Development of novel long noncoding RNA MALAT1 near-infrared optical probes for in vivo tumour imaging

With the advent of next-generation sequencing technology, there is rapidly increasing interest in long noncoding RNAs (lncRNAs). The objectives of this study were to develop a novel lncRNA MALAT1 near-infrared optical probe, to evaluate the characteristics of this optical imaging probe in vitro and to determine whether it can be used for imaging MALAT1 expression in malignant tumours in vivo. Conjugation of Cy5.5 to MALAT1 ASO was accomplished using standard NHS (N-hydroxysuccinimide) ester procedures, and the labelled MALAT1 ASO was purified with a Glen-Pak DNA Purification Cartridge and reversed-phase high performance liquid chromatography (HPLC). The in vitro cellular uptake results showed that the percentage of cell binding increased with an increasing final concentration and increased with increasing incubation time for the MHCC-LM3 tumour cell flow cytometry analyses. in vivo optical imaging exhibited 5’ (Cy5.5)-MALAT1 ASO uptake in the tumour with a maximum at 30 min p.i. that slowly washed out over time. High contrast to normal tissue was gradually observed from 4 h to 48 h p.i. Tumour-to-normal ratios of fluorescence intensities were plotted as a function of time. The in vivo competition assay showed little uptake of the probe into the tumours at any time point, indicating effective competition, selectivity of probe binding and retention by tumours in vivo. Our proposed Cy5.5 labelling of MALAT1 ASO can serve as a potent optical probe for in vivo imaging of tumour expressing MALAT1. Importantly, the successful development of optical probes provides a basis for specific molecular diagnoses in the field of lncRNAs.


INTRODUCTION
With the advent of next-generation sequencing technology, there is rapidly increasing interest in long noncoding RNAs (lncRNAs). Despite the outdated opinion that lncRNAs may be mere transcriptional noise, evidence suggests that lncRNAs play a vital role in regulating gene expression, cell metabolism, tumour development and progression [1]. The long noncoding RNA Malat1 (Metastasis-Associated Lung Adenocarcinoma Transcript 1), also known as MALAT-1 or nuclear-enriched abundant transcript 2 (NEAT2), is one www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 49), pp: 85804-85815 Research Paper of the few biologically well-studied lncRNAs, located on chromosome 11q13, and it was originally found to be overexpressed in early-stage non-small cell lung cancer (NSCLC) [2]. Recently, MALAT1 was found to be overexpressed in many human carcinomas, including hepatocellular carcinomas, pancreatic cancer, bladder cancer, breast cancer, colorectal cancer, gastric cancer, and osteosarcoma [3,4], suggesting that MALAT1 dysregulation is implicated in the development of many types of cancers [5]. Inhibition of MALAT1 expression may suppress cell proliferation, migration, invasion, and the epithelial-mesenchymal transition, and it may induce cell apoptosis and G2/M cell cycle arrest in vitro [6]. These associations suggest that targeting MALAT1 may have important clinical implications because it selectively affects cancer cells or residual cancer cells. Several studies demonstrated that MALAT1 may be a promising target for therapeutic potential in vivo [7]. For example, Gutschner et al. [8] reported that injection of antisense oligonucleotide (ASO) into subcutaneous tumours of nude mice can effectively inhibit MALAT1 expression in vivo and block the metastasis of lung cancer cells. Based on these studies, we hypothesised that MALAT1 is a potential target for a specific molecular diagnosis.
Functional and non-invasive molecular imaging techniques, such as optical imaging, nuclear imaging, and magnetic resonance imaging (MRI), are gradually being incorporated into every aspect of cancer management. Of these, in vivo optical imaging has advantages of not requiring ionising radiation, low cost, and easily generated optical imaging probes [9]. Additionally, optical molecular imaging techniques have become essential tools for providing unique insights into disease pathogenesis, drug development, effects of therapy, and the pharmacokinetic behaviour of drug candidates. In this way, optical imaging is making a substantial impact on basic and translation medical research [10,11]. Therefore, the development of optical imaging molecules that are selective for tumours for in vivo studies is becoming an important field for cancer research [12].
The purpose of the study was to explore the feasibility of development of molecular probes for lncRNA. This study reports the method for preparing molecular optical imaging probes (antisense probes with a Cy5.5 emitter on its 5′equivalent end) for specific targeting of overexpressing MALAT1 tumours cells. We evaluated the characteristics of this optical imaging probe in vitro and determined whether it can be used for the imaging of MALAT1 expression in malignant tumours in vivo.

