Hepatocellular Carcinoma Cells Carrying a Multimodality Reporter Gene for Fluorescence, Bioluminescence, and Magnetic Resonance Imaging In Vitro and In Vivo

ARTICLE IN PRESS

Original Investigation

Hepatocellular Carcinoma Cells Carrying a Multimodality Reporter Gene for Fluorescence, Bioluminescence, and Magnetic Resonance Imaging In Vitro and In Vivo Xiaoxiao Qin, M.D., Xiaojun Hu, M.D., Chun Wu, M.D., Ph.D., Mingyue Cai, M.D., Zhengran Li, M.D., Ph.D., Lina Zhang, M.D., Liteng Lin, M.D., Wensou Huang, M.D., Kangshun Zhu, M.D., Ph.D. Rationale and Objectives: The study aimed to evaluate the feasibility of imaging or tracking hepatocellular carcinoma cells by modifying these cells to carry a multimodality reporter gene, enabling fluorescence, bioluminescence, and magnetic resonance imaging (MRI) in vitro and in vivo. Materials and Methods: HepG2 cells were labeled with the enhanced green fluorescent protein (EGFP)/luciferase2/ferritin—the multimodality reporter gene (labeled HepG2 cells). The labeled and unlabeled HepG2 cells were cultured in vitro and then injected subcutaneously into mice as a hepatoma model in vivo. The expressions of EGFP, luciferase2, and ferritin in HepG2 cell suspensions and hepatoma model were investigated using fluorescence, bioluminescence, and MRI. Results: Individual HepG2 cells expressing EGFP were identified under blue laser excitation. The linear coefficient between the optical signal intensity of luciferase2 and the number of labeled cells was 0.993. MRI was used to distinguish the T2* signal of 2 × 107 cells/mL between the labeled (6.67 ± 1.88 ms) and unlabeled cells (10.66 ± 2.22 ms) (P = 0.034). In vivo, individual HepG2 cells expressing EGFP in frozen sections were observed. Labeled cells expressing luciferase2 have been detected since the second day after injection, and the bioluminescence increased with the tumor size. The T2* signal was significantly different between the labeled (6.04 ± 1.60 ms) and unlabeled cells (17.06 ± 2.17 ms) (P < 0.001). Conclusions: A multimodality reporter gene consisting of EGFP, luciferase2, and ferritin was successfully integrated into the HepG2 cell genome via a lentiviral vector and was highly expressed in the daughter cells. These cells could be detected by fluorescence, bioluminescence, and MRI in vitro and in vivo. Key Words: Hepatocellular carcinoma; Gene imaging; Enhanced green fluorescent protein; Luciferase2; Ferritin. © 2016 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.

INTRODUCTION

Acad Radiol 2016; ■:■■–■■ From the Department of Minimally Invasive Interventional Radiology, The Second Affiliated Hospital of Guangzhou Medical University, 250 East Changgang Road, Guangzhou 510260, Guangdong Province (X.Q., W.H., K.Z.); Department of Radiology, The First Affiliated Hospital, College of Medicine, Zhejiang University, Hangzhou, Zhejiang Province (X.Q.); Department of Radiology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong Province (X.H., C.W., M.C., Z.L., L.Z., L.L.); Interventional Radiology Institute, Sun Yat-sen University, Guangzhou, Guangdong Province, China (C.W., M.C., Z.L., L.L.). Received January 21, 2016; revised June 27, 2016; accepted July 1, 2016. Address correspondence to: K.Z. e-mail: [email protected] © 2016 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.acra.2016.07.009

H

epatocellular carcinoma (HCC) is one of the most common forms of cancer and is the fifth most prevalent cause of tumor-related death globally (1). The absolute number of new HCC cases continues to increase, and HCC is projected to remain as the largest cancer burden for the next several decades. Although advancements in gene diagnosis and treatment modalities have helped improve the survival and prognosis of HCC patients, HCC diagnosis predominantly depends on imaging studies such as computed tomography and magnetic resonance imaging (MRI), limiting the early diagnosis of the disease (2). Several biomarkers have been proposed (3), but no available theory can explain how these malignant cells are capable of evading immune 1

