Clara, CA, USA) with NP-S tips (Veeco, Inc., Oyster Bay, NY, USA).

Preparation of MAG3-ssDNA (A20)

The experimental procedure of MAG3-ssDNA (A20) preparation followed the study published previously in our lab.3 In brief, 300 μg NHS-MAG3 ester and 250 μg ssDNA (A20) were mixed gently in 0.3 M HEPES (100 μmol/L, pH 8.0) at room temperature and incubated for 2 h. The reaction was terminated by adding 50 uL ammonium acetate (2 mol/L) to the mixture. MAG3-ssDNA (A20) was purified through a NAP-5 column (GE) with 0.25 mol/L ammonium acetate as eluent.

Radiolabeling of DCN

First, 99mTc-MAG3-ssDNA (A20) was obtained by labeling MAG3-ssDNA (A20) with technetium-99m by using a stannous chloride reduction method. MAG3-ssDNA (A20) was mixed with 10 μL tartrate buffer (50 mg/mL), 5 μL SnCl2 solution (4 mg/mL, freshly prepared), and finally 25 μL sodium pertechnetate [99mTc] (~10 mCi) for 20 min at 95°C. Then, when the temperature of the reaction solution droped to room temperature, DCN with side chains (T20-DCN) was added into the reaction solution to hybridize with 99mTc-MAG3-ssDNA (A20), which was showed in Scheme 1. Last, 99mTc-DCN were analyzed and purified through Radio-HPLC system.

Scheme 1

Schematic display of the assembly and radiolabeling process of 99mTc-DCN.

In-vivo Biodistribution Study

The biodistribution study of DCN was conducted by injecting 200 uL 99mTc-DCN (~10 μCi) into KM mice via tail vein. The mice (n=5 at each time point) were sacrificed by cervical dislocation and dissected, and interested tissues or organs were collected to measure radioactivity by a γ-counter at 60 sec counting periods. Uptake values were corrected for radio-decay and expressed as the percentage of injected dose per gram of tissue (%ID/g).

Small Animal SPECT/CT Imaging

In SPECT/CT imaging, 99mTc-DCN (~500 μCi for each mouse) was injected into KM mice (n=5) via the tail vein without anesthesia. After 1 h post injection, the mice were anesthetized by inhaling the isoflurane (1.5%)-oxygen mixture, and placed prone on the scanning bed for SPECT/CT imaging by nanoScan SPECT/CT with large FOV (12 cm X 10 cm). CT scanning was performed following each SPECT acquisition. SPECT and CT images were reconstructed using the MEDISO software and then merged and analyzed using the InVivoScope software (Bioscan).

Results

Preparation and Characterization of DCN

DCN was assembled by six oligonucleotides with designated sequences (a, b, c, d, e, and f) mixed at an equal mole ratio of about 1 μM. The as-prepared DCN was characterized by PAGE. With each single-strand oligonucleotide added to the reaction solution, the migration of as-prepared DNA samples would gradually slow down until the final formation of DCN, which is shown in Figure 1A. The size of DCN measured with AFM was about 5 nm (Figure 1B). DCN was assembled correctly, which was proved by FRET. The peak at 665 nm, which is Cy5 emission peak, appeared obviously, while there was almost no peak at 565 nm, which is Cy3 emission peak (Figure 1C). The crude product was purified by HPLC, and the retention time (RT) of DCN was about 14 min (Figure 1D).

Figure 1

Preparation and Characterization of DCN. (A) Characterization of the DCN self-assembly process by PAGE. M1, 50 bp marker. M2, 20 bp marker; (B) AFM image of DCN. Scale bar: 50 nm; (C) FRET analysis of DCN. A FRET pair (Cy3 and Cy5) was designed on the same vertex of DCN, and FRET effect could be observed during the assembly of DCN; (D) HPLC analysis of DCN, and the retention time was about 14 min.

