The present protocol describes the implantation and evaluation of melanoma in the murine choroid utilizing optical coherence tomography.
Establishing experimental choroidal melanoma models is challenging in terms of the ability to induce tumors at the correct localization. In addition, difficulties in observing posterior choroidal melanoma in vivo limit tumor location and growth evaluation in real-time. The approach described here optimizes techniques for establishing choroidal melanoma in mice via a multi-step sub-choroidal B16LS9 cell injection procedure. To enable precision in injecting into the small dimensions of the mouse uvea, the complete procedure is performed under a microscope. First, a conjunctival peritomy is formed in the dorsal-temporal area of the eye. Then, a tract into the sub-choroidal space is created by inserting a needle through the exposed sclera. This is followed by the insertion of a blunt needle into the tract and the injection of melanoma cells into the choroid. Immediately after injection, noninvasive optical coherence tomography (OCT) imaging is utilized to determine tumor location and progress. Retinal detachment is evaluated as a predictor of tumor site and size. The presented method enables the reproducible induction of choroid-localized melanoma in mice and the live imaging of tumor growth evaluation. As such, it provides a valuable tool for studying intraocular tumors.
Uveal melanoma (UM) is the most frequent intraocular primary malignancy in adults. Approximately 90% of ocular melanomas originate from melanocytes in the choroid region of the uveal tract1. UM is a major cause of morbidity and mortality, as it is estimated that close to 50% of patients develop metastatic disease, with the liver being the major site of metastasis2. Early treatment of primary lesions may reduce the chance of metastases, yet no effective treatment prevents metastases formation3.
The standard treatment of uveal melanoma includes irradiation therapy, which is associated with loss of vision due to optic neuropathy, retinopathy, dry eye syndrome, and cataract. Surgical resection is typically delayed until the growth of the lesion is recognized and characterized. However, such a delay may allow metastatic disease development4. In some cases, futile enucleation is required. Of course, this radical procedure compromises vision and results in dramatic aesthetic deterioration.
There have been many efforts dedicated to developing experimental models to study uveal melanoma. Preclinical animal models that allow accurate assessment of this malignancy are key for investigating novel diagnostic and therapeutic strategies for uveal melanoma. Experimental animal models of ocular melanoma are mainly based on the inoculation of tumor cells in mice, rats, and rabbits5,6. Mouse models are cost-effective and widely used for melanoma studies due to their rapid reproduction rate and high genome similarity to humans. The murine cutaneous melanoma cell line B16 is commonly utilized to inoculate C57BL6 mice and induce syngeneic tumors. When using this model to induce uveal melanoma, tumor-bearing eyes typically need to be enucleated 7-14 days after inoculation. Further, B16 is a highly invasive model. The immune-privileged nature of the eye supports metastasis, and metastases may typically be detected 3-4 weeks after tumor cell inoculation. Subcultures of the original B16 line display distinct metastatic properties6. For example, the Queens melanoma line has a high metastatic rate7,8. The B16LS9 cell line has dendritic cell morphology and was derived from liver metastases of C57BL/6 mice injected with the parental cutaneous melanoma line B16F19. When injected into the posterior compartment of the eye, these cells were shown to form intraocular tumors, which histologically resemble human uveal melanoma and form liver-specific metastases in C57BL/6, but not Balb/C, mice10,11,12. Genetically, the cells are characterized by higher expression of the c-met proto-oncogene, which acts as a cellular receptor for hepatocyte growth factor13. In contrast, B16F10, the 10th passage of the parental B16, primarily metastasizes to the lungs when inoculated intraocularly14. Both B16F10 and B16LS9 are pigmented12.
Several key challenges limit the success of murine uveal melanoma models. First, tumor cell reflux may lead to extraocular or subconjunctival melanoma. Second, tumor growth after intraocular inoculation of melanoma cells is often highly variable, posing difficulties in evaluating treatment and progress. Another major difficulty is the limited ability to follow tumor growth in vivo. While bioluminescent imaging, such as of luciferase expressing tumors, is commonly used to monitor ocular tumor growth15,16, it cannot provide information on the intraocular location of the tumor. Therefore, evaluation of the tumor is typically performed following enucleation of the eye10,17. This greatly limits the ability to characterize tumor progression and response to treatments extensively. Another major hurdle in studying uveal melanoma is the difficulty in monitoring lesions in pigmented mice. New approaches, which overcome these difficulties, are required to promote the research of uveal melanoma in animal models.
Optical coherence tomography (OCT) provides distinctive capabilities to image deep into the different sections of the eye in high resolution, which is unparalleled by other methodologies, including ultrasound18,19. OCT imaging has been used in animal models to study various ocular diseases20. Recently, OCT imaging was demonstrated as noninvasive means to evaluate intraocular tumor growth21. The protocol described here depicts the implantation of melanoma cells in the murine choroid and the utilization of OCT to predict intraocular tumor localization and size at the time of cell inoculation.
