Evaluation of bioluminescent imaging for noninvasive monitoring of colorectal cancer progression in the liver and its response to immunogene therapy
- 10k Downloads
Bioluminescent imaging (BLI) is based on the detection of light emitted by living cells expressing a luciferase gene. Stable transfection of luciferase in cancer cells and their inoculation into permissive animals allows the noninvasive monitorization of tumor progression inside internal organs. We have applied this technology for the development of a murine model of colorectal cancer involving the liver, with the aim of improving the pre-clinical evaluation of new anticancer therapies.
A murine colon cancer cell line stably transfected with the luciferase gene (MC38Luc1) retains tumorigenicity in immunocompetent C57BL/6 animals. Intrahepatic inoculation of MC38Luc1 causes progressive liver infiltration that can be monitored by BLI. Compared with ultrasonography (US), BLI is more sensitive, but accurate estimation of tumor mass is impaired in advanced stages. We applied BLI to evaluate the efficacy of an immunogene therapy approach based on the liver-specific expression of the proinflammatory cytokine interleukin-12 (IL-12). Individualized quantification of light emission was able to determine the extent and duration of antitumor responses and to predict long-term disease-free survival.
We show that BLI is a rapid, convenient and safe technique for the individual monitorization of tumor progression in the liver. Evaluation of experimental treatments with complex mechanisms of action such as immunotherapy is possible using this technology.
KeywordsLuciferase Activity Light Emission Mifepristone Luciferase Expression MC38 Cell
The liver is the most frequent site for metastases from colorectal cancer. Approximately 10–25% of colon cancer patients present one or multiple liver metastases at the time of diagnose . At least in 30% of these cases the liver is the only organ affected, apart from the tumor in the gastrointestinal tract. Moreover, recurrence after surgical removal of the primary lesion occurs mainly in the liver, with a 20–25% rate of metachronous liver metastases. Potentially curative resection of hepatic tumors is not feasible in more than 75% of the cases due to large size, elevated number and/or unfavourable localization of lesions, or poor liver function. Nonsurgical approaches including systemic chemotherapy and regional treatments are the only options for these patients. Local control is often achieved and these techniques are rapidly improving [2, 3], but a significant increase in long-term survival is not guaranteed. Therefore, hepatic metastases from colon cancer are frequently observed in the clinic and they are the most frequent cause of death in these patients. Advances in the management of this disease will probably require the combination of standard care and new therapies that are still in the experimental stage.
Immunotherapy is one of these alternatives . Systemic or local administration of vectors driving expression of immunostimulatory cytokines such as interleukin-12 (IL-12) has demonstrated potent antitumor effects in pre-clinical studies [5, 6, 7, 8]. However, further optimization of this approach is required, and improvement in animal models is needed so that research in this area can generate more clinically relevant results [9, 10]. In a previous study , we described a High-Capacity (gutless) adenoviral vector carrying a liver-specific inducible system for the expression of murine IL-12 (GL-Ad/RUmIL-12). Intravenous administration of this vector eliminated intrahepatic colon cancer in a murine model when intense production of IL-12 was induced at early stages of the disease. If more restrictive conditions are used (larger tumors and lower dose of vector that leads to moderate IL-12 concentration) the antitumor response was heterogeneous (manuscript in preparation), as observed with many other experimental approaches .
In these cases, a more detailed characterization of the partial responses would be desirable, and longitudinal monitoring of individual subjects could identify transient antitumor effects. Implantation of certain colon cancer cell lines in the liver of syngeneic mice constitutes one kind of intrahepatic cancer model . Although each model has its own limitations, progressive growth and extra hepatic dissemination of these tumors often leads to the death of the animal, recapitulating some aspects of the natural history found in humans. However, monitoring progression in these internal tumors by direct measurement requires repeated laparotomy or large groups of animals to be sacrificed at different time points, thus precluding an individualized follow-up. Different noninvasive imaging techniques have been developed to overcome these limitations. Some of them such as ultrasonography (US) , computerized tomography (CT) , positron emission tomography (PET) , single photon emission computed tomography (SPECT)  and magnetic resonance imaging (MRI) [18, 19] are adaptations of clinical imaging devices to the use in small animals. Others such as fluorescence imaging (FLI)  and bioluminescent imaging (BLI) [21, 22] have been specifically developed for the in vivo monitoring of gene expression in experimental animals, mostly rodents.
