This protocol describes xenograft and orthotopic mouse models of human thyroid tumorigenesis as a platform to test microRNA-based inhibitor treatments. This approach is ideal to study the function of non-coding RNAs and their potential as new therapeutic targets.
MicroRNAs (miRNAs) are important regulators of gene expression through their ability to destabilize mRNA and inhibit translation of target mRNAs. An ever-increasing number of studies have identified miRNAs as potential biomarkers for cancer diagnosis and prognosis, and also as therapeutic targets, adding an extra dimension to cancer evaluation and treatment. In the context of thyroid cancer, tumorigenesis results not only from mutations in important genes, but also from the overexpression of many miRNAs. Accordingly, the role of miRNAs in the control of thyroid gene expression is evolving as an important mechanism in cancer. Herein, we present a protocol to examine the effects of miRNA-inhibitor delivery as a therapeutic modality in thyroid cancer using human tumor xenograft and orthotopic mouse models. After engineering stable thyroid tumoral cells expressing GFP and luciferase, cells are injected into nude mice to develop tumors, which can be followed by bioluminescence. The in vivo inhibition of a miRNA can reduce tumor growth and upregulate miRNA gene targets. This method can be used to assess the importance of a determined miRNA in vivo, in addition to identifying new therapeutic targets.
Thyroid cancer is an endocrine malignancy with an increasing incidence, although in general terms it has a good outcome1. Nevertheless, some patients develop aggressive forms of the disease that are untreatable and the molecular bases are poorly understood2.
miRNAs are 22-nucleotide-long non-coding RNAs that regulate gene expression in many tissues, typically by base-pair binding to the 3' untranslational region (3’UTR) of target messenger RNAs (mRNAs), triggering mRNA degradation or translational repression3,4. There is increasing evidence demonstrating that the deregulation of microRNA expression is a hallmark of cancer, as these molecules modulate proliferative signaling, migration, invasion and metastasis, and can provide resistance to apoptosis5,6. In recent years, many studies have identified miRNAs as potential biomarkers for cancer diagnosis and prognosis as well as therapeutic targets7, providing a new dimension to cancer evaluation and treatment.
miRNAs have taken center stage in human molecular oncology as key drivers of human thyroid neoplasms8,9,10,11,12. Among the miRNAs up-regulated, miR-146b is highly overexpressed in Papillary Thyroid Carcinoma (PTC) tumors and was shown to significantly increase cell proliferation, and to be associated with aggressiveness and dismal prognosis6,12,13,14,15. Furthermore, miR-146b regulates several thyroid genes involved in differentiation12, and also important tumor suppressor genes such as PTEN16 and DICER117. Despite their importance in cancer biology, miRNA-based cancer therapy is still in its early stages, and very few studies have addressed thyroid cancer – the most frequent of the endocrine tumors18. Here we describe a protocol using two different mouse models with human-derived tumors, in which the administration of a synthetic miRNA-inhibitor (antagomiR) that specifically inhibits a cellular miRNA can block tumor growth. We first used a common xenograft model, and the local intratumor administration of an antagomiR decreased tumor growth measured as a reduction in tumor bioluminescence16. Because the establishment of robust mouse models mimicking human tumor progression is essential to develop unique therapeutic approaches, orthotopic implantation of primary human tumors is a more valuable platform for clinical validation of new drugs than subcutaneous implantation models. Thus, in order to better assess the therapeutic potential of the antagomiR, we used an orthotopic mouse model with systemic delivery in the blood stream, obtaining the same results.
Animal experimentation was performed in compliance with the European Community Law (86/609/EEC) and the Spanish law (R.D. 1201/2005), with approval of the Ethics Committee of the Consejo Superior de Investigaciones Científicas (CSIC, Spain).