Radiolabeling of MAG3-ssDNA

First, S-acetyl-MAG3-NHS was conjugated with NH2-ssDNA A(20), and the product MAG3-ssDNA A(20) was radiolabeled by technetium-99m with a single reduction reaction to obtain 99mTc-MAG3-ssDNA (A20). The yield of the radiolabeling reaction was approximately 90%, and the specific activity (SA) of 99mTc-MAG3-ssDNA (A20) was 1.0–1.2 mCi/nmol. Radio thin-layer chromatography (Radio-TLC) was used to analyze the radiolabeling yield. Three components would occur during the radiolabeling process: free technetium-99m, technetium hydrogel, and radiolabeled ssDNA. Two developing system was applied during radio-TLC analysis. When developed in 100% acetone, radiolabeled ssDNA and technetium hydrogel would both stay at the original point, while free technetium would move to the solvent front. Moreover, when developed by saline (0.9% sodium chloride), only technetium hydrogel would stay at the original spot, while radiolabeled ssDNA and free technetium would migrate to the solvent front. Then, by a simple subtraction, we could calculate the radiochemistry purity of radiolabeled ssDNA. As shown in Figure 2, almost no free technetium was observed, and technetium hydrogel was about 8% in the reaction solvent. Thus, the radiochemistry purity of 99mTc-MAG3-ssDNA (A20) prepared was about 92%.

Figure 2

Radio-TLC image of 99mTc-MAG3-ssDNA (A20) under 2 developing system. Black line: 100% Acetone; Red line: 0.9% saline.

In Scheme 1, a sequence of T(20) was designed to the 5ʹ end of oligonucleotide a, and the as-formed DCN thus had a T(20) side chain (T20-DCN), which made it available for hybridization with an A(20) single-stranded DNA (ssDNA). Radiolabeling of DCN could be readily achieved by hybridization between 99mTc-MAG3-ssDNA (A20) and T20-DCN. T20-DCN mixed with 99mTc-MAG3-ssDNA (A20) in phosphate saline buffer (PBS, pH 7.4) under mild shaking conditions at room temperature for 5–10 min could generate 99mTc-DCN upon hybridization, and the estimated SA of radiolabeled DCN was in the range of 6.0–7.2 mCi/nmol. Radio-HPLC was used for analysis and purification of radiolabeled DCN for in vivo investigation. As shown in Figure 3, the RT of 99mTc-MAG3-ssDNA (A20) was at ~22 min, and the RT of as-prepared 99mTc-DCN was at ~12 min, indicative of the successful preparation of radiolabeled DCN (99mTc-DCN).

Figure 3

Radio-HPLC analysis of DNA samples. The RT of 99mTc-DCN was at ~12 min, and 99mTc-MAG3-ssDNA (A20) was at ~22 min.

We investigated the biodistribution pattern of 99mTc-DCN in healthy mice, denoted as the percent injected dose per gram organ (%ID/g). As shown in Figures 4, 99mTc-DCN in the blood pool peaked at 5 min after injection, and then rapidly decreased later on, suggesting a rapid distribution of DCN in vivo. Within 4 hours after injection, the tracer uptake in the heart, lung, intestine, stomach and brain was never higher than 5%ID/g, while accumulation in the liver dropped from about 16.33 ± 1.86%ID/g at 15 min to around 0.18 ± 2.29%ID/g at 4 h after injection. In the kidneys, 99mTc-DCN had reached about 27.84 ± 1.66%ID/g at 5 min after injection, and then decreased rapidly to about 7.62 ± 0.26%ID/g and 0.59 ± 0.08%ID/g at 60 min and 4 h after injection, respectively. As for the tracer uptake in the spleen, 99mTc-DCN dropped down from 4.25 ± 0.76%ID/g at 5 min to 0.59 ± 0.27%ID/g at 4 h after injection.

Figure 4

Biodistribution profile (%injected dose per gram organ, %ID/g organ) of 99mTc-DCN in healthy mice.

Finally, SPECT/CT imaging of 99mTc-DCN in healthy mice was performed, and metabolic pathway of DCN in vivo was analysed. As shown in Figure 5, after intravenous injection of 99mTc-DCN for 1 h, a significant amount of 99mTc-DCN was seen in the bladder. Moreover, the gallbladder and intestine showed a high uptake of 99mTc-DCN.

Figure 5

SPECT/CT images of 99mTc-DCN in a healthy mouse at 1 h p.i. Middle: sagittal position; Right, coronal position.

