We present a method for studying tumor cell redissemination from lung metastases involving a surgical protocol for the selective photoconversion of lung metastases, followed by the identification of redisseminated tumor cells in tertiary organs.
Metastasis – the systemic spread of cancer – is the leading cause of cancer-related deaths. Although metastasis is commonly thought of as a unidirectional process wherein cells from the primary tumor disseminate and seed metastases, tumor cells in existing metastases can also redisseminate and give rise to new lesions in tertiary sites in a process known as "metastasis-from-metastases" or "metastasis-to-metastasis seeding." Metastasis-to-metastasis seeding may increase the metastatic burden and decrease the patient's quality of life and survival. Therefore, understanding the processes behind this phenomenon is crucial to refining treatment strategies for patients with metastatic cancer.
Little is known about metastasis-to-metastasis seeding, due in part to logistical and technological limitations. Studies on metastasis-to-metastasis seeding rely primarily on sequencing methods, which may not be practical for researchers studying the exact timing of metastasis-to-metastasis seeding events or what promotes or prevents them. This highlights the lack of methodologies that facilitate the study of metastasis-to-metastasis seeding. To address this, we have developed – and describe herein – a murine surgical protocol for the selective photoconversion of lung metastases, allowing specific marking and fate tracking of tumor cells redisseminating from the lung to tertiary sites. To our knowledge, this is the only method for studying tumor cell redissemination and metastasis-to-metastasis seeding from the lungs that does not require genomic analysis.
Metastasis is the leading cause of cancer-related deaths1. Metastatic cancer arises when cells from the primary tumor disseminate throughout the body and proliferate into clinically detectable tumors in distant organs2,3.
Although metastasis is commonly thought of as a unidirectional process wherein tumor cells disseminate from the primary tumor and colonize distant organs4, increasing clinical and experimental evidence suggests that a more complex, multi-directional process is at play. It has been shown that circulating tumor cells can reseed the primary tumor (if still in place)5,6,7,8,9, and tumor cells from existing metastatic foci can travel to tertiary sites and give rise to new lesions10,11,12,13. Indeed, evidence from recent genomic analyses suggests that some metastatic lesions arise not from the primary tumor, but from other metastases-a phenomenon known as "metastasis-from-metastases" or "metastasis-to-metastasis seeding"14,15,16. Metastasis-to-metastasis seeding can perpetuate the disease process even after removal of the primary tumor, increasing metastatic burden, and decreasing patient quality of life and survival. Therefore, understanding the processes behind metastasis-to-metastasis seeding is crucial to refining treatment strategies for patients with metastatic disease.
Despite the potentially severe clinical implications, little is known about metastasis-to-metastasis seeding, due in part to logistical and technological limitations. Human studies are limited by a paucity of clinical samples. Clinical resection and biopsy of metastatic lesions are uncommon, as is the biopsy of seemingly healthy organs, where single disseminated tumor cells may lurk. This means that human studies are typically only possible using autopsy samples from individuals whose primary tumors are either still in place or were previously resected but are still available to researchers. When such samples are available, lineage analyses of cancer progression must be performed using sequencing methods14. However, bulk sequencing of matched primary tumors and metastases does not have the sensitivity needed for comprehensive lineage tracing. For instance, bulk sequencing of one lesion may reveal a subclone that is undetectable in any of its matched lesions. In this case, one would be unable to determine the origin of this subclone. It may have been present in the primary tumor or another metastasis at a frequency below the limit of detection, or it may have arisen after the initial colonization of the metastatic lesion it was found in. Single-cell sequencing provides increased sensitivity, but its high cost limits the large-scale application of this technique. The retrospective nature of these studies also means that they provide limited insight into transient metastatic events and the disease landscape at different time points.
In animal models, recent technological advances now allow for prospective phylogenetic mapping with high spatial and temporal resolution17,18,19,20. These techniques utilize CRISPR/Cas9 genome editing to engineer cells with an evolving barcode – heritable mutations that accumulate over time. Upon sequencing, the lineage of each cell can be traced based on the mutational profile of its barcode17,18,19,20. Indeed, such technology is already being used to map metastasis-to-metastasis seeding. In a recent paper, Zhang et al. demonstrated that breast and prostate cancer cells in bone metastases redisseminate from the bone to seed secondary metastases in multiple organs21.
While these novel methods have great potential to generate detailed, high-resolution phylogenetic maps of cancer progression, they are highly impractical for those studying the exact timing of metastasis-to-metastasis seeding events and what promotes or prevents them. Filling these knowledge gaps is crucial to refining our understanding and treatment of metastatic cancer, but there is a noticeable lack of technologies to facilitate such studies. To address this need, we recently developed – and present herein – a novel technique that allows us to specifically mark tumor cells via photoconversion in a metastatic site (the lung) and subsequently reidentify them in tertiary organs. Using this technique, we recently showed that breast cancer cells do redisseminate from lung metastases and seed tertiary organs13. This technique may also be used to determine the timing of redissemination events within a narrow window and quantify redisseminated tumor cells, facilitating the study of organotropism of redisseminated cells and what promotes/prevents redissemination.