1. Flank inoculation of cells and intratumoral antagomiR treatment
2. Thyroid orthotopic inoculation of cells and systemic antagomiR treatment
We used two different mice models to determine whether the neutralization of a miRNA could suppress tumor growth. Accordingly, human tumor thyroid Cal62-luc cells were subcutaneously injected into the flanks of nude mice to generate a xenograph model. After two weeks, tumors were established and could be measured with calipers. At that time point, mice were injected intratumorally with the miR-146b-inhibitor, or an appropriate control, and tumor volume was followed for a further two weeks (Figure 1A). As shown in Figure 1B, the growth of tumors intratumorally injected with the miR-146b-inhibitor (anti-146b) (n=8) was significantly suppressed with respect to the negative control group (n=4) or the saline control group (not shown). miRNAs are an important feature of gene regulation, as they regulate several target mRNAs. Accordingly, oncomiRs are important regulators of tumor suppressor genes. This protocol allows for the analysis of intratumoral expression levels of these genes, which can be tested after treatment with the miRNA-inhibitor. In addition, the levels of some proliferation markers can be also studied, illustrated by the lower expression of proliferating cell nuclear antigen (PCNA) in antagomiR-treated tumors than in control-treated tumors (Figure 2). Also the recovery of the miRNA-targets can be studied through the analysis of tumor-extracted RNA or protein, as illustrated by the higher expression of PTEN in the antagomiR(anti-146b) treated tumors (Figure 2). Collectively, these data reveal that the in vivo inhibition of a miRNA is effective and may be exploited therapeutically for thyroid cancer treatment.
To better assess the therapeutic potential of a miRNA-inhibitor and improve the mouse model, we generated an orthotopic model, in which human tumor thyroid Cal62-luc cells were directly seeded into the right thyroid lobe of nude mice. After 3 weeks, the thyroid tumors were established, as demonstrated by bioluminescence signals in the neck of the mice and by gross tissue analysis and immunohistochemistry, with hematoxylin and eosin and GFP staining of cells, respectiviely. As shown in Figure 3, the epithelial cells surrounding the colloid demonstrate the thyroid follicle architecture from the mouse. GFP-positive cells interspersed with the follicles, unequivocally demonstrating that human Cal62 cells have been injected correctly and proliferate within the mouse thyroid. Once the tumors were formed, we injected 13 mice intravenously with the miR-146b-inhibitor (n=8) or an appropriate control (n=5), and tumor volume was followed for a further two weeks (Figure 4A). We found that the tumor growth was significantly decreased in the systemic antagomiR-treated group (Figure 4B). Notably, the expression of the newly described miR-146b-target DICER1 in the primary tumor was increased after the anti-miR-146b treatment (Figure 5). These data demonstrate that the inhibition of endogenous miRNA expression and therefore the restoration of its target genes could be used as a therapy in thyroid cancer16,17.
Figure 1. The miR-146b-inhibitor impairs established human thyroid tumor growth. Cal62-luciferase expressing cells (Cal62-luc) were injected subcutaneously. After xenografts were established, a synthetic miR-inhibitor (anti-146b) (n=8) or a negative control (n=4) was administered intratumorally. (A) Timeline. (B) Left: The image shows the endpoint bioluminescent signal of the treated tumors imaged with in vivo imaging software. Right: The graph shows tumor radiance increase at the indicated times in xenografts from treatment onset with the miRNA-inhibitor (dark grey) or the negative control (grey). (C) Representation of the endpoint radiance increase of the treated tumors. Values represent mean ± SEM. *p < 0.05; **p <0.01. Figure adapted from Ramírez-Moya et al.16. Please click here to view a larger version of this figure.
Figure 2. PCNA protein expression decreases and PTEN protein expression increases in miR-146b-inhibitor-treated tumors. Protein from the tumors treated with control or miR-146b inhibitor was obtained for western blotting with antibodies to PCNA or PTEN. Figure adapted from Ramírez-Moya et al.16. Please click here to view a larger version of this figure.
Figure 3. Intratumoral mouse thyroid follicles in the orthotopic model. Staining with hematoxylin and eosin (upper panel) and or an antibody to GFP (bottom panel) of the orthotopic mice tumors 28 days after inoculation with Cal62-luc GFP-positive cells. Please click here to view a larger version of this figure.
Figure 4. The miR-146b-inhibitor impairs established human thyroid tumor growth. Cal62-luciferase expressing cells (Cal62-luc) were injected in the thyroid gland of the mice. After the tumors were established (3 weeks), a synthetic miRNA-inhibitor (anti-146b) (n=8) or a negative control (n=5) was administered systemically. (A) Timeline. (B) Left: The image shows the endpoint bioluminescent signal of the treated tumors imaged with in vivo imaging software. Right: The graph shows tumor radiance increase at the indicated times from treatment onset with the miRNA-inhibitor (blue) or the negative control (green). (C) Representation of the endpoint radiance increase of the treated tumors. Values represent mean ± SEM. *p < 0.05. Figure adapted from Ramírez-Moya et al.17. Please click here to view a larger version of this figure.