The experiments in the protocol were approved by the Israeli National Council on Animal Experimentation and comply with the ARVO Statement for using Animals in Ophthalmic and Vision Research. Female C57BL/6 mice, aged 8-10 weeks, were used for the present study and were exposed to 12/12 h light-dark cycles. The animals were obtained from a commercial source (see Table of Materials).
1. Cell culture
2. Animal preparation
3. Creating a conjunctival peritomy and scleral tract into sub-choroidal space
Figure 1: Tumor cell inoculation. (A) The supero-temporal limbal conjunctiva is held using intraocular forceps and pulled toward an infra-nasal position. (B) The tip of a 30 G needle is inserted to penetrate through the sclera, and excision is made to create a track into the sub-choroidal space. (C) A syringe loaded with cells and mounted with a 32 G needle is inserted into the track, and the cells are injected. Please click here to view a larger version of this figure.
4. Inoculation of melanoma cells
5. Assessing the location of the injection
6. Predicting tumor size based on RD height
7. Postoperative procedures
Eyes were examined via OCT immediately after injection of the B16LS9 cells. Local retinal detachment was observed after injection. The mice exhibited three patterns of RD: focal (Figure 2, upper panel), leakage to the vitreous (Figure 2, middle panel), and extended RD (Figure 2, bottom panel). Extended RD is likely caused by damage from the injection. There was an association between the pattern of RD immediately after injection and the localization of the tumors 5-7 days after injection. As demonstrated in Figure 2, focal RD was associated with tumor cell growth that was limited to the choroid. However, observation of cells in the vitreous after injection in animals displaying focal RD indicated tumor growth in the vitreal cavity in addition to the choroid. Finally, when extended RD, or RD at multiple sites, was observed after injection, tumors were dispersed throughout the choroid and vitreous after 5 days. Previous studies demonstrated that the characterization of tumors by OCT fully correlated with the histological examination21.
Figure 2: OCT-based prediction of tumor growth following sub-choroidal cell injection. A total of 7 × 104 B16LS9 cells in 2 µL of PBS were injected into the sub-choroid space of the mouse eyes. The injection site was visualized by OCT and fundus imaging immediately after cell injection (left). Dashed lines denote the detection of focal RD (top), cell material at the vitreous (middle), and extended RD (bottom). On the right, dashed lines indicate the tumor mass 5 days after injection. This figure is adapted from Zaks et al.21. Please click here to view a larger version of this figure.
The range of tumor sizes induced in this model may be divided into small, medium, and large (Figure 3). The height of the RD after injection was associated with the tumor size at 5 days post-injection, as measured by the horizontal and vertical OCT scans (Figure 4). The RD heights associated with the tumor sizes are shown in Table 1.
Figure 3: Determining tumor size. Tumor growth was evaluated by OCT scans (left) and real-time camera view of the fundus (right) 5-7 days following inoculation. Shown are representative images of small, medium, and large tumors. This figure is adapted from Zaks et al.21. Please click here to view a larger version of this figure.
Figure 4: Classification of tumor size based on OCT live imaging. Tumors were induced by sub-choroidal injection of 7 × 104 B16LS9 cells. Tumor growth was evaluated in live OCT imaging immediately and 5 days following injection. (A) A representative OCT measurement of RD height after injection (day 0, left, yellow line indicates height), tumor height and width in a horizontal OCT scan (day 5, middle, yellow lines indicate height and width), and an H&E-stained eye section (right). Inset: fundus image of the RD area. (B) The range of tumor sizes induced was divided into three groups. Tumor volume measurements of mice that presented small (S, n = 15), medium (M, n = 9), or large (L, n = 7) RD immediately after cell injection is represented. *p < 0.05. This figure is adapted from Zaks et al.21. Please click here to view a larger version of this figure.
RD height | Tumor volume |
<300 μm (small) | 0.0059 mm3 to 0.07 mm3 (mean 0.027 ± 0.005 mm3) |
300-400 μm (medium) | 0.015 mm3 to 0.15 mm3 (mean 0.056 ± 0.016 mm3) |
> 400 μm (large) | 0.05 mm3 to 0.36 mm3 (mean 0.017 ± 0.06 mm3) |
Table 1: The associated RD heights with the tumor size.