QIN ET AL

clearance or when and where they prefer to reside and metastasize. This limited knowledge has hindered our understanding of the pathologic changes caused by HCC and our identification of potential therapeutic targets. Thus, tracing or imaging HCC cells in vitro and in vivo at the molecular level is a challenge for the treatment of HCC. Gene imaging (4,5), which is a distinct imaging from exogenous tracers, can be visualized stably in cells, enabling the precise localization of labeled cells. Both optical imaging (6) and MRI (7) facilitate the noninvasive tracking of transplanted cells in living organisms. Due to its advantages in terms of sensitivity and specificity, bioluminescent reporters are among the most widely used, as they can provide qualitative and quantitative information about the cells (8). However, due to photo attenuation and light scattering within deep tissues, bioluminescence imaging (BLI) lacks tomographic resolution and limits its application to small animals or superficial areas (9). Ferritin has been used as a T2*-weighted MRI probe to dynamically track cells in vivo because of the low-intensity signals generated from T2*-weighted MRI (10,11). In contrast to optical imaging, MRI displays the advantages of unlimited depth penetration, high spatial resolution, and multiple imaging parameters; however, MRI cannot distinguish whether or not the transplanted cells are surviving (12). Apparently, the combination of these two imaging tools may overcome their disadvantages while benefiting from each of their advantages to provide improved cell tracking in vivo (13). Moreover, because the expression of enhanced green fluorescent protein (EGFP) can be detected in single cells, the sensitivity of EGFP enables its function as a sorting and qualitative marker for cell tracing. Therefore, the expression of reporter genes that can be used for several imaging modalities in targeted cells would have complementary advantages in sensitivity and specificity and would provide a complete and precise description of the biological processes in these cells. A lentiviral vector carrying the optical reporter gene EGFP, the bioluminescent reporter gene luciferase2, and the MR reporter gene ferritin has been established in this study. These reporter genes were successfully integrated into the genome of HCC cells (HepG2) via lentiviral vector transduction (labeled HepG2 cells). The nude mouse HepG2 hepatoma model was established via the subcutaneous injection of labeled or unlabeled HepG2 cells. The expression of EGFP, luciferase2, and ferritin in HepG2 cell suspensions and hepatoma samples was individually investigated using a fluorescence microscope, a Caliper IVIS Lumina XR imaging system, and an MRI system.

MATERIALS AND METHODS Instruments and Reagents

The BD Influx Cell Sorter (BD Biosciences, Two Oak Park, MA) was employed to select the EGFP-positive cells. The Bruker BioSpec 94/20USR (Bruker BioSpin MRI, Ettlingen, BadenWürttemberg, Germany) was employed to detect the expression 2

Academic Radiology, Vol ■, No ■■, ■■ 2016

of ferritin in vivo and ex vivo. The Axio Observer A1 (Carl Zeiss, Göttingen, Lower Saxony, Germany) was employed to detect EGFP in labeled cells. The IVIS Spectrum system (Caliper Life Sciences, Alameda, CA) was employed for in vivo and ex vivo BLI. The plasmid pLenti7.3-cmv-ferritin-IRES-luciferase2SV40-EGFP, which co-expressed EGFP, luciferase2, and ferritin, was used in this study. The ferritin cDNA was donated by Prof Guillem Genove (Department of Biological Sciences and Pittsburgh NMR Center for Biomedical Research, Carnegie Mellon University, Pittsburgh, PA). It was amplified by reverse transcription polymerase chain reaction with ferritin forward (GGGGACAAGTTTGTACAAAAAAGCAGGCTGCCACC ATGACGACCGCGTCCACCTC) and reverse (TTAGGG GGGGGGGAGGGAGAGGGGCTTAGCTTTCATTATCA CTGTCTCCCAGGGT) primer pairs. The pLenti7.3-V5DEST and IRES were purchased from Invitrogen (Carlsbad, CA). The IRES forward primer was TGGGAGACAGTGA TAATGAAAGCTAAGCCCCTCTCCCTCCCCCC, and the reverse primer was GGGCCCTTCTTAATGTTTT TGGCATCTTCCATGGTGGCTGTGGCCATATTATCA TCGTGTTTTTC. The HCC cell line HepG2 was donated by Shuai Xintao Laboratory (Sun Yat-sen University, Guangzhou, Guangdong, China). High-glucose Dulbecco’s modified Eagle medium (DMEM), 0.25% trypsin-ethylene diamine tetraacetic acid (EDTA) solution, fetal bovine serum (FBS), and Dulbecco’s phosphate-buffered saline (PBS) were purchased from Gibco (Life Technologies, Carlsbad, CA). The rabbit polyclonal anti-GFP antibody and the donkey anti-rabbit IgG (H+L) antibody were purchased from Abcam (Cambridge, England, UK). ProLong Gold Antifade Reagent and Lipofectamine 2000 were purchased from Life Technologies. The nuclear fluorescent staining agent 4′,6-diamidino-2-phenylindole (DAPI) was purchased from Roche (F. Hoffmann-La Roche Ltd, Basel, Basel-Stadt, Switzerland). D-luciferin was purchased from Cold Spring Biotech Corp (Taipei, China), and gelatin was purchased from Bio Sharp (Shanghai, China). Matrigel and 1% crystal violet solution were purchased from BD Biosciences. Plasmid Construction

The lentiviral expression vector pLenti7.3-V5-DEST was constructed to carry a multimodality reporter gene consisting of EGFP, luciferase2 (14) (Luc2, Promega Inc, Madison, WI), and ferritin according to the manufacturer’s protocol (Invitrogen). The expression vector pLenti7.3-cmv-ferritinIRES-luciferase2-SV40-EGFP) and ViraPower Packaging Plasmid Mix (Invitrogen, Carlsbad, CA) were cotransfected into 293FT cells (8000 cells/well) using Lipofectamine 2000 (Invitrogen). The culture supernatants were collected, concentrated, and used as a viral stock at a concentration ranging from 2 × 10−4/mL to 2 × 10−8/mL. Culture and Transfection of HepG2 Cells