Discussion

The innovation of nanotechnology has made considerable progress in the field of biomedical research, including drug delivery, targeted therapy, molecular imaging, and biosensing. Among them, DNA nanotechnology uses natural or artificial DNA strands as constituent elements to prepare 2D and 3D nanostructures through specific interactions between DNA molecules, namely Watson–Crick base pairing. DNA nanostructures have the advantages of easy construction, controllable structure and size/shape, as well as multiple chemical modification pathways, and exhibit unprecedented versatility in terms of complexity and programmability, which provides a versatile platform for biomedical applications. The oligonucleotide strands constituting DNA nanostructures can be covalently linked with small molecules, proteins or nucleic acid, endowing DNA nanostructures with a stable structure under physiological conditions, high targeting and affinity, high cellular uptake rate, or improved delivery capacity of molecular cargo, etc., which enhanced the function of DNA nanostructures in biological environments. DCN, which is assembled with 6 designed DNA oligonucleotides, is one of DNA nanostructures, and possesses the characteristics of DNA nanostructure described above. We would like to further understand the feasibility of DCN as molecular imaging probes in this study.

DCN was successfully assembled by six oligonucleotides with designated sequences via a single annealing procedure. The as-prepared DCN was characterized by PAGE and FRET, purified by HPLC, and further used for technetium-99m labeling experiment. Technetium-99m labeled DCN was prepared by side-arm chain hybridization, which is consistent with the preparation method of technetium-99m labeled DNA bipyramid reported in the literature.3 MAG3-ssDNA (A20) was labeled with technetium-99m to obtain 99mTc-MAG3-ssDNA (A20) using a stannous chloride reduction method. 99mTc-MAG3-ssDNA (A20) was hybridized with DCN with side chains (T20-DCN) to prepare 99mTc-DCN.

Intravenous injection of 99mTc-DCN was done to reveal its biodistribution in healthy mice, and the half-life of 99mTc-DCN was calculated to be around 5 min, which was similar to the reported blood half-life of technetium-99m labeled DBNs3 and was about two-fold longer than the blood half-life of double-stranded DNA (dsDNA) reported previously.6 DNA analogs, in the form of single-stranded antisense oligonucleotides or aptamers, have been radiolabeled for cancer imaging. However, these tracers showed extremely fast blood clearance in vivo, with an estimated circulation half-life of 2–3 min. Limited by this rapid excretion, DNA analogs-based imaging probes presented limited tumor accumulation. Also, single-stranded or double-stranded oligonucleotides have poor serum stability. Studies have found that after folding into 3D nanostructures (such as tetrahedron, bipyramid, or cubes), their serum stability improved significantly. The prolonged circulation of 99mTc-DCN in vivo suggested different biological properties of DCN, indicating an intrinsic physiological property of folded 3D nanostructures in vivo. By measuring the accumulation of 99mTc-DCN in organs of interest, we defined the specific biodistribution pattern of DCN in vivo, which showed an efficient renal excretion (from 27.84 ± 1.66%ID/g at 5 min after injection to 0.59 ± 0.08%ID/g at 4 h after injection) and an elongated retention than common dsDNA reported previously,6 yet displayed some similarity with other nanoparticles with diameters less than 5 nm (such as quantum dots22) in mice. As for DBNs, the uptake of tracer in the liver remained at a high level (above 20%ID/g) until 2 hours after injection. In the kidney, spleen and small intestine, the distribution trend of the two imaging agents (99mTc-DCN and 99mTc-DBNs) was similar.3

The results of SPECT/CT imaging were consistent with the results of biodistribution experiment, that the gallbladder showed a high uptake of 99mTc-DCN, suggesting that a significant proportion of DCN was indeed metabolized within the liver. However, in SPECT/CT imaging, the intestine exhibited a relatively indicative signal either, which might be explained by the enterohepatic circulation of DCN via the liver and gallbladder.

Conclusion

In conclusion, we have successfully radiolabeled DNA cube nanoparticles with a gamma-ray emission isotope (Technetium-99m) and prepared 99mTc-DCN as a SPECT/CT imaging probe via a side-chain hybridization strategy. The biodistribution pattern of the probe has been investigated, and SPECT/CT imaging of 99mTc-DCN was performed for preliminary evaluation. The probe was metabolized mainly by the liver and excreted primarily to the bladder. The relatively short half-life of this probe would minimize the radiation dose to ease the concern of radiation hazard, and the excellent biodegradability of DCN would cause minimal toxicity to animals. Due to the superior properties of DNA cube nanoparticles, we believe DCN may potentially be translated into a preclinical setting for diagnosis and treatment of cancer-related diseases.