Figure 5. DICER1 protein expression increases in miR-146b-inhibitor-treated tumors. Immunohistochemistry with a DICER1 antibody in the orthotopic tumors after treatment with miR-146b-inhbitor or control. Please click here to view a larger version of this figure.
This paper describes a method for studying the in vivo function of a miRNA in order to better understand its role in tumor initiation and progression, and its potential as a therapeutic target in thyroid cancer. The tumor xenograft models here described are based on the use of cells that can be tracked by their bioluminescence signal, permitting the measurement of tumor growth in vivo under the influence of a treatment. In addition, we describe the use of a miRNA-based treatment for thyroid cancer, which is currently in the early stages.
We first tested the feasibility of a miRNA as a therapeutic target using a subcutaneous xenograft model, in which thyroid tumor cells were injected into the flanks of immunodeficient mice. We then intratumorally injected the antagomiR. This model can be used for several cancer types to test the significance of specific miRNAs or oncomiRs in vivo in human cancer cells. Advantages of this technique are that it is relatively simple to perform and also quick, with minimal animal suffering. A major disadvantage of the technique, however, is that it does not closely mimic the human condition because the cells are injected subcutaneously and not into the organ-of-origin. To overcome this limitation, we used a more robust model by performing surgical orthotopic implantation of bioluminescent thyroid tumor cells into the thyroid of the mice. Although this technique is more complicated and time consuming, it allows the study of tumor growth in its “native” environment, the thyroid. A strength of this method is that it permits the study of tumoral cell dissemination as metastasis, usually in the lungs, which can be evaluated with in vivo imaging software. Another strength is that the systemic delivery of the antagomiR mimics a potential human treatment and allows for the analysis of the thyroid response. Furthermore, the systemic injection of the antagomiR did not cause adverse effects on the animals, as parameters such as serum glucose levels or liver morphology were not altered17.
There are a few critical steps. In the orthotopic mouse model, the injection of the tumoral cells should be precisely performed in the thyroid gland and not in neighboring tissue. Thus, it is necessary to demonstrate with immunohistochemistry that the thyroid architecture is maintained and that the cells injected have infiltrated the thyroid and are not present out with the gland. In addition, for both xenograft and orthotopic models, the size of the tumors in the different groups of animals should be similar at the beginning of the treatment in order to follow a similar growth pattern in all cases. Finally, it is important to demonstrate that the expression of the targets downstream of the microRNA (in our case miR-146b) are altered after antagomiR treatment
As a possible future application of this model, patient cells could be injected into mouse models in order to analyze how the inhibition of a miRNA could affect the tumor growth of a specific patient, a useful approach for precision and personalized medicine based on miRNAs.
The authors have nothing to disclose.
We are grateful to Raquel Arocha Riesco for her assistance with the treatment and care of mice. We thank Dr. J. Blanco (Catalonian Institute for Advance Chemistry-CSIC) and Dr E. Mato (Institut de Reserca de l’Hospital de la Santa Creu i Sant Pau) Barcelona (Spain) for gifting the CMV-Firefly luc- IRES-EGFP and Cal62-Luc cells, respectively. Funding: SAF2016-75531-R (MICIU), Fondo Europeo de Desarrollo Regional FEDER, B2017/BMD-3724 (Comunidad de Madrid), and GCB14142311CRES (Fundación Española contra el Cáncer, AECC).
AntagomiR: mirVana miRNA inhibitor | Thermo Fisher | 4464088 | In Vivo Ready |
Basement Membrane Matrix: Matrigel Basement Membrane Matrix High Concentration | Corning | #354248 | |
DICER antibody | Abcam | ab14601 | IHQ: 1/100 |
In vivo delivery reagent: Invivofectamine 3.0 Reagent | Thermo Fisher | IVF3005 | |
In vivo imaging software: IVIS-Lumina II Imaging System | Caliper Life Sciences | ||
Negative control: mirVana miRNA Inhibitor, Negative Control #1 | Thermo Fisher | 4464077 | In Vivo Ready |
PCNA antibody | Abcam | ab92552 | WB: 1/2,000 |
PTEN antibody | Santa Cruz | sc-7974 | WB: 1/1,000 |
XenoLight D-Luciferin – K+ Salt Bioluminescent Substrate | PerkinElmer | 122799 | Diluted in PBS |