Human HCC HepG2 cells were grown and maintained in High Glucose/DMEM medium supplemented with 10% FBS,

Academic Radiology, Vol ■, No ■■, ■■ 2016

and the cells were maintained at 37°C under a humidified atmosphere of 5% CO2. HepG2 cells were seeded in a 12well plate at a density of 3 × 104 cells per well 24 hours prior to transfection. The multiplicity of infection (MOI) ranged from 2 to 200. After various complexes were mixed for 30 minutes at room temperature, they were added to each well, and the cells were incubated for an additional 72 hours at 37°C in a CO2 incubator. All processes were conducted in the dark. After the cells were incubated in the vector, they were washed twice with PBS and cultured in 200 mL of HG/DMEM. Cells cultured as usual without transfection (unlabeled) were used as a control for background calibration. The transfection efficiency was evaluated using a confocal laser scanning microscope (Carl Zeiss), and images were concurrently recorded. After the EGFP, luciferase, and ferritin were integrated into some of the genome of some HCC cells (HepG2) via lentiviral vector transduction, we selected the positive cells (labeled HepG2) with flow cytometry: 2 × 106 cells were suspended in 2-mL PBS. When the cells passed the flow cytometry one by one under the blue fluorescence excitation (488 nm), the cells that emitted green fluorescence (505–530 nm) were regarded as EGFP+ and collected in a round-bottom tube filled with 2-mL culture medium. After fluorescence-activated cell sorting, the percentage of EGFP+ cells (with no obvious descends in 2 weeks) increased to 95%. These purified cells were treated as the stable cell line for further experiment. Transwell Migration/Invasion Assay

To evaluate whether the transfection of the lentivirus carrying the multimodality reporter gene affected the behavior of HepG2 cells, we performed an invasion assay to confirm their invasive properties (15). For the migration assay, labeled and unlabeled HepG2 cells were harvested in serum-free HG/DMED, and the cell suspension was added to the upper well of the transwell insert (Corning Inc, Corning, NY) at a density of 1 × 105 cells per well (24-well plate). The lower well was treated with 500 μL of HG/DMEM containing 10% FBS as a chemoattractant. After incubation for 24–48 hours, the cells that penetrated the bottom of the transwell membrane were stained with crystal violet and photographed under a microscope. Then, five microscopic fields for the two groups, respectively, were randomly captured and used for cell counting. To evaluate the effect of transfection on cell migration, the cells were starved using HG/DMEM containing 0.5% FBS for 24 hours and were then collected via trypsinization. Both plates were incubated for 24–48 hours. For the invasion assay, the upper well of the transwell insert was precoated with a layer of Matrigel (1:7 dilution in serum-free medium) before adding the cells. The cells were incubated and followed by crystal violet staining. Fluorescence Analysis of EGFP for In Vitro Targeting of Labeled Cells

The expression of EGFP was evaluated via confocal laser scanning microscopy using an Axio Observer A1 microscope. The

MULTIMODALITY REPORTER GENE IMAGING FOR HCCS

labeled HepG2 cells were seeded at a density of 1 × 103 cells per well in a confocal dish. After a 2-hour incubation at room temperature, the DNA staining agent DAPI (1 mg/mL) was added to the dish for 30 minutes. Subsequently, the observations were performed. Nuclear staining with DAPI was detected under a fluorescence microscope at an excitation of 340 nm and an emission of 488 nm. EGFP was detected under a fluorescence microscope at an excitation of 488 nm and an emission of 505–530 nm using a band-pass filter. The unlabeled control cells were also measured in the same way. Bioluminescence Analysis of Luciferase2 In Vitro

Labeled HepG2 cells were seeded in a black 96-well plate at 4 × 106, 2 × 106, 1 × 106, 5 × 105, 2.5 × 105, and 1.25 × 105 cells per well, and the cells were incubated in D-luciferin (150 μg/mL) at room temperature. Cellular bioluminescence signals were detected using an acquisition time of 1 second. BLI was performed using a Xenogen IVIS Lumina II imaging system and was analyzed using IVIS Living Imaging 4.2 software (Caliper Life Sciences). The BLI results were normalized and expressed as photons per second per centimeter squared per steradian (p/s/cm2/sr) (14). Unlabeled HepG2 cells cultured at the same density were used as a control. Prussian Blue Iron Staining to Detect the Fe3+ Content in Transfected Cells

To increase the extracellular iron content in culture media, both the labeled and unlabeled HepG2 cells were incubated in the media supplemented with ferric ammonium citrate (FAC, 250 μmol/mL) for 5 days, and were washed twice with PBS before being tested. To investigate whether the protein expression of the reporter transgene ferritin produces a functional iron storage molecule, we used Prussian blue iron staining to compare the levels of elemental iron in transformed cells expressing ferritin (16) to those in unlabeled cells. After fixation with 4% paraformaldehyde solution for 10 minutes, the labeled and unlabeled HepG2 cells were washed three times with PBS and then incubated for 30 minutes in a 1:1 solution of 2% aqueous potassium ferrocyanide:2% hydrochloric acid. An optical microscope was used to evaluate the iron staining. MRI Analysis of Ferritin In Vitro