Bioluminescence of cells is based on a chemical reaction catalyzed by the luciferase enzyme in which a substrate (D-luciferin) is converted into an excited oxyluciferin intermediate in the presence of Oxygen, Magnesium and ATP . When oxyluciferin returns to its relaxed state, it emits a photon in the visible wavelength range. The most common source for luciferase is the firefly Photinus pyralis. Since no luciferase expression is found in mammalian cells and there is no need for external light excitation, this method of cell labelling has a very low background. The intensity of light is proportional to the amount of luciferase expressed in each individual cell, and the number of cells in which the gene has been transferred. In addition, actual light detectors have a very high dynamic range. Therefore, luciferase-based assays have been widely used to study changes in gene expression intensity in vitro and in vivo. For tumor imaging, a luciferase expressing gene is transferred to cancer cells and stable clones are implanted into permissive hosts. These can be syngeneic animals [24, 25, 26, 27], but in the case of human cancer cell lines, immunodeficient mice or rats are required [28, 29, 30, 31, 32]. A cooled charge-coupled device (CCD) camera attached to a light-tight chamber is then used to detect the intensity and location of light emission. The high sensitivity of BLI has been extensively demonstrated in different tumor models, including colon cancer [30, 33]. However, stability of luciferase expression and maintenance of a good correlation between light emission and tumor burden needs further investigation, especially if orthotopic (intrahepatic) immunocompetent models are used [26, 27]. The influence of anatomical barriers and the fact that luciferase is a foreign gene could interfere with this correlation. Therefore, in this study we have stably transfected MC38 murine colon cancer cells with a plasmid carrying luciferase gene and have monitored the evolution of bioluminescence and tumor volume after subcutaneous or intrahepatic inoculation of cells in C57BL/6 mice. We found a direct correlation between these two parameters in both cases, although higher dispersion was observed at late stages of the experiment. Long-term maintenance of the genetically modified cells and stability of luciferase expression were observed.
In addition, we provide evidence showing that this model is suitable for evaluation of individual responses to immunotherapy. Using BLI monitoring, we were able to identify subsets of animals that experienced transient reduction in tumor progression, and to clearly distinguish this group from others having no effect, or long-term complete responses.
Results and discussion
Characterization of MC38 colon cancer cells stably transfected with the luciferase gene
In order to verify that the MC38Luc1 cells retained the tumorigenic properties of the parental cells, we inoculated 106 MC38 or MC38Luc1 cells subcutaneously in two groups of C57BL/6 mice. We could not detect any difference in tumor progression between both groups. In figure 1D, we represent the average tumor volume 3 weeks after inoculation of cells, determined by direct calliper measurement. Therefore, integration of the luciferase expression cassette in the genome of the MC38Luc1 clone did not change their ability to form tumors in vivo. In a previous report by an independent group , MC38 cells transduced with retroviruses encoding luciferase or GFP showed moderate differences in tumor progression, although both cell lines gave rise to tumors that grew progressively for at least 20 days. The size of tumors observed in our experiments (367 mm3 after 21 days) coincides remarkably with the clone described by Choy et al . This and other studies in syngeneic animals clearly demonstrate that tumor cells expressing luciferase are not selectively eliminated by the immune system.
Comparison of US and BLI for noninvasive monitorization of liver metastases
Limitations of BLI in the advanced stage of tumors
Hypoxia has been recently proposed as a potential inhibitor of the luciferase reaction . Experiments performed in vitro demonstrated that intense hypoxia (0,2% pO2) can reduce ATP levels in cells and impair bioluminescence. Differences in the extension and intensity of hypoxic areas could contribute to the lack of correlation between light emission and tumor volume. To investigate this possibility, we obtained lysates from advanced tumors and performed in vitro luciferase reactions. Under these circumstances, luciferase activity (measured in Relative Luciferase Units, RLUs in a standard luminometer) only depends on the amount of functional enzyme, because all other components of the reaction, including ATP, are contained in the assay buffer. Therefore, if lack of ATP is limiting light emission in some of the MC38Luc1 tumors in vivo, luciferase activity should be restored in vitro. However, we did not observe any sample in which low photon emission in vivo was accompanied by an unexpectedly high RLU value in vitro (figure 5C). This suggests that hypoxia is not interfering with BLI in the MC38Luc1 model. However, the correlation between both parameters was modest (r2 = 0.42) due to considerable dispersion of values. Therefore, we cannot definitely rule out the possibility that availability of ATP in vivo is affecting the accuracy of BLI.