RESULTS
The schematic molecular structure of the 5' (Cy5.5)-MALAT1 ASO conjugate is shown in Figure 1. The conjugation of dyes to the ASO was accomplished through standard NHS-ester procedures, and the labelled ASO was purified using a Glen-Pak DNA Purification Cartridge and the ion-pair reversed-phase HPLC method. The isolated fluorescent modified oligonucleotides showed very high purity, nearly 100%, as shown in Figure   Figure 1: Scheme of Cy5.5 labelled MALAT1 ASO synthesis. www.impactjournals.com/oncotarget 2, and were suitable for further study. ESI high-resolution mass spectrometry (M+H) − for 5' (Cy5.5)-MALAT1 ASO (calculated, 7227), and m/z=6149 (M+H) − for MALAT1 ASO. Elution times on the HPLC column were 16.61 min for 5' (Cy5.5)-MALAT1 ASO, and 16.28 min for MALAT1 ASO.

In vitro cellular uptake results
To determine whether the new imaging probes result in specific cellular uptake, uptake studies in vitro were performed. The binding specificity of 5' (Cy5.5)-MALAT1 ASO to tumour cells was determined using fluorescence activated cell sorter (FACS) studies.   Figure 4 under fluorescence microscope.
The in vitro cellular distribution study showed that 5' (Cy5.5)-MALAT1 ASO is specific and selective for MHCC-LM3 cells using a fluorescence microscope, and the accumulation of 5' (Cy5.5)-MALAT1 ASO is retained in the nucleus ( Figure 5).

In vivo fluorescence imaging
Fluorescence images of tumour-bearing mice administered 5' (Cy5.5)-MALAT1 ASO probes are shown in Figure 6. 5' (Cy5.5)-MALAT1 ASO uptake in the tumour was maximum at 30 min p.i. and slowly washed out over time. High contrast to background tissue was gradually observed from 4 h to 48 h p.i., whereas normal tissue had faster probe binding and washout. Tumourto-normal ratios of fluorescence intensities were plotted as a function of time, and were 1.03±0.08, 1.04±0.07, 1.05±0. 22 Figure 6J). At early time points, the signals from mouse urine were strong but not retained and became minimal at later time points, suggesting that the probe is mainly excreted from the urinary tract.
To determine whether the in vivo selectivity of probe binding is retained in the tumours, we also performed an in vivo competition assay. These studies showed little uptake of the probe in the tumours at any time points, indicating effective competition and selectivity of probe binding and retention by the tumours in vivo ( Figure 7). Interestingly, signals from the mouse kidney were increased after administration and slowly washed out over time.
Furthermore, excised organs were evaluated ex vivo and quantified using ROI that encompassed the entire organ. The 5' (Cy5.5)-MALAT1 ASO was predominantly taken up by the MHCC-LM3 tumour at 24 h p.i ( Figure  8), as seen in the imaging results in vivo. In the blocked group, the unlabelled MALAT1 ASO reduced the overall 5'(Cy5.5)-MALAT1 ASO probe uptake, which is significantly lower in the tumour (p=0.002), suggesting the targeted specificity of the 5'(Cy5.5)-MALAT1 ASO probe. The kidneys showed increased uptake of the 5' (Cy5.5)-MALAT1 ASO probe, which was identical to the imaging in vivo. The contrast ratios of tumour to normal organs for non-blocked and blocked groups were calculated and presented in Figure 8C. Comparison of data between groups demonstrated that 5' (Cy5.5)-MALAT1 ASO has a tumour-to-muscle ratio of 17.39 ± 2.8) at 24 h p.i. in the non-blocked group and a significantly decreased ratio (2.88 ± 0.11) in the blocked group ( Figure 8D). In addition, the non-blocked group ratio of tumour-to-kidney and tumour to-liver uptake at 24 h p.i. was calculated as 15.63 ± 1.36 and 14.60 ± 1.82, respectively, whereas the corresponding values for the blocked group were 0.14 ± 0.03 and 2.02 ± 0.30, respectively.

DISCUSSION
LncRNAs are a group of RNAs that do not encode proteins, and the lengths of these molecules are greater than 200 nucleotides [16]. Recent observations showed that lncRNAs represent a new frontier in the molecular biology of complex organisms as it is increasingly evident that they are involved in the regulation of almost every stage of gene expression, as well as implicated in cancer onset, progression and a variety of disease states [1], showing a promising future for lncRNA  studies. MALAT1, a highly conserved nuclear-abundant lncRNA of approximately 8000 nucleotides, received great interest among lncRNAs since its discovery as an independent prognostic biomarker for early stage metastasising NSCLC, and many studies showed the mechanisms of MALAT1 in regulating gene expression, cell proliferation, invasion, and tumour formation in a broad range of human cancers, including pancreas, breast, colon, liver, oesophageal and prostate cancers [17,18,19]. For example, our previous study [15,19] revealed that MALAT1 overexpression is a predictive factor for tumour recurrence following liver transplantation in hepatocellular carcinoma. The depletion of MALAT1 by siRNA in the HepG2 cell line resulted in a reduction of cell viability, mobility and invasiveness, as well as an increase of sensitivity to apoptosis. Yang et al. [20] suggested that MALAT1 expression in colorectal cancer tissues of human patients with lymph node metastasis was higher than those without metastasis, and MALAT1 may promote colorectal cancer development via its target protein AKAP-9. Several studies showed the aberrant expression of MALAT1 in tumour tissues compared with normal tissues and its association with clinical progression in human cancers, suggesting an important role in cancer pathogenesis and progression [5,21].
MALAT1 is also an independent prognostic marker for poor outcomes and patient survival in several cancer types, including stage I non-small cell lung cancer and hepatocellular carcinoma [22,23]. Ma et al. [24] reported that the over-expression of MALAT1 in glioma tissues was positively correlated with grade and tumour size, suggesting that MALAT1 may serve as an authentic prognostic biomarker for patients with glioma. Shuai et al. [24] performed a meta-analysis with a total of 2094 patients from 17 studies to assess the effects of MALAT1 on cancer prognosis and lymph nodes, and the results indicated that the upregulation of MALAT1 expression was significantly associated with poor disease-free survival (HR = 2.29, 95 % CI 1.24-3.35), and recurrencefree survival (HR = 2.09, 95 % CI 0.81-3.37).
MALAT1 has many characteristics that make it attractive as a biomarker and therapeutic targets [1,25,26]. MALAT1 can be targeted therapeutically by a variety of approaches, including antisense oligonucleotides, RNAi mediated gene silencing, small molecule inhibitors and by gene therapy [7]. Schmidt et al. [27] reported that RNAi-mediated suppression targeting MALAT1 suppressed migration and clonogenic growth in A549 NSCLCs in vivo, exhibiting significant anticancer effects. ASO, which is relatively smaller than siRNAs, give a higher rate of delivery into the nucleus where most lncRNAs are confined [26]. Gutschner et al. [28] observed an accumulation of MALAT1 ASO in both tumour and tumour-associated stromal cells and effectively reduced both human and mouse MALAT1 expression compared with control ASO. Tripathi et al. [13] reported that compared to control cells, MALAT1-depleted cells using ASO showed a consistent reduction in the RNA and protein levels of several of the mitotic genes analysed. To the best of our knowledge, up to now no molecular probes of lncRNA have been reported.