A total of 2 × 107 labeled or unlabeled HepG2 cells incubated in the media supplemented with ferric ammonium citrate (FAC, 250 μmol/mL) for 48 hours and washed twice with PBS before MRI were suspended in 100 μL of 2% gelatin and transferred to a 200-μL Eppendorf tubes. The cells were scanned under a 9.4T Bruker BioSpec 94/20USR imaging system (Bruker BioSpin MRI) equipped with a commercially available volume coil (RF RES 1H 75/40 Q TR, Germany) of 40 mm in diameter. T2*-weighted imaging was performed using a gradient-echo sequence at the following parameters: Echo Time/Repetition Time = 3/200 ms, flip 3

QIN ET AL

angle: 30°, matrix: 256 × 256 × 15, field of view = 30 × 30 mm2, averages 8, and slice thickness: 1 mm. The total scan time was 17 minutes. The signal intensity (SI) of the labeled and unlabeled HepG2 cells was evaluated within a circular 9 mm2 region of interest.

Nude Mouse Hepatoma Model

A total of 28 six- to seven-week old male mice were used in bioluminescence, MRI, and histologic studies. The animal procedures were performed according to the protocol of the National Institutes of Health Guide for Care and Use of Laboratory Animals and were approved by the Sun Yat-sen University Institutional Animal Care and Use Committee on September 10, 2012. Six- to seven-week-old male nude mice (nu/nu) were purchased from WuShi Laboratory Animal Technology Co, Ltd (Fujian, China). The nude mice were subcutaneously injected into the right thigh with 2 × 106 labeled HepG2 cells/100 μL of cell suspension and into the left thigh with 2 × 106 unlabeled HepG2 cells/100 μL of cell suspension. The mice were anesthetized with 10% chloral hydrate 100~150 μg before cell transplant. All measurements were performed under inhalation anesthesia induced using 3–4% isoflurane and maintained using 0.75% isoflurane in 70% O2 and 30% N2O at an air flow rate of 0.2–0.4 L/min.

Fluorescence Analysis of EGFP for In Vivo Targeting of Labeled Cells

EGFP imaging of frozen sections of tumors harvested at 2 weeks after injection was acquired. Frozen tissue sections (5-μm thick) were fixed with 4% paraformaldehyde and subsequently processed for immunostaining. After washing twice with 0.1% Triton X-100 in PBS for 5 minutes, the specimens were blocked with blocking buffer (10% FBS, 0.3% Triton X-100, 0.1 M PBS) for 60 minutes, followed by incubation in a specific rabbit polyclonal antibody against GFP (ab290, Abcam) for 12 hours. Next, the frozen sections were washed three times with PBS, incubated in Alexa Fluor 488 donkey antirabbit IgG (H+L) antibody (A-21206) for 60 minutes while shielded from light, washed with PBS, and then treated with the nuclear stain DAPI (1 mg/mL) for 15 minutes. The sections were mounted using ProLong Gold Antifade Reagent. All experiments were performed at room temperature. Images were obtained via fluorescence microscopy.

Bioluminescence Analysis of Luciferase2 In Vivo

On the first, second, third, seventh, and fourteenth day after transplant, the nude mice were administered 150 mg/kg body weight D-luciferin substrate via intraperitoneal injection and were then anesthetized via isoflurane inhalation. The mice were placed in the imaging chamber, and bioluminescence signals were detected using an acquisition time of 5 seconds. 4

Academic Radiology, Vol ■, No ■■, ■■ 2016

MRI Analysis of Ferritin In Vivo

For in vivo imaging, the mice that were transplanted with tumor cells for 2 weeks were anesthetized, placed on a mechanical ventilator, and maintained on 0.75% isoflurane in 70% O2 and 30% N2O inhalation gas during the imaging sessions. The mice were scanned under a 9.4T Bruker BioSpec 94/20USR imaging system (Bruker BioSpin MRI) equipped with a commercially available volume coil (RF RES 1H 75/40 Q TR, Germany) of 40 mm in diameter. T2*-weighted imaging was performed using a gradient-echo sequence at the following parameters: Echo Time/Repetition Time = 3/200 ms, flip angle: 30°, matrix: 256 × 256 × 15, field of view = 40 × 60 mm2, averages 8, and slice thickness: 1 mm. The total scan time was 25 minutes. The SI of the labeled and unlabeled HepG2 cells was evaluated within a 6 mm2 circular region of interest. Statistical Methods

Correlation analysis between the number of cells and the bioluminescence produced over 1 second following treatment with luciferin was performed by calculating the correlation coefficients. The difference in the T2* values between the two groups was compared using a test in the SPSS 19.0 software (IBM, Armonk, NY). The quantitative data are expressed as the mean ± standard deviation. The means were compared using Student’s t test. P values of <0.05 were considered to be statistically significant. All statistical tests were two-sided. RESULTS Plasmid Construction and Verification and Lentivirus Packaging