Validity of BLI to monitor the efficacy of immunotherapy
We have validated an immunocompetent model of hepatic tumors that allows the noninvasive monitorization of tumor progression using BLI. A good correlation between functional imaging and tumor volume was observed until mice reached an advanced stage of the disease. This model is suitable to characterize transient and long-term antitumor responses obtained with new therapeutic approaches such as immunotherapy.
The MC38 murine colon carcinoma cell line was derived from 3-methylcholanthrene-treated C57BL/6 mice . It was grown in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 μg/ml streptomycin and 100 μg/ml penicillin (all from of Gibco, Invitrogen) and 2 mM glutamine (Cambrex). The MC38Luc1 clone was maintained in the same media supplemented with 400 μg/ml G418 (geneticin, Gibco, Invitrogen).
DNA cloning and stable transfection
pCDNA3-luc was generated by inserting the luciferase gene from the plasmid pGL3-Basic (Promega) into the Hind III-Xba I sites of pCDNA3.1 vector (Invitrogen). MC38 cells were grown in 10 cm dishes and were transfected with 4 μg of pCDNA3-luc by the calcium phosphate precipitation method when they reached 50–60% confluence. Post-transfection culture medium was replaced 16 hours later by DMEM 10% FBS supplemented with 400 μg/ml G418. Resistant clones were isolated by individual trypsinization and expanded.
Different clones transfected with the pCDNA3-luc plasmid were plated in 96 well-black plates (Packard) at a density of 100; 500; 1,000; 5,000; 12,500; 25,000; 50,000; 75,000 or 100,000 cells per well for MC38Luc1 cells, or at cell densities above 12,500 cell/well for clones MC38Luc4 and MC38Luc8. Twenty-four hours later, cells were washed in PBS and D-luciferin substrate (Promega) was added at a final concentration of 150 μg/ml. Light emission from culture plates was detected immediately using the IVIS CCD camera system (Xenogen) and analyzed with the Living Image 2.20 software package (Xenogen). For in vivo imaging and quantification of light emission, mice were anesthetized with a mixture of Xylacine and Ketamine and 150 mg/kg D-luciferin (100 μl of a 30.3 mg/ml solution dissolved in phosphate-buffered saline) were injected intraperitoneally. Ten minutes later, animals were placed in the dark chamber for light acquisition. Typically, a circular region of interest measuring 3 cm in diameter was defined in the abdomen of mice, and quantification of light emission was performed in photons/second. Time exposure ranged from 1 second to 5 minutes depending on light intensity. A maximum of 10 animals were analyzed at a time.
Determination of luciferase activity in solution
Fifty to 150 mg of frozen tumor samples were lysed in Reporter Lysis Buffer (Promega) and centrifuged at 13.000 rpm in a cooled microfuge for 10 minutes. The luciferase activity was measured in the supernatant using the Luciferase Reporting Assay System (Promega) in a Berthold microplate luminometer. The luciferase Relative Lights Units (RLUs) obtained for each sample were multiplied by the total tumor weight to obtain an estimation of the RLU/tumor.
RNA and genomic DNA isolation and qRT-PCRs
Genomic DNA and total RNA was extracted from 50–100 mg frozen tumor samples using QIAmp DNA mini kit only (Qiagen) or TRI reagent (Sigma), respectively, following manufacturer's instructions. Three μg of RNA were treated with DNase I and retro transcribed to cDNA with M-MLV RT in the presence of RNase out reagent (all from Invitrogen). For real time PCR reactions, 2 μl of genomic DNA or cDNA were mixed with specific primers using iQ SYBR Green Supermix (Bio-Rad). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used to normalize gene expression and murine albumin was used to normalize genomic DNA content. Gene expression (specific mRNA content) is represented by the formula 2ΔCt where ΔCt indicates the difference in the threshold cycle between GAPDH and luciferase. Plasmid copy number refers to the copies of the neo resistance gene relative to the copies of the endogenous albumin gene. Standard curves were used to calculate copy numbers in each case. The name and sequence of primers used is as follows: GAPDH forward CCAAGGTCATCCATGACAAC; reverse TGTCATACCAGGAAATGAGC. Albumin forward GATGCTGCTCTTTGGCTATGA; reverse CAGCAGTCAGCCAGTTCACC Neo forward AGATGGATTGCACGCAGGT; reverse TTGCATCAGCCATGATGGA. Firefly Luciferase forward AGAGATACGCCCTGGTTC; reverse ATAAATAACGCGCCCAACAC.
