Analysis of Lymph Node Volume by Ultra-High-Frequency Ultrasound Imaging in the Braf/Pten Genetically Engineered Mouse Model of Melanoma

Summary

Melanoma is a very aggressive disease that quickly spreads to other organs. This protocol describes the application of ultra-high-frequency ultrasound imaging, coupled with 3D rendering, to monitor the volume of the inguinal lymph nodes in the Braf/Pten mouse model of metastatic melanoma.

Abstract

Tyr::CreER+,Braf^CA/+,Pten^lox/lox genetically engineered mice (Braf/Pten mice) are widely used as an in vivo model of metastatic melanoma. Once a primary tumor has been induced by tamoxifen treatment, an increase in metastatic burden is observed within 4-6 weeks after induction. This paper shows how Ultra-High-Frequency UltraSound (UHFUS) imaging can be exploited to monitor the increase in metastatic involvement of the inguinal lymph nodes by measuring the increase in their volume.

The UHFUS system is used to scan anesthetized mice with a UHFUS linear probe (22-55 MHz, axial resolution 40 µm). B-mode images from the inguinal lymph nodes (both left and right sides) are acquired in a short-axis view, positioning the animals in dorsal recumbency. Ultrasound records are acquired using a 44 µm step size on a motorized mechanical arm. Afterward, two-dimensional (2D) B-mode acquisitions are imported into the software platform for ultrasound image post-processing, and inguinal lymph nodes are identified and segmented semi-automatically in the acquired cross-sectional 2D images. Finally, a total reconstruction of the three-dimensional (3D) volume is automatically obtained along with the rendering of the lymph node volume, which is also expressed as an absolute measurement.

This non-invasive in vivo technique is very well tolerated and allows the scheduling of multiple imaging sessions on the same experimental animal over 2 weeks. It is, therefore, ideal to assess the impact of pharmacological treatment on metastatic disease.

Introduction

Melanoma is an aggressive form of skin cancer that often spreads to other skin sites (subcutaneous metastases), as well as to lymph nodes, lungs, liver, brain, and bones¹. In the last decade, new drugs have been introduced into clinical practice and have contributed to improving the life expectancy of metastatic melanoma patients. However, limitations remain, including variable time to and degree of response, severe side effects, and the insurgence of acquired resistance¹. Therefore, it is crucial to detect metastatic spreading at its early stages, i.e., when it gets to the local lymph nodes.

A biopsy of the local lymph nodes (sentinel lymph nodes) is usually performed to check for the presence of melanoma cells. However, ultrasound imaging is taking hold as a non-invasive method of detecting metastatic involvement, as it outperforms clinical evaluation and can help avoid an unnecessary biopsy^2,3,4. Furthermore, ultrasound imaging seems appropriate for lymph node surveillance, especially in the case of advanced age and/or comorbidities^5,6. The features that are detected by ultrasound analysis and allow the differentiation between normal and metastatic lymph nodes comprise increased size (volume), change of shape from oval to round, irregular margin, altered echogenic pattern, and altered (increased) vascularization⁷.

Tyr::CreER+,Braf^CA/+,Pten^lox/lox genetically engineered mice (Braf/Pten mice) have recently been made available to the scientific community as a tissue-specific and inducible model for metastatic melanoma⁸. In this animal model, primary tumors develop very quickly: they become visible within 2-3 weeks after the induction of the switch from wild-type (wt) Braf to BrafV600E and of the loss of Pten, while they reach a volume of 50-100 mm³ within 4 weeks. In the following 2 weeks, the growth of the primary tumor is accompanied by a progressive increase in metastatic burden in other skin sites, lymph nodes, and lungs.

Braf/Pten mice have been extensively used for multiple purposes including the dissection of signaling pathways involved in melanomagenesis^9,10, the identification of melanoma cells of origin^11,12,13, and the testing of new therapeutic options in terms of both targeted therapy and immunotherapy^8,14,15,16. Specifically, we used Braf/Pten mice to demonstrate that attenuated Listeria monocytogenes (Lmat) works as an anti-melanoma vaccine. When systemically administered in the therapeutic setting, Lmat is not associated with overall toxicity as it selectively accumulates at tumor sites. Furthermore, it causes a remarkable decrease in primary melanoma mass and a reduction in metastatic burden in the lymph nodes and lungs. At the molecular level, Lmat causes apoptotic killing of melanoma cells, which is due, at least in part, to non-cell-autonomous activities (recruitment on-site of CD4+ and CD8+ T lymphocytes)¹⁶.