Uveal melanoma is a devastating disease for which novel therapeutic approaches are greatly needed. However, research on uveal melanoma and potential treatments is limited by the technical challenges of uveal melanoma animal models1,25. Ocular tumors, which are induced by intraocular injection of cancer cells, are highly variable in both localization and size, likely due to the small dimensions of the mouse eye. Such variability is an obstacle to the comprehensive evaluation of tumor progression. The experimental approach described herein allows one to assess intraocular tumor localization and predicted size immediately after sub-choroidal B16LS9 melanoma cell injection, based on live OCT imaging21. Over the years, the development of live imaging techniques provided advantages in cancer research by allowing to follow tumor progression throughout experiments, not just at pre-defined endpoints. Such methods include, for example, bioluminescent imaging utilizing reporter cells, which enables monitoring of the location of tumors and metastases in living animals, including identifying ocular tumors16. Another sophisticated method is photoacoustic imaging, which enables high-resolution imaging of cells and distinguishing melanoma cells from healthy cells26,27. However, the distinct advantage of OCT is the ability to identify the intraocular location of tumors and evaluate their size in live animals.
The protocol described here depicts the intraocular inoculation of melanoma cells by forming a peritomy and a tract into the sub-choroidal space, followed by the gentle insertion of a blunt needle to the tip of the tract and cell injection. This eliminates excessive puncture and directs the location of cell inoculation. The injection induces retinal detachment, which reflects the localization and size of the tumor that will be developed. Our observations suggest that tumors formed at the position of a local RD tend to be focal, and their size corresponds to the height of the RD after injection. In contrast, multiple RDs, or detecting cells at the vitreous after injection, typically result in dispersed tumors21.
It is plausible that focal RD reflects that the injection compacted most cells in a certain location, allowing for the formation of a localized tumor. On the other hand, multiple RD sites imply that the injection reached several locations across the retina, increasing the likelihood that tumor cells were implanted in numerous sites and formed dispersed tumors.
Evaluating the localization and predicted tumor size early after injection is particularly important when assessing tumor progressions, such as examining the effect of specific factors or potential treatments. Early determination can allow the inclusion of a homogenous study group, enhancing the model's reproducibility, thereby improving the study's accuracy, and saving valuable time and costs. Further, as OCT imaging of intraocular tumors provides accurate information on both the location and size of tumors in living animals, mice may be monitored for long-term experiments, for example, to evaluate metastases. Another difficulty in mouse models of melanoma is that mouse pigmentation, such as in C57BL/6 mice, often restricts the analysis of pigmented tumors. Therefore, another advantage of OCT imaging is that it is independent of pigmentation and may be applied to pigmented mice or lesions.
It must be noted that various parameters, such as the mouse strain, cell line, or injection technique, may affect tumor initiation and growth. Therefore, optimization of the model for specific strains or cells is advised. It must also be considered that secondary cataracts may develop over time1, limiting the ability to use OCT. While this is unlikely to occur in the short timeframe described herein, one should consider increasing the group size if performing longer experiments to allow for the exclusion of cataract eyes.
In summary, the described protocol utilizes a common model of B16LS9 melanoma, assessed by live OCT imaging, to develop a reproducible model for evaluating choroidal tumors in live animals. This approach may be utilized for future studies investigating the underlying mechanisms of uveal melanoma, new experimental models, and potential new therapies.
The authors have nothing to disclose.
This study was supported in part by grant 1304/20 from the Israel Science Foundation (ISF), Israel, for Arie Marcovich. We thank Shahar Ish-Shalom and Ady Yosipovich, from the Department of Pathology, Kaplan Medical Center, Rehovot, Israel, for histology analysis.
10 μL glass syringe (Hamilton Co., Bonaduz, Switzerland) | Hamilton | 721711 | |
30 G needles | BD Microbalance | 2025-01 | |
Atipamezole hydrochloride | Orion Phrma | ||
B16LS9 cells | from Hans Grossniklaus USA | ||
Buprenorphine | richter pharma | 102047 | |
C57BL/6 female mice | Envigo | ||
Essential vitamin mixture | satorius | 01-025-1A | |
Fetal bovine serum | rhenium | 10270106 | |
HEPES | satorius | 03-025-1B | |
Hydroxyethylcellulose 1.4% eye drops | Fisher Pharmaceutical | 390862 | |
InSight OCT segmentation software | Phoenix Micron, Inc | ||
Ketamine | bremer pharma GMBH (medimarket) | 17889 | |
L-glutamine | satorius | 03-020-1B | |
Medetomidine | zoetis (vetmarket) | 102532 | |
Ofloxacin 0.3% eye drops | allergan | E92170 | |
Optical coherence tomography | Phoenix Micron, Inc | ||
Oxybuprocaine 0.4% | Fisher Pharmaceutical | 393050 | |
Penicillin-streptomycin-amphoteracin | satorius | 03-033-1B | |
Phosphate buffered saline (PBS) | satorius | 02-023-1a | |
RPMI cell media | satorius | 01-104-1A | |
Sodium pyruvate | satorius | 03-042-1B | |
Surgical microscope | Zeiss | OPMI-6 CFC | |
Tropicamide 0.5% | Fisher Pharmaceutical | 390723 |