MC38 and MC38Luc1 cells were cultured in glass coverslides and fixed in pre-chilled acetone for 4 min. Samples were kept at -20°C before being analyzed. For luciferase protein detection, coverslides were air-dried and washed tree times in phosphate-buffered saline (PBS). Endogenous peroxidase was quenched with Peroxidase Blocking Reagent (DAKO) for 15 min at room temperature. Samples were then washed and incubated with 5 μg/ml of the primary goat anti-luciferase antibody (Cortex) for 1 h at room temperature. After additional washing, samples were incubated with the polyclonal biotinylated anti-goat antibody diluted 1:600 in PBS for 45 min and then were incubated for 45 min with HRP-streptavidin (Amersham). The peroxidase activity was revealed using DAB Substrate Chromogen System (DAKO).
Mice and tumor cell inoculation
Five to 8 weeks old C57BL/6J female mice were purchased from Harlan (Barcelona, Spain) and were kept in the animal facility at least one week before starting the experiments. Tumors were established by subcutaneous (s.c.) or intrahepatic (i.h.) implantation of cells. For s.c. tumor formation, a total of 106 cells were injected in the right hind flank. For liver metastases establishment, 5 × 105 cells were injected into the left liver lobe of mice following medial laparotomy in isofluorane-anesthesized animals. In both cases, cells were resuspended in a total volume of 50 μl saline solution. Tumor size was monitored at indicated time points measuring two perpendicular tumor diameters using a precision calliper. Tumor volume was calculated using the following formula: V = length × width2 × 0.5. In the case of intrahepatic tumors, laparotomies were performed in a subset of the animals at weekly intervals in order to estimate the average tumor volume of the group by direct calliper measurement. Individual animals underwent a maximum of two exploratory laparotomies before they were sacrificed. Survival was checked daily and mice were euthanized if general status was deteriorated. All in vivo experiments were performed in accordance with the local animal commission.
Mice were anesthetized with 1.5% isofluorane in oxygen at 0.6 L/min delivered via nose cone and allowed to breathe spontaneously. They were placed in supine position on a feedback-controlled heating pad. Abdomen was shaved, and pre-warmed ultrasound gel was applied on the skin of mice. Recordings were made under continuous ECG monitoring by fixing the electrodes on the limbs. Two-dimension mode U.S imaging was performed by using a dedicated small-animal high resolution imaging unit (VEVO 770; Visualsonics, Toronto, Canada) and a Visualsonics RMV 700-series scanhead with a 30–40 MHz high frequency linear transducer (depth of field 1.5–2.2 mm, field of view-max 14.5–16.5 mm). Tumor diameter measurements were made from digital images captured on cineloops at the time of the study using the software incorporated in the VEVO 770 device. Tumor volumes were calculated as described in the previous section.
Treatment of colorectal liver metastases by adenoviral-directed expression of IL-12
The High-Capacity (gutless) adenoviral vector GL-Ad/RUmIL-12 carrying a liver-specific Mifepristone-inducible system for the expression of murine IL-12 has been previously described . MC38Luc1 cells were inoculated in the liver of C57BL/6 mice, and 3 days later 2.5 × 108 iu of the virus were injected intravenously dissolved in 200 μl saline solution. One week later, expression of IL-12 was activated by daily intraperitoneal injection of 250 μg/kg Mifepristone for 10 days. Quantification of IL-12 concentration in the serum of animals was performed by ELISA 10 hours after the first induction. Tumor progression was monitored by BLI or direct tumor measurement following laparotomy or necropsy.
We used GraphPad Prism software for statistical analysis. Two-tailed unpaired t-test was used to compare groups of values when n>10. For smaller groups, Mann-Whitney non-parametric test was used.
We thank Dr. Lieping Chen (Department of Oncology and Dermatology, Johns Hopkins University School of Medicine, Baltimore, Maryland) for the MC38 cells, and Maria Bunuales for technical assistance. This work was funded in part by FMMA, Grant SAF2006-04755 from the Spanish Department of Education and Science, and the UTE project CIMA.
- 5.Narvaiza I, Mazzolini G, Barajas M, Duarte M, Zaratiegui M, Qian C, Melero I, Prieto J: Intratumoral coinjection of two adenoviruses, one encoding the chemokine IFN-gamma-inducible protein-10 and another encoding IL-12, results in marked antitumoral synergy. J Immunol. 2000, 164: 3112-3122.CrossRefPubMedGoogle Scholar
- 7.Melero I, Duarte M, Ruiz J, Sangro B, Galofre J, Mazzolini G, Bustos M, Qian C, Prieto J: Intratumoral injection of bone-marrow derived dendritic cells engineered to produce interleukin-12 induces complete regression of established murine transplantable colon adenocarcinomas. Gene Ther. 1999, 6: 1779-1784. 10.1038/sj.gt.3301010CrossRefPubMedGoogle Scholar
- 8.Mazzolini G, Narvaiza I, Bustos M, Duarte M, Tirapu I, Bilbao R, Qian C, Prieto J, Melero I: Alpha(v)beta(3) integrin-mediated adenoviral transfer of interleukin-12 at the periphery of hepatic colon cancer metastases induces VCAM-1 expression and T-cell recruitment. Mol Ther. 2001, 3: 665-672. 10.1006/mthe.2001.0317CrossRefPubMedGoogle Scholar
- 9.Mazzolini G, Alfaro C, Sangro B, Feijoó E, Ruiz J, Benito A, Tirapu I, Arina A, Sola J, Herraiz M, Lucena F, Olagüe C, Subtil J, Quiroga J, Herrero I, Sádaba B, Bendandi M, Qian C, Prieto J, Melero I: Intratumoral injection of dendritic cells engineered to secrete interleukin-12 by recombinant adenovirus in patients with metastatic gastrointestinal carcinomas. J Clin Oncol. 2005, 23: 999-1010. 10.1200/JCO.2005.00.463CrossRefPubMedGoogle Scholar
- 10.Sangro B, Mazzolini G, Ruiz J, Herraiz M, Quiroga J, Herrero I, Benito A, Larrache J, Pueyo J, Subtil JC, Olagüe C, Sola J, Sádaba B, Lacasa C, Melero I, Qian C, Prieto J: Phase I trial of intratumoral injection of an adenovirus encoding interleukin-12 for advanced digestive tumors. J Clin Oncol. 2004, 22: 1389-1397. 10.1200/JCO.2004.04.059CrossRefPubMedGoogle Scholar
- 11.Wang L, Hernandez-Alcoceba R, Shankar V, Zabala M, Kochanek S, Sangro B, Kramer MG, Prieto J, Qian C: Prolonged and inducible transgene expression in the liver using gutless adenovirus: a potential therapy for liver cancer. Gastroenterology. 2004, 126: 278-289. 10.1053/j.gastro.2003.10.075CrossRefPubMedGoogle Scholar
- 18.Judenhofer MS, Wehrl HF, Newport DF, Catana C, Siegel SB, Becker M, Thielscher A, Kneilling M, Lichy MP, Eichner M, Klingel K, Reischl G, Widmaier S, Röcken M, Nutt RE, Machulla HJ, Uludag K, Cherry SR, Claussen CD, Pichler BJ: Simultaneous PET-MRI: a new approach for functional and morphological imaging. Nat Med. 2008, 14: 459-465. 10.1038/nm1700CrossRefPubMedGoogle Scholar
- 19.Marzola P, Degrassi A, Calderan L, Farace P, Nicolato E, Crescimanno C, Sandri M, Giusti A, Pesenti E, Terron A, Sbarbati A, Osculati F: Early antiangiogenic activity of SU11248 evaluated in vivo by dynamic contrast-enhanced magnetic resonance imaging in an experimental model of colon carcinoma. Clin Cancer Res. 2005, 11: 5827-5832. 10.1158/1078-0432.CCR-04-2655CrossRefPubMedGoogle Scholar
- 25.Miretti S, Roato I, Taulli R, Ponzetto C, Cilli M, Olivero M, Di Renzo MF, Godio L, Albini A, Buracco P, Ferracini R: A mouse model of pulmonary metastasis from spontaneous osteosarcoma monitored in vivo by Luciferase imaging. PLoS ONE. 2008, 3: e1828- 10.1371/journal.pone.0001828PubMedCentralCrossRefPubMedGoogle Scholar
- 31.Rozemuller H, Spek van der E, Bogers-Boer LH, Zwart MC, Verweij V, Emmelot M, Groen RW, Spaapen R, Bloem AC, Lokhorst HM, Mutis T, Martens AC: A bioluminescence imaging based in vivo model for preclinical testing of novel cellular immunotherapy strategies to improve the graft-versus-myeloma effect. Haematologica. 2008, 93: 1049-1057. 10.3324/haematol.12349CrossRefPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.