When Braf/Pten mice are used for melanoma modeling, the growth of primary tumors and subcutaneous metastases can be monitored by caliper measurements. However, the involvement of lymph nodes and lungs needs to be investigated using an alternative technique, possibly a non-invasive one that allows researchers to follow the same animal over time. This paper describes the use of ultrasound imaging (Figure 1), coupled with a subsequent 3D volumetric analysis of the obtained data, for the longitudinal monitoring of the increase in size (volume) of inguinal lymph nodes.

Protocol

All methods described here have been approved by the Italian Ministry of Health (animal protocols #754/2015-PR and #684/2018-PR).

1. Melanoma induction

NOTE: Six-week-old Tyr::CreER+,Braf^CA/+,Pten^lox/lox mice [B6.Cg-Braf^tm1Mmcm Pten^tm1Hwu Tg(Tyr-cre/ERT2)13Bos/BosJ (Braf/Pten)] were used in this study (see the Table of Materials).

1. Treat the mice with 4-hydroxytamoxifen (4-HT) by applying 3 µL of 5 mM 4-HT on ~1 cm² of shaved skin of the upper back, as described previously^11,16,17, for 3 days in a row.
   NOTE: This will activate the Cre enzyme and cause a switch from wt Braf to BrafV600E and the loss of Pten. These two hits are enough to induce melanoma formation.
2. Observe that primary tumors develop at the site of the skin painting in 2-3 weeks and reach a volume of 50-100 mm³ in 4 weeks. Also, observe metastases to other skin sites, lymph nodes, and lungs at this time point (t₀).
3. Use calipers to measure the volume of the primary tumor and of subcutaneous metastases, and ultrasound imaging to measure the volume of the inguinal lymph nodes. Repeat these measurements after one week (t₁, 5 weeks after skin painting) and after two weeks (t₂, 6 weeks after skin painting).
4. At the last time point, euthanize the mice by overdosing with gaseous sevorane.
5. Analyze the primary tumor and lymph nodes by visual inspection, then excise them for histological studies, as reported in¹⁶.

2. Imaging procedure

1. Place the mouse in an induction chamber for gas anesthesia and supply 3% isoflurane in pure oxygen until the animal is fully anesthetized. Verify the depth of anesthesia by lack of response to paw pinch.
2. Transfer the animal to a heated board - a constitutive part of the UHFUS imaging station - holding the animal in a supine position. Use a rectal probe lubricated with petroleum jelly to measure the body temperature. Adjust the board temperature to maintain the mouse's body temperature in the physiological range (36 ± 1°C).
3. Moisten the mouse's eyes with vet ointment to prevent dryness during anesthesia. Supply narcotic gas (1.5% isoflurane in pure oxygen) through a mouse's nose mask. Adjust the percentage of isoflurane to maintain the correct depth of anesthesia.
4. Coat the fore and hind paws with conductive paste and tape them to the ECG plate electrodes embedded in the board. Check that the physiological parameters (heart rate, respiration signal, and core body temperature) are correctly acquired and displayed.
5. Remove hair from both inguinal areas by applying a depilatory agent and coat them with an acoustic coupling medium.
6. Clamp the UHFUS linear probe (40 MHz center frequency) into a specialized 3D motor embedded in the UHFUS imaging station, allowing automated and stepwise movement of the probe.
7. Properly orient and adjust the position of the ultrasound probe to obtain short-axis images of the inguinal lymph node (left/right), and place the region of interest in the focal zone.
8. Scan the entire volume of the inguinal lymph node as a sequence of 2D B-mode images, as described previously¹⁸. Acquire images at multiple levels of the lymph node by linear movement of the transducer with step sizes on a micrometer scale, to generate 3D data in terms of automatically respiration- and cardiac-gated cine loops.
9. Set the image recording with the following parameters: scan distance ranging between 2 and 5 mm (depending on lymph node size); step size 44 µm, with an outcome of 46-114 scan steps/lymph node slices and an acquisition time of 1-3 min per animal. Digitally store the acquired images in raw format (DICOM) for further offline analyses.
10. At the end of the imaging session, discontinue the gas anesthesia and allow the animal to recover on the heating board in sternal recumbency. Take care of the animal until it has regained sufficient consciousness to maintain the prone position.