The results of DNA sequencing confirmed the accuracy of the gene sequence of the recombinant plasmid pLenti7.3ferritin-IRES-luciferase2-SV40-EGFP, which was successfully constructed without gene mutations or deletions (Fig 1a). Seventy-two hours after the recombinant plasmid was transfected into the 293FT cells, green fluorescence emission (505–530 nm) could be observed in some of the cells under blue laser excitation (488 nm) (Fig 1b), illustrating that the recombinant plasmid had been successfully transferred to some of the 293FT cells. Culture and Transfection of HepG2 Cells

The number of cells emitting green fluorescence in the two wells whose viral load was 2 × 10−7 mL was 16 and 17, resulting in the following virus titer: (17 + 16)/2 = 16.5 TU; 16.5 TU/(2 × 10−7 mL) = 8.5 × 107 TU/mL. At this titer, we found that viral infection was effective at an MOI of 50, as demonstrated by green fluorescence emission from blue laser excitation (Fig 1c). The HepG2 cells appeared in a good state, as no mass scarring of necrotic cells occurred based on the survival of the host HepG2 cells 72 hours after lentiviral transfection (Fig 1c). No increase in the percentage of transfected HepG2

Academic Radiology, Vol ■, No ■■, ■■ 2016

MULTIMODALITY REPORTER GENE IMAGING FOR HCCS

23.60 ± 1.21 per well, and this result was similar to that in the unlabeled cells (24.20 ± 1.16 per well) (P > 0.05) (Fig 1d and e). These results indicated that this transfection had no significant effect on cancer cell migration. The cells in both groups were epithelial-like, extending few processes; this observation confirmed that reporter gene expression had little influence on cell morphology. Fluorescence Analysis of EGFP Imaging In Vitro

Green fluorescence emission (505~530 nm) was detected in the labeled HepG2 cells under blue laser excitation (488 nm) (Fig 2a), but no fluorescence signal was detected in the unlabeled HepG2 cells. Confocal analysis of the labeled HepG2 cells revealed EGFP expression predominantly in the cytoplasm (Fig 2a). This result indicated that EGFP was effectively expressed in the labeled HepG2 cells. Bioluminescence Analysis of Luciferase2 In Vitro

The bioluminescence over 1 second following treatment with D-luciferin was 1.039 × 10 9 , 4.035 × 10 8 , 2.388 × 10 8 , 1.207 × 108, 5.677 × 107, and 3.168 × 107 in the wells containing 4 × 106, 2 × 106, 1 × 106, 5 × 105, 2.5 × 105, and 1.25 × 105 cells, respectively (Fig 2b). With the reduction of the cell number, the photon count decreased, indicating a linear relationship between the cell number and the bioluminescence signal (r = 0.993) (Fig 2b). At a cell density of 1.25 × 105 cells/mL, the bioluminescence was undetectable. In the wells containing unlabeled HepG2 cells at the same densities, no bioluminescence signal was detected. Figure 1. (a) Schematic diagram of the recombinant plasmid. (b) At 72 hours after transfecting 293FT cells with the recombinant plasmid, the cells had grown to confluence (left), and green fluorescence emission was detected in some of the cells under blue laser excitation (right), indicating that the recombinant plasmid was successfully transferred to some of the 293FT cells (×40; scale bar: 200 μm). (c) At 72 hours after lentiviral transfection of HepG2 cells. At a multiplicity of infection of 50, the HepG2 cells were in a good state, as no mass scarring of necrotic cells occurred (left). Green fluorescence emission was observed under blue laser excitation on the same field of view (right) (×100; scale bar: 100 μm). (d,e) The number of labeled HepG2 cells (left) that succeeded in migrating across the micro-pores of the base membrane was 23.60 ± 1.21 per well, similar to that of the unlabeled cells (right) (24.20 ± 1.16 per well) (×200; scale bar: 100 μm); this difference was not significant (P > 0.05). The cells in both groups were epithelial-like, extending few processes.

cells was observed at an MOI > 50. The labeling efficiency was 5.7% based on flow cytometric analysis of EGFPpositive cells (labeled HepG2 cells). After fluorescenceactivated cell sorting, the percentage of labeled HepG2 increased to 90.8%.

Prussian Blue Iron Staining for Iron Assessment In Vitro

Prussian blue staining consistently showed extensive blue staining in the cytoplasm of the labeled HepG2 cells; alternatively, the blue granules in the unlabeled HepG2 cells were fewer and were more scattered (Fig 2c). The cells in both groups were adherent, epithelial-like monolayers that formed small aggregates; this result confirmed that reporter gene expression increased the iron content of the cells. MRI for Ferritin In Vitro

The average T2* value of the four fields of labeled HepG2 cells was 6.67 ± 1.88 ms, which was significantly lower than that of the unlabeled cells (10.66 ± 2.22 ms) (P = 0.034) (Fig 2d). These results indicated that the labeled HepG2 cells were able to transport and integrate more Fe3+ than the unlabeled HepG2 cells. Fluorescence Analysis of EGFP Imaging In Vivo

Cancer Cell Migration

The number of labeled HepG2 cells that succeeded in migrating across the micro-pores of the base membrane was

For in vivo detection, nude mice harboring a tumor of a size beyond 1 × 1 cm and lacking necrosis were sacrificed via an overdose of anesthesia (4% chloral hydrate, 200 μg), on the 5

QIN ET AL

Academic Radiology, Vol ■, No ■■, ■■ 2016

Figure 2. (a) Green fluorescence emission was observed in individual labeled HepG2 cells under blue laser excitation, and enhanced green fluorescent protein (EGFP) was predominantly expressed in the cytoplasm (left); nuclei counterstained with DAPI are shown in blue (middle), and the fused image of EGFP fluorescence and nuclear staining (right) (×630; scale bar: 50 μm). (b) With the reduction in the cell number, the photon count decreased, revealing a linear correlation between the number of cells and the bioluminescence signal (r = 0.993). (c) Prussian blue staining showed extensive blue staining in the labeled HepG2 cells, whereas the blue granules in the unlabeled HepG2 cells were fewer and were more scattered (×200; scale bar: 100 μm). (d) Comparison of the normalized T2* signal intensity between the labeled (left) and unlabeled HepG2 cells (middle). The average T2* value of the four fields of labeled HepG2 cells was 6.67 ± 1.88 ms, which was significantly lower than that of the unlabeled cells (10.66 ± 2.22 ms) (P = 0.034) (right).

6

Academic Radiology, Vol ■, No ■■, ■■ 2016

MULTIMODALITY REPORTER GENE IMAGING FOR HCCS

Figure 3. (a) A frozen tumor section 14 days after labeled HepG2 cell transplant. Green fluorescence emission was detected in the cytoplasm of individual labeled HepG2 cells under blue laser excitation (left), nuclei counterstained with DAPI (middle), and the fused image of enhanced green fluorescent protein fluorescence and nuclear staining (right) (×400; scale bar: 50 μm). (b) The first, second, third, seventh, and fourteenth day after transplant, the bioluminescence signal was detected using an acquisition time of 5 seconds. Until the 14th day, the signal increased as the tumor size increased. (c) At 14 days after injection, the T2* value of the tumors originating from the labeled cells was 6.04 ± 1.60 ms, which was significantly lower than that from the unlabeled cells (17.06 ± 2.17 ms) (P < 0.001). This result indicated that the labeled HepG2 cells (ROI 1) transported and integrated more Fe3+ than the unlabeled HepG2 cells (ROI 2). ROI, region of interest.

14th day after transplant on average. Under blue laser excitation, the observations showed EGFP was distributed in the frozen sections (Fig 3a), and EGFP expression was predominantly observed in the cytoplasm of the tumor cells (Fig 3a). In contrast, no fluorescence signal was detected in the unlabeled HepG2 cells.

signal from the tumors on the first day after transplant. The bioluminescence increased with the tumor size, peaking on the 14th day (Fig 3b). Thereafter, as the tumor size increased, greater necrosis and festering emerged in the tumor, resulting in a decrease in the bioluminescence signal.

MRI for Ferritin In Vivo Bioluminescence Analysis of Luciferase2 in Vivo

Mice were intraperitoneally injected with D-luciferin at a dose of 150 mg/kg body weight. We detected bioluminescence

On the 14th day, the nude mice transplanted with tumors were scanned. The average T2* value of the tumors originating from labeled HepG2 cells was 6.04 ± 1.60 ms, which 7

QIN ET AL

was significantly lower than that from the unlabeled HepG2 cells (17.06 ± 2.17 ms) (P < 0.001) (Fig 3c). DISCUSSION In vivo visualization of transplanted HCC cells using noninvasive methods is essential in the study of the proliferation and transformation of HCC cells; such methods would facilitate the study of molecular functions targeted cells (17). Molecular imaging enables in vivo cell tracking in a realtime (18), longitudinal (19), noninvasive (20) manner, and potentially facilitates the assessment, optimization, and guidance of clinical targeted therapies (21). In the current study, we constructed a new plasmid, pLenti7.3-ferritin-IRESluciferase2-SV40-EGFP, which was used to insert the triple reporter gene sequence into the HepG2 genome to provide the basis for in vivo research at the histologic level (fluorescence), the animal level (bioluminescence), and the histologic anatomic level (MRI). The results of the transwell invasion assay verified that the transfection of the lentivirus carrying the triple reporter gene exerted negligible cytotoxicity and did not alter cell morphology or invasiveness. EGFP displays short excitation and emission wavelengths and low background signal. For adherent EGFP-positive cells and biopsy specimens, EGFP expression was detected in the cytoplasm of primary and daughter HepG2 cells under blue laser excitation (22,23). In transfected cells, the stable expression and high sensitivity of EGFP render it as a distinguishing marker. The expressed protein remained excitable in frozen tumor sections originating from labeled HepG2 cells; therefore, EGFP served as a qualitative indicator in subsequent experiments. However, we were unable to detect EGFP expression deep in the nude mice because of poor signal penetration and a low signal-to-noise ratio, and we failed to excite EGFP in tumors in vivo. However, the clearly positive immunostaining results demonstrated the potential of EGFP as a qualitative indicator to confirm whether the imaged cells are descendants of the transplanted cells. Bioluminescence is one of the most commonly used methods for in vivo imaging. In this method, a luciferase gene is inserted into the chromosomal DNA of the targeted cells to induce luciferase expression. In the presence of abundant ATP, oxygen, and exogenously administered luciferin, luciferase catalyzes luciferin to produce luminescence, which displays low background and a high signal-to-noise ratio. In our experiment, the correlation coefficient between the number of cells and the bioluminescence signal is 0.993, verifying the linear relationship between the optical SI and the number of cells. Thus, BLI may assist in quantitative cellular analysis (23,24). Furthermore, the results of in vivo BLI revealed bioluminescence signal from the tumor, and the SI increased with the tumor size. This result demonstrated that we could determine the number of tumor cells without sacrificing the mice. Optical imaging is a very sensitive technique, but it cannot provide accurate spatial positioning (9). However, MRI can compensate for this deficiency. 8