Considering the above factors, we hypothesised that MALAT1 is a potential molecular target and that fluorophore-conjugated MALAT1 ASO would generate a novel molecular probe for the optical imaging of tumours. To create the probe, we selected the Cy5.5 dye, which is used as a promising contrast agent for the in vivo demarcation of tumours, to conjugate with the MALAT1 ASO [14,29]. The synthesis method is convenient, high purity is obtained (nearly 100%) using a Glen-Pak DNA purification cartridge and HPLC method, and it is suitable for further study.
The present study identified a near-infrared molecular imaging probe for cells over-expressing MALAT1 that allows for non-invasive specific detection of tumours in vitro and in vivo. in vitro cellular uptake experiments with 5' (Cy5.5)-MALAT1 ASO showed that it can selectively bind to and be taken up by MHCC-LM3 cells, and this retention is time-dependent. FACS analyses revealed that the percentage of cell binding (M2) increased with increasing incubation times from 30 min to 8 h, with a positive correlation existing between uptake and increasing incubation time (r=0.97). By contrast, 5' (Cy5.5)-MALAT1 ASO bound to MHCC-LM3 cells in a dose-dependent manner, and there was increased uptake of MALAT1 antisense oligonucleotides by MHCC-LM3 cells with increasing concentrations, ranging from 20 nM to 250 nM. In addition, the intracellular distribution of 5' (Cy5.5)-MALAT1 ASO was also examined, and the results Fluorescence intensity ratio of tumour-to normal tissue based on the ROI analysis (D). The error bar is calculated as the standard deviation.
The whole-body fluorescent images of the MHCC-LM3 cell xenograft mouse model presented in Figure 6 provide further evidence to support the hypothesis that the 5' (Cy5.5)-MALAT1 ASO can target tumours in vivo. The 5' (Cy5.5)-MALAT1 ASO uptake in the tumour showed a maximum at 30 min p.i. and slowly washed out over time. High contrast to background tissue was gradually observed from 4 h to 48 h p.i. The normal tissue demonstrated rapid probe uptake. However, probe washout in the tumour is much slower than in normal tissue, leading to excellent tumour-to-normal tissue contrast at later time points. 5' (Cy5.5)-MALAT1 ASO exhibited high contrast tumour-to tissue ratios for the imaging of dissected tissues and organs, which were consistent with the in vivo imaging findings.
For the development of new molecular probes for in vivo tumour imaging, high specificity binding is an important property [32]. To further demonstrate that binding of our 5' (Cy5.5)-MALAT1 ASO probe is specific for MALAT1 in vivo, competition studies were conducted by injecting unlabelled MALAT1 ASO before injecting 5' (Cy5.5)-MALAT1 ASO. Significantly reduced tumour uptake of 5' (Cy5.5)-MALAT1 ASO was observed for the blocked group at all times p.i. (P < 0.05), indicating that it is a target-specific probe. In the blocked experiments, signals from the mouse kidney were gradually increased, reaching a peak 24 h after administration, and then gradually decreasing, suggesting that the probe is mainly excreted from the urinary tract.
In summary, our results provide evidence that Cy5.5 labelling of MALAT1 ASO is a convenient approach and that these novel optical probes can serve as specific optical probes for the in vivo imaging of tumours expressing MALAT1. More importantly, the successful development of optical probes provides for specific molecular diagnoses in the field of lncRNAs, which deserve further study.