3. Post-processing of ultrasound images

1. Open the dataset of DICOM 3D images of the left/right inguinal lymph node in the software platform for ultrasound image post-processing.
2. Segmentation:
   1. Select Multi-slice Method to visualize both the current frames and thumbnails of all frames corresponding to each image captured during the 3D acquisition.
   2. Select the thumbnail of the first frame to load it into the contouring view. In the contouring view, left-click on the mouse to drop points along the border of the lymph node. Once the desired number of points has been set (range 10-15), right-click to complete the contour.
   3. After the first contour is completed, use the thumbnail view to select the next image for contouring. If required, skip over several images (average of 3 frames) between contours to reduce the number of manual traces needed for each 3D volume.
      NOTE: The software platform for ultrasound image post-processing will automatically generate contours between manual traces, thereby reducing the time of analysis.
   4. Repeat this process until the entire volume has been outlined. Once complete, click on Finish.
3. Generation of the 3D wireframe and volume measurement:
   1. While in the 3D mode window workspace, click on the Volume measurement icon beneath the image display area to activate the surface view.
      NOTE: The surface view creates a compilation view that maps the user-generated volume to the acquired image. The surface view can be rotated into any desired position.
   2. Take note of the volume measurement listed in the lower left-hand corner of the cube view.
      NOTE: Segmentation and 3D volume generation can also be obtained using custom-developed software and/or freely available/commercial software for general image processing. Starting from manual segmentation, the software should provide a mathematical and/or pixel-level description of the lymph node contours. These contours would be combined in a 3D space to render the external surface of lymph nodes. All steps described in the imaging procedure and the post-processing of ultrasound images are summarized in Figure 2.

Results

After skin painting of Tyr::CreER+,Braf^CA/+,Pten^lox/lox mice with 4-HT, Cre activity is induced, due to which there is a switch at the genomic level from wt Braf to BrafV600E, while Pten is lost (Figure 3A). In 2-3 weeks, mice develop on-site primary tumors with 100% penetrance. After four weeks from 4-HT treatment (t₀), primary tumors reach a volume of 50-100 mm³, and their growth can be measured by calipers for an additiona...

Discussion

The data obtained in this study attest the ability of ultrasound imaging to monitor the metastatic involvement of inguinal lymph nodes of the Braf/Pten mouse model of metastatic melanoma. As shown previously¹⁶, this technique is especially useful to assess the efficacy of drug treatment. This is because it allows the monitoring of the change in lymph node volume in the same animal over time, by comparing the measurements collected at t₁ and t₂ with those colle...

Materials

Name	Company	Catalog Number	Comments
4-hydroxytamoxifen	Merck	H6278	drug used for tumor induction
B6.Cg-Braftm1Mmcm Ptentm1Hwu Tg(Tyr-cre/ERT2)13Bos/BosJ (Braf/Pten) mice	The Jackson Laboratory	013590
Blu gel	Sooft Ialia		ophthalmic solution gel
BRAFV600E antibody	Spring Bioscience Corporation	E19290
IsoFlo (isoflorane)	Zoetis		liquid for gaseous anaesthesia
MLANA antibody	Thermo Fisher Scientific	M2-7C10
Sigma gel	Parker		electrode gel
Transonic gel clear	Telic SAU		ultrasound gel
Veet	Reckitt Benckiser IT		depilatory cream
Compact Dual Anesthesia System	Fujifilm, Visualsonics Inc.		Isoflurane-based anesthesia system equipped with nose cone and induction chamber
MX550S	Fujifilm, Visualsonics Inc.		UHFUS linear probe
Vevo 3100	Fujifilm, Visualsonics Inc.		UHFUS system
Vevo Imaging Station	Fujifilm, Visualsonics Inc.		UHFUS imaging station and Advancing Physiological Monitoring Unit endowed with heated board
Vevo Lab	Fujifilm, Visualsonics Inc.		software platform for ultrasound image post-processing