Academic Radiology, Vol ■, No ■■, ■■ 2016

MRI displays a high spatial resolution and provides anatomic and physiological information, which can be semiquantitative by obtaining the T2* value (25,26). The most common MRI methods involve exogenous negative MRI contrast agents, which are difficult to track over the course of long-term biological processes. This is primarily due to the deficiency of cell labeling, as the contrast agent cannot replicate itself during cell division, resulting in a decrease in the cellular signal. Therefore, MRI using contrast agents can only trace the distribution of the original transplanted cells in the short term (27). Molecular imaging using reporter genes could overcome this barrier to provide evidence regarding the distribution of target cells (28–30). The lentivirus integrated the ferritin reporter gene into the genome of the target cells, allowing the ferritin reporter gene to be replicated as the host cell proliferates (26). Ferritin specifically integrates to iron, and the expression level of ferritin negatively correlates with the MRI signal intensity (25,26). This finding is consistent with the conclusions of our experiment, in which the T2* signal of the labeled HepG2 cells (6.67 ± 1.88 ms) was lower than that of the unlabeled HepG2 cells at the same density (10.66 ± 2.22 ms). Moreover, the results of in vivo MRI showed that the T2* value of tumors originating from labeled HepG2 cells (6.04 ± 1.60 ms) was much lower than that of tumors originating from unlabeled HepG2 cells (17.06 ± 2.17 ms). This result illustrated that the ability of the labeled HepG2 cells to integrate iron is greater than that of the unlabeled HepG2 cells and confirmed the expression of ferritin in the labeled HepG2 cells. This study has some limitations. We failed to detect the expression of EGFP in vivo, as we only detected EGFP fluorescence in frozen sections; this result limited the sensitivity of EGFP for cell imaging in vivo. Even when ferritin was successfully expressed in labeled HepG2 cells, the decrease in the T2* value was distinguished using the 9.4T MR scanner, which would be a limitation to its application. We only established a nude mouse subcutaneous hepatoma model and did not transplant the labeled cells into the liver. Moreover, we once thought the tumor under the skin would give us an intuitive evidence of the tumor growth. However, on the in vivo MR images, we found the tumor could infiltrate into the deep surrounding tissue, and this part of tumor was also a large part which could not be observed from the body surface. So a surveillance of the tumor growth using in vivo BLI and MRI might lead to a more persuasive result. But in this study, we did not obtain these data. This is our lapses in planning. However, our results demonstrated that the bioluminescence reporter gene and the MR reporter gene were efficiently delivered to HCC cells using the lentiviral vector, enabling the detection of these cells via fluorescence, bioluminescence, and MRI in vitro and in vivo. In conclusion, we successfully constructed a multimodality reporter gene for the sensitive detection of tumor cells via fluorescence, bioluminescence, and MRI. Both in vitro and in vivo studies demonstrated that this viral vector efficiently delivered the bioluminescence reporter gene and the MR reporter gene to HCC cells, which resulted in evident gene

Academic Radiology, Vol ■, No ■■, ■■ 2016

expression in labeled tumor cells and significantly different MR signal between the labeled and unlabeled cells. In this study, EGFP served as a qualitative indicator for target identification, luciferase2 served as a quantitative marker in vivo, and MRI revealed the location of the groups of labeled cells in vivo. Our results revealed that the multimodality reporter may serve as an effective platform for the targeted bioluminescence and MRI of HCC cells; thus, this tool warrants further investigation. ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grant Nos. 81070349, 81371655, and 81571774) and the Natural Science Foundation of Guangdong Province, China (Grant No. 2014A030313171). REFERENCES 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2015; 65:5–29. 2. Tejeda-Maldonado J, García-Juárez I, Aguirre-Valadez J, et al. Diagnosis and treatment of hepatocellular carcinoma: an update. World J Hepatol 2015; 7:362–376. 3. Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012; 379:1245–1255. 4. Youn H, Chung JK. Reporter gene imaging. AJR Am J Roentgenol 2013; 201:206–214. 5. Brader P, Serganova I, Blasberg RG. Noninvasive molecular imaging using reporter genes. J Nucl Med 2013; 54:167–172. 6. Lin Y, Molter J, Lee Z, et al. Bioluminescence imaging of hematopoietic stem cell repopulation in murine models. Methods Mol Biol 2008; 430:295–306. 7. Hasegawa S, Furukawa T, Saga T. Molecular MR imaging of cancer gene therapy: ferritin transgene reporter takes the stage. Magn Reson Med Sci 2010; 9:37–47. 8. Shen L, Li Y, Chen J, et al. Generation of a recombinant classical swine fever virus stably expressing the firefly luciferase gene for quantitative antiviral assay. Antiviral Res 2014; 109:15–21. 9. Wu JC, Sundaresan G, Iyer M, et al. Noninvasive optical imaging of firefly luciferase reporter gene expression in skeletal muscles of living mice. Mol Ther 2001; 4:297–306. 10. Amsalem Y, Mardor Y, Feinberg MS, et al. Iron-oxide labeling and outcome of transplanted mesenchymal stem cells in the infarcted myocardium. Circulation 2007; 116:38–45. 11. Sun JH, Teng GJ, Ju SH, et al. MR tracking of magnetically labeled mesenchymal stem cells in rat kidneys with acute renal failure. Cell Transplant 2008; 17:279–290.

MULTIMODALITY REPORTER GENE IMAGING FOR HCCS

12. Choi SH, Cho H, Kim HS, et al. Imaging and quantification of metastatic melanoma cells in lymph nodes with a ferritin MR reporter in living mice. NMR Biomed 2012; 25:737–745. 13. Wu C, Li J, Pang P, et al. Polymeric vector-mediated gene transfection of MSCs for dual bioluminescent and MRI tracking in vivo. Biomaterials 2014; 35:8249–8260. 14. Li Z, Hu X, Mao J, et al. Optimization of mesenchymal stem cells (MSCs) delivery dose and route in mice with acute liver injury by bioluminescence imaging. Mol Imaging Biol 2015; 17:185–194. 15. Sani IK, Marashi SH, Kalalinia F. Solamargine inhibits migration and invasion of human hepatocellular carcinoma cells through down-regulation of matrix metalloproteinases 2 and 9 expression and activity. Toxicol In Vitro 2015; 29:893–900. 16. Kim HS, Joo HJ, Woo JS, et al. In vivo magnetic resonance imaging of transgenic mice expressing human ferritin. Mol Imaging Biol 2013; 15:48– 57. 17. Iordanova B1, Goins WF, Clawson DS, et al. Quantification of HSV-1mediated expression of the ferritin MRI reporter in the mouse brain. Gene Ther 2013; 20:589–596. 18. Sheth RA, Mahmood U. Optical molecular imaging and its emerging role in colorectal cancer. Am J Physiol Gastrointest Liver Physiol 2010; 299:G807–G820. 19. Cho IK, Moran SP, Paudyal R, et al. Longitudinal monitoring of stem cell grafts in vivo using magnetic resonance imaging with inducible maga as a genetic reporter. Theranostics 2014; 4:972–989. 20. Youn H, Hong KJ. In vivo noninvasive small animal molecular imaging. Osong Public Health Res Perspect 2012; 3:48–59. 21. Alberti C. From molecular imaging in preclinical/clinical oncology to theranostic applications in targeted tumor therapy. Eur Rev Med Pharmacol Sci 2012; 16:1925–1933. 22. Pei Z, Lan X, Cheng Z, et al. A multimodality reporter gene for monitoring transplanted stem cells. Nucl Med Biol 2012; 39:813– 820. 23. Wu Y, Liao Q, Yang R, et al. A novel luciferase and GFP dual reporter virus for rapid and convenient evaluation of hepatitis C virus replication. Virus Res 2011; 155:406–414. 24. Shimizu T, Ohta Y, Tsutsuki H, et al. Construction of a novel bioluminescent reporter system for investigating Shiga toxin expression of enterohemorrhagic Escherichia coli. Gene 2011; 478:1–10. 25. Vande Velde G, Raman Rangarajan J, Vreys R, et al. Quantitative evaluation of MRI-based tracking of ferritin-labeled endogenous neural stem cell progeny in rodent brain. Neuroimage 2012; 62:367– 380. 26. Iordanova B, Ahrens ET. In vivo magnetic resonance imaging of ferritinbased reporter visualizes native neuroblast migration. Neuroimage 2012; 59:1004–1012. 27. Qi Y, Feng G, Huang Z, et al. The application of super paramagnetic iron oxide-labeled mesenchymal stem cells in cell-based therapy. Mol Biol Rep 2013; 40:2733–2740. 28. Sun N, Lee A, Wu JC. Long term non-invasive imaging of embryonic stem cells using reporter genes. Nat Protoc 2009; 4:1192–1201. 29. Kang JH, Chung JK. Molecular-genetic imaging based on reporter gene expression. J Nucl Med 2008; 49(suppl 2):164S–179S. 30. Gilad AA, Winnard PT, Jr, van Zijl PC, et al. Developing MR reporter genes: promises and pitfalls. NMR Biomed 2007; 20:275–290.

9