General
Single strand 20-base DNA sequences were modified as phosphorothioates (Shanghai Integrated Biotech Solutions Co., Ltd, China), and the base sequence used was as follows: GGGAGTTACTTGCCAACTTG (MALAT1 ASO) [13]. The Cy5.5 NHS ester was purchased from GE Healthcare Bioscience/GE. The Glen-Pak DNA purification cartridge was obtained from Glen Research (Sterling, VA). Triethylammonium acetate solution (2 M, TEAA), pH 7.0, was purchased from Glen Research (Sterling, VA). All other chemicals were from Sigma-Aldrich (St. Louis, MO).
The HPLC system was equipped with a quaternary pump, variable wavelength detector, vacuum degasser, and ChemStation for system control and chromatography analysis. HPLC was performed using a C18 column (Zorbax ODS 4.6 * 250 mm). For the HPLC linear gradient, the following buffers were used: A, 2% acetonitrile in 0.1 M TEAA, pH 7; B, 50% acetonitrile in 0.1 M TEAA, pH 7. The column was eluted at 1 mL/ min at room temperature. Mass spectra were obtained on a Q-Tof premier-UPLC system equipped with an electrospray interface (ESI) (Waters, USA). The product was collected, lyophilised, and stored in the dark at -20°C until use.

Synthesis of 5' (Cy5.5)-MALAT1 ASO
Ten nanomoles of the MALAT1 ASO (GGGAGTTACTTGCCAACTTG) bearing 5'-aminoalkyl linkers was dissolved in 15 μL of 0.1 M NaHCO 3 . A volume of 7 μL of 49 mM Cy5.5-NHS in DMSO was added, and the mixture was incubated at room temperature for 12 h in the dark [14]. Then, 40 μL of water was added, and the modified MALAT1 ASO was separated from the excess dye using a Glen-Pak DNA purification cartridge according to the manufacturer's recommendations. Highperformance liquid chromatography (HPLC) (1200 series, Agilent, USA) was used for the isolation and purification of the product. After Glen-Pak DNA purification, the modified MALAT1 ASO was precipitated by standard ethanol-sodium acetate treatment and purified by HPLC as described above. The purified modified MALAT1 ASO was concentrated in a centrifugal vacuum concentrator.

In vitro cellular uptake study
The human MHCC-LM3 cell line, which was selected for its high MALAT1 expression [15], was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). It was cultured at 37°C in a humidified atmosphere of 5% CO 2 in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal calf serum with 100 U/mL penicillin and 100 μg/mL streptomycin.

In Vivo and Ex Vivo optical imaging
All animal work was performed in accordance with and approved by the Zhejiang University Institutional Animal Care and Use Committee guidelines. BALB/c nu/ nu mice (female, 20 ± 3 g, 4-to 6-wk-old; Department of Laboratory Animal Science, Zhejiang University) were used in this study. BALB/c nu/nu mice with MHCC-LM3 cell tumours were randomly divided into 3 groups of 5 mice each. The MHCC-LM3 cell xenograft model was generated by subcutaneous injection of 1.0 * 10 7 MHCC-LM 3 cells resuspended in 100 μL of PBS into the right axilla, and the cells were allowed to grow 3-5 weeks until tumours were 200-500 mm 3 in volume. The mice were maintained using a standard diet, bedding, and environment, with free access to food and drinking water.
In vivo fluorescence imaging was performed using a filter set with an excitation of 658 nm and emission of 719 nm on a Clairvivo OPT plus (Shimadzu Co., Kyoto, Japan). Identical illumination settings (fluorescent dye, excitation mode, exposure time, and field of views) were used for acquiring all images. Fluorescence images were acquired at 15 min, 30 min, 1 h, 2 h, 4 h, 8 h, 24 h, and 48 h after injection of 5' (Cy5.5)-MALAT1 ASO using 1 s exposure time, anesthetised with 2% isoflurane. The non-blocked mice (n=5) received 100 nmol of 5'(Cy5.5)-MALAT1 ASO injected into each mouse via tail vein and were subjected to optical imaging at various time points post-injection. For the blocked group of mice (n=5), unlabelled MALAT1 ASO (100 nmol) was first injected, and 100 nmol 5' (Cy5.5)-MALAT1 ASO was injected 30 min later via tail vein. Ellipsoid regions of interest (ROI) of equal area were drawn on the tumour and on the contralateral sites of the left axilla, and the fluorescent intensity was quantified (C/mm 2 ).
At the end of the in vivo imaging experiments, the mouse was sacrificed by cervical dislocation under isoflurane anaesthesia. The tumours and major organs or tissues (heart, liver, spleen, lung, kidney, muscle, bone, and blood) were collected for ex vivo imaging of the biodistribution of the fluorescent probe, and the fluorescence intensity for each sample was reported (C/mm 2 ).

Statistical analysis
Data are presented as the means ± SD (standard deviation) of n independent measurements. Statistical analysis was performed using Student's t test. A p value less than 0.05 was considered statistically significant. Analyses were performed using SAS v8.0 (SAS, Cary, NC, USA).

Author contributions
Dong MJ and Wang CQ are the principal investigators for the study. Dong MJ also wrote the manuscript and collected the data. Liu ZF, and Wang YH conceived of this study and carried out the major part of this project. Wang GL participated in this project and collected and analyzed the data.