Synthesis and Preclinical Evaluation of 68Ga-MALAT-1 ASO for PET Imaging of MALAT-1 expressing Tumours

Background: MALAT-1 (Metastasis-Associated Lung Adenocarcinoma Transcript 1) is a large long nuclear noncoding RNA (lncRNA) that is overexpressed in an array of cancers. In this study, we designed a range of positron probes for MALAT-1 to evaluate its distribution, pharmacokinetics, and to explore whether the probe can be used for the imaging of malignant tumors with high MALAT-1 expression in vivo. Methods: 68Ga labelling of MALAT-1 antisense oligonucleotides (68Ga–MALAT-1 ASO) was synthesized by the conjugation of MALAT-1 NOTA-ASO and 68Ga3+. Purity was assessed by radio-HPLC. Pharmacokinetic studies and cell uptake were assessed. The biodistribution and metabolism of 68Ga– MALAT-1 ASO in normal ICR and MHCC-LM3 xenograft-bearing nude mice were studied. Results: 68Ga–MALAT-1 ASO was obtained at a radiochemical yield of 98% from a 10 min synthesis with 100 ± 50 MBq/nmol activity and > 99% purity once synthesized. The Log P was -2.53±0.19. The tracer displayed excellent stability in vitro. 68Ga–MALAT-1 ASO showed satisfactory binding ability to MHCC-LM3 cells; the biodistribution of 68Ga-MALAT-1 ASO in MHCC-LM3 tumour-bearing mice showed high levels of uptake (3.04 ± 0.11%ID/g). Micro-PET scans demonstrated the tumor specific uptake of 68Ga-MALAT-1 ASO in mouse models. Conclusions: We conclude that 68Ga labelling of MALAT-1 ASO is a convenient approach to label tumors overexpressing MALAT-1. vein. The images were performed at 30, 60 and 90 min after injection using a micro PET/CT scanner (Siemens Inveon MultiModality system) while the mice were maintained under isoflurane anaesthesia (1.5% isoflurane, 3% oxygen). A blocking study was performed in which 100 nmol of unlabelled MALAT-1 ASO was intravenously administered 30 min before the intravenous injection of 68 Ga-MALAT-1 ASO (3.7MBq). Three-dimensional volumes of interest (VOIs) were used for the assessment of %ID/g and standard uptake value (SUV) in selected organs.

To our knowledge, lncRNA MALAT-1 ASOs have not been radiolabeled with positron nuclides and has not been investigated for PET applications. Herein, we describe the development of positron probes for lncRNA MALAT-1, to evaluate its distribution, pharmacokinetics, and tumor targeting in vivo.

Reagents
The 20-mer PO oligonucleotides bearing a 5-aminohexyl tether were purchased from Beijing Tsing

Ga-labelling of MALAT-1 antisense oligonucleotides ( Ga-MALAT-1 ASO)
The synthetic route is shown in Figure 1. We produced 68 Ga as [ 68 Ga]Cl 3 in the 68 Ge/ 68 Ga generator via elution with 0.05 N aq. HCl. The 68 Ga 3+ eluate was mixed with the solubilized bioconjugate (5-12 nmol) that was dissolved in 1M HEPES (final pH 4.0-4.2). Reactions proceeded for 10 min at room temperature. 68 Ga-MALAT-1 ASO was separated on purification cartridges and sense oligodeoxynucleotides (SO) were treated as described for ASO. Radio-HPLC was used for the confirmation of yields.

Quality control and stability
The synthesized 68 Ga-MALAT-1 ASO was assayed for purity via HPLC, pH using indicator paper and the absence of suspended precipitates by analytical HPLC.

In Vitro Stability Analysis
In vitro stability studies of 68 Ga-MALAT-1 ASO were performed in foetal bovine serum as well as in phosphate-buffered saline (PBS) solution. Briefly, 50 µL of 68 Ga-MALAT-1 ASO (1 MBq) was mixed with FBS at 37°C for 30 min, 60 min, and 2h. Acetonitrile (0.5 mL) was then added to precipitate serum proteins and radio-HPLC was performed to determine serum stability. Fifty microliters of 68 Ga-MALAT-1 ASO (1 MBq) was mixed with PBS (10 mM, 450 µL) for 2 h and radio-HPLC was used to confirm stability.

Partition coefficient studies
Partition coefficients (Log P) of 68 Ga-MALAT-1 ASO were measured through the assessment of the distribution of radioactivity in 1-octanol and phosphate buffer in a 2 mL centrifuge tube. A 10 µL of 68 Ga-MALAT-1 ASO solution was added to PBS + 1-octanol (total volume: 1 mL) and centrifuged (5 min at 5000 rpm). A total of 2 samples (50 μL) taken from each layer were assayed in a γ counter.
Partition coefficients (log Po/w) are shown as the log-counts in 1-octanol vs PBS layers (n=3).

Animals
ICR female mice (body weight 18-20 g), and BALB/c nude mice (6 to 8 weeks of age) were purchased from Jiangsu Gempharmatech. The mice were raised and managed on a standard diet with free water access at our institute. All animal protocols were approved by our internal review board and followed the standard care procedures for animal use (National Research Council of USA, 1996).

Pharmacokinetic studies
For the pharmacokinetic studies, six female ICR mice weighting 18-20 g was administrated with 68 Ga-MALAT-1 ASO (7.4 MBq, 0.2 mL) by IV injection into the caudal vein. Blood samples (10 μL) were obtained from the tail at various time points from 3 min to 120 min following 68 Ga-MALAT-1 ASO injection. Radioactivity was calculated as the % of dose per g of tissue per body weight (%ID/g) expressed over time. Pharmacokinetics were assayed using DAS, version 2.1.1.
MHCC-LM3 cells seeded into 6-well plates (2.5× 10 6 cells per well) to 80% confluency were treated with 68 Ga-MALAT-1 ASO (final concentration of 100 nM, 0.55 MBq 68 Ga-MALAT-1 ASO) for different times (30, 60, 90,120, and 240 min). Mixtures were then washed in ice-cold PBS and centrifuged to discard the supernatants. The radioactivity in the precipitate was quantitatively measured on a γcounter. Blocking cellular uptake by the addition of excess PO-ASO (1 µM) was studied. These experiments were repeated four times under the same conditions.
Values are shown as the percentage of the injected dose per gram (% ID/g) which corrected for background and decay to maintain consistency.

In vivo micro-PET/CT imaging
Mice with MHCC-LM3 tumours were divided into group 1 (antisense group), 2 (sense group) and 3 (blocked group), (n=4 for each group). Mice were injected with 0.1 mL of 68 Ga-MALAT-1 ASO (3.7 MBq) via tail vein. The images were performed at 30, 60 and 90 min after injection using a micro PET/CT scanner (Siemens Inveon MultiModality system) while the mice were maintained under isoflurane anaesthesia (1.5% isoflurane, 3% oxygen). A blocking study was performed in which 100 nmol of unlabelled MALAT-1 ASO was intravenously administered 30 min before the intravenous injection of 68 Ga-MALAT-1 ASO (3.7MBq). Three-dimensional volumes of interest (VOIs) were used for the assessment of %ID/g and standard uptake value (SUV) in selected organs.

Statistics
Quantitative data were showed as the mean ± SD, and means were compared via one-way ANOVA with SPSS V.22.0 software. P-values ≤ 0.05 were considered significant differences.

Results
The synthesis procedure for the NOTA-ASO conjugate is shown in Figure 1 68 Ga-MALAT-1 ASO was synthesized using a single stage method from NOTA-ASO. We obtained chemical yields of 98% in 10 min using HEPES. Mild radiolabelling conditions were required for oligonucleotide radiolabelling due to the highly acidic conditions of the reaction system at high temperatures. Lower yields were observed for the sodium acetate buffer (83% after 30 min). HEPES at room temperature for thus selected for all further experiments. The specific activity of the tracer can reach 100±50 MBq/nmol. As shown in Figure 2, the radiochromatogram displayed a single radioactive and UV peak at 15 min. 68 Ga-MALAT-1 ASO was successfully synthesized with radiochemical purity greater than 98%. The synthesized 68 Ga-MALAT-1 ASO showed high levels of clarity and purity. The pH value was 6-7.
The 68 Ga-oligonucleotide remained stable in PBS at room temperature for 2 h by repeated radio-HPLC analyses, and the radiochemical purities were 99.7%, 99.5%, and 99.1% (all with radiochemical purity greater than 98%) at 30 min, 60 min and 120 min, respectively. The radiochemical purities were 99.9%, 99.6%, and 99.1% in PBS at 37°C analyses at 30 min, 60 min and 120 min, respectively.

The biodistribution of 68 Ga-MALAT-1 ASO in xenograft-bearing nude mice
To investigate the tumour uptake of 68 Ga-MALAT-1 ASO in MHCC-LM3 xenograft-bearing nude mice, tumours, blood, and tissue/organs were excised for the measurement of radioactivity. The tumour distribution was 3.04 ± 0.11%ID/g at 30 min, 2.04 ± 0.04%ID/g at 60 min, and 0.40 ± 0.10%ID/g at 120 min, respectively. The tumour-to-blood (T/B) and tumour-to-muscle (T/M) ratios in the non-blocked and blocked groups were calculated for 68 Ga-MALAT-1 ASO (Table 4). Table 3 shows the time-activity relationship derived from the VOI analysis in the main organs (heart, liver and kidney) in normal mice after intravenous administration by micro PET. Figure 5 shows representative PET images. 68 Ga-MALAT-1 ASO was observed to bind to MHCC-LM3 tumours peaked at 30 min post-injection, then declined over time. The overall image contrast was reduced starting at 60 min.

Micro PET Imaging
PET analysis was used to measure blood clearance rates and the tissue specific uptake of 68 Ga-MALAT-1 ASO. All Micro-PET findings were consistent with the biodistribution studies. As seen in Table   5, the tumour/non-tumour (T/NT) ratios were shown as a function of time. The tumour/muscle ratios were 1.639, 2.397, 2.342, 1.592, 1.445, and 1.254 at 30, 60, 120, 240, 360, and 480 min, respectively.
In the blocked group, study was performed in which 100 nmol of unlabelled PO MALAT-1 ASO was intravenously administered 30 min before the intravenous injection of 68 Ga-MALAT-1 ASO. The results showed that the unlabelled ASO significantly reduced 68 Ga-MALAT-1 ASO tumour uptake (p=0.002), confirming the specificity of the labelled probe ( Figure 6). No tumor uptake was observed in the sense group.

Discussion
Since the association between MALAT-1 and NSCLC has been identified, its important roles in cancer have been considered a paradigm. The expression of MALAT-1 was found to be upregulated in numerous types of tumours, and MALAT-1 promotes tumour cell growth and metastasis [6] . MALAT-1 is upregulated in hepatocellular carcinoma where it activates Wnt signalling and SRSF1, both of which promote cancer growth [7] . Guo and colleagues [8] demonstrated that MALAT-1 regulates caspase-3, caspase-8, Bax, Bcl-2, and BclxL in cervical cancer cells to promote growth and survival. Tano and coworkers [9] showed that MALAT-1 enhances cell motility through upregulating cell motility-related genes including CTHRC1, CCT4, HMMR and ROD1 both at the transcriptional and post transcriptional level. Ying et al. [10] reported the pro-metastatic role of MALAT-1 in bladder cancer. In these cells,
MALAT-1 can be targeted therapeutically in vivo through a number of approaches, including RNAi, antisense oligonucleotides, and small molecule inhibitors [18] . Compared with siRNAs, antisense oligonucleotides provide higher silencing efficiencies due to their high levels of nuclear targeting, to which cellular lncRNAs translocate. [12] . Gutschner and colleagues [20] showed a loss of tumor growth in the lungs of MALAT-1 ASO vs. control ASO groups, concluding that MALAT-1 targeting with ASO represents a novel therapeutic approach for the prevention of lung cancer metastasis. Our previous study identified 5' (Cy5.5)-MALAT-1 ASO, a near-infrared molecular imaging probe, showed promising approach for specific detection of tumours over-expressed MALAT-1 in vivo. The probe can selectively bind to MHCC-LM3 cells and be absorbed by cells, and its retention is time-dependent and concentration-dependent [3] .
To develop novel PET molecular imaging probes, it is necessary to consider its chemical yield and radiochemical yield. In the present study, the chemical yield using HEPES buffer reached 98% within 10 min, and 68 Ga-MALAT-1 ASO was successfully synthesized with radiochemical purity greater than 98%. The 68 Ga-oligonucleotide remained stable in PBS and serum, and the radiochemical purity was higher than 98% for 120 min, which meets the requirements of a molecular probe to be used for further animal exploratory research.
LogP represents a compounds lipid-water partition coefficient, which is the ratio of the concentration of a compound in equilibrium between the lipid phase and the water phase. The smaller the lipidwater partition coefficient is, the more soluble the compound is in water, whereas the conversely, less soluble it is in the lipid phase. In this study, the lipid-water partition coefficient of the 68 Ga-MALAT-1 ASO radioactive probe was calculated (logP=-2.5287 ± 0.1906), suggesting that the 68 Ga-MALAT-1 ASO molecular probe had strong water solubility, which inferred that the molecular probe was highly radioactive in the urinary system, mainly through excretion by the urinary system, and this conclusion was further confirmed by the study of normal animals in vivo.
In this study, a positron probe of 68 Ga-MALAT-1 ASO for high expression of MALAT-1 was synthesized.
The results of the biodistribution studies disclosed that 68 Ga-MALAT-1 ASO initially distributed to the liver, but then predominantly accumulated in the kidneys (~ 30 min). The signal then decreased over time. After 120 min, over 30% of the signal (3.30 ± 1.02%ID/g) remained in the kidney compared to 30 min (8.82 ± 1.63%ID/g). This demonstrated that 68 Ga-MALAT-1 ASO was mainly excreted by the urinary tract and hepatobiliary system. Micro PET images from the MHCC-LM3 cell xenograft mouse models (Fig. 5) provided further evidence for the in vivo tumor targeting of 68 Ga-MALAT-1 ASO. The 68 Ga-MALAT-1 ASO uptake in the tumour showed an increased tumour/muscle ratio from 30, 60, to 120 min, which were1.639, 2.097, and 2.362, respectively, and then deceased over time. The in vivo experiments in tumour-bearing mice showed that the radioactivity uptake in the tumour tissues was the highest 30 min after intravenous injection of 68 Ga-MALAT-1 ASO (3.04 ± 0.11% ID/g), but the ratio of tumour/muscle (2.397 ± 0.304) after 60 min was significantly higher than the ratios at the other observation times.
For tumor probe development, high specificity is an important prerequisite. We performed competition assays to further confirm the in vivo specificity of the probe. Before injection of 68 Ga-MALAT-1 ASO, the competitive inhibition test was carried out by injecting MALAT-1 ASO. The uptake of 68 Ga-MALAT-1 in the tumour tissue significantly decreased (60 min, 0.73 ± 0.00%ID/g), and the tumour/blood ratio and tumour/muscle ratio both decreased significantly (0.58 ± 0.61, 0.66 ± 0.57, respectively), suggesting that this is a target-specific probe. Similar to a previous study, the kidney signal was high in the blocking experiment, suggesting that the probe was primarily cleared from the urinary tract. Moreover, it is necessary to conduct further research on the modification of molecular structure in the future to reduce the radioactive uptake by the kidney and increase the radioactive uptake by the tumour tissue.

Conclusions
In the present research, we present a radiolabelled 68 Ga-MALAT-1 ASO to image liver cancer in the MHCC-LM3 tumour model by a facile method. The radiolabelling was completed within 10 min, and the radiochemical yield reached 98%. The biodistribution of 68 Ga-MALAT-1 ASO was characterized by rapid blood clearance through the urinary system. In vivo PET imaging further confirmed that 68 Ga-MALAT-1 ASO had high tumour uptake (3.04 ± 0.11%ID/g) 30 min after i.v. injection. Moreover, the pharmacokinetic parameters of 68 Ga-MALAT-1 ASO were obtained and showed a fast CL. Therefore, the high accumulation of 68 Ga-MALAT-1 ASO in tumours expressing MALAT-1 demonstrates that the radio compound can be used as a potential positron molecular probe.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests. The funding body only provided financial support, and the design of the study, analysis and interpretation of data and the manuscript writing were independently completed by the authors of   Table 3.Time-activity profile derived from the VOI analysis in the main organs (heart, liver and kidney) in normal mice after intravenous administration by micro PET (n = 5)

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.