While photoconversion and locally inducible cre/lox systems that permanently replace one fluorescent protein with another have been previously used to mark and track tumor cells11,22,23, to our knowledge, no approach for spatiotemporal marking of tumor cells has been optimized to target the lung – one of the most common sites of metastasis among men and women diagnosed with any of the 14 most common cancers24. Any cancer cell type and any protocol for lung metastasis generation may be used with our procedure, making it broadly useful for metastasis researchers. All cancer cells used to generate lung metastases should express a photoconvertible or photo-switchable protein, and researchers may choose which protein to use based on their specific needs and resources. In this study, we used 6DT1 breast cancer cells that stably expressed the photoconvertible green-to-red fluorescent protein Dendra2 (6DT1-Dendra2 cells)25 tagged to the histone H2B. We injected 5.0 × 104 6DT1-Dendra2 cells into the fourth mammary fat pad of female Rag2-/- mice. Primary tumors were palpable between 12 and 16 days after injection and were not resected for the duration of the experiment. Spontaneous lung metastases developed between 19 and 26 days after tumor cell injection. Photoconversion surgeries were performed between 26 and 29 days after tumor cell injection. Mice were sacrificed by 72 h post surgery due to lung metastasis burden.
All procedures described in this protocol have been performed in accordance with guidelines and regulations for the use of vertebrate animals, including prior approval by the Albert Einstein College of Medicine Institutional Animal Care and Use Committee.
Prior to surgery, lung metastases are to be generated in mice using cancer cells that express a photoconvertible/photo-switchable protein; several protocols for lung metastasis generation have been published26,27,28.
1. Preparation for surgery
2. Surgery to expose the lung
NOTE: Perform all steps of the surgery (Figure 1), including photoconversion, in a hood or laminar flow cabinet to avoid contamination of the surgical field.
3. Lung metastasis photoconversion
NOTE: Details and variations on the following steps can be found in the Discussion.
4. Procedure to close chest wall
5. Sample processing and detection of photoconverted cells using tissue clearing
6. Sample processing and detection of photoconverted cells using tissue disaggregation
The steps of the surgery described in this protocol are illustrated in Figure 1. In brief, the mouse is anesthetized, and hair is removed from the left thorax. The mouse is then intubated and ventilated, which allows the mouse to receive oxygen while the thoracic cavity is open. Soft tissue is removed to expose the ribcage, and an incision is made in the 6th or 7th intercostal muscle. A retractor is inserted in the intercostal breach and released to spread the adjacent ribs and expose the left lung and any photoconvertible metastases thereupon. The lung and metastases are then exposed to blue light to photoconvert the exposed lesions. Following photoconversion, the retractor is removed, the ribs are sutured back together, and the overlying skin is closed. Finally, excess air is removed from the thoracic cavity to alleviate pressure on the lungs and allow the mouse to resume independent, spontaneous respiration.
The protocol described here can be adapted to a number of different photoconvertible/photo-switchable proteins. Ensuring that photoconverted cells reach their peak fluorescence intensity in the photoconverted channel can facilitate their identification in other organs. The conditions needed to achieve peak fluorescence intensity after photoconversion should be determined empirically, as they may vary with other aspects of the experiment such as the photoconvertible protein being used and the intensity of the light used to photoconvert. Figure 2B,C demonstrate how the fluorescence intensity changed in both the FITC (non-photoconverted) and TRITC (photoconverted) channels as a function of the duration of exposure to blue light in an ex vivo sample. We determined that 6 min of exposure to blue light produced the brightest signal in the photoconverted channel and that additional exposure led to photobleaching. We then measured the fluorescence intensity in both channels before and after 6 min of blue light exposure, in vivo, and obtained similar results (Figure 2D).
Next, the brain, liver, and non-photoconverted right lung were harvested from mice that underwent this surgery with (experimental group) and without (control group) the photoconversion lamp turned on. The tissues were optically cleared as previously described31, placed in a glass bottom dish, and imaged on a custom-built multiphoton microscope32 using a 25x 1.0 NA objective lens and an excitation wavelength of 880 nm. Fluorescence was captured for the green and red channels using the bandpass filters 525/35 nm and 580/60 nm, respectively. Photoconverted and non-photoconverted cells could be clearly identified in all three tissue types from mice that underwent the surgery with blue light exposure (Figure 3). As expected, only non-photoconverted tumor cells were found in the tissues of mice that underwent the surgery without blue light exposure.
Figure 1: Overview of the surgical protocol for the lung photoconversion surgery. (A) Depilated mouse placed on the surgical stage and taped in position. (B) A 10 mm circular clearing of skin and soft tissue over the ribcage. (C) Initial breach of intercostal muscle with blunt-edge micro-dissecting scissors. (D) Closed retractor inserted in 10 mm intercostal incision. (E) Open retractor exposing lung tissue. (F) Photoconversion lamp in place directly above open thorax of the mouse. (G) Simple continuous suture that incorporates the rib on both sides of the intercostal incision. (H) Suture tightened and secured to reestablish chest wall integrity. (I) Removal of air from thoracic cavity. Please click here to view a larger version of this figure.
Figure 2: Determining the conditions needed to achieve peak fluorescence intensity in photoconverted channel. (A) Representative images of exposed Dendra2-expressing lung metastases prior to blue light exposure in room lighting without any filters and at low magnification (1) and high magnification (2), and high magnification in the FITC (3) and TRITC channels with widefield illumination (4). (B) Fluorescence images of a Dendra2-expressing lung metastasis before blue light exposure, and after 2 min, 4 min, 6 min, and 8 min of blue light exposure, ex vivo. (C) Normalized mean fluorescence intensity of a Dendra2-expressing lung metastasis as a function of the duration of exposure to blue light, ex vivo. (D) Normalized mean fluorescence intensity of a Dendra2-expresssing lung metastasis before and after 6 min of exposure to blue light, in vivo. Scale bars = 2 mm (A), 400 µm (B). Please click here to view a larger version of this figure.
Figure 3: Representative images of photoconverted (top row) and non-photoconverted (bottom row) tumor cells found in the contralateral lung, brain, and liver of mice that underwent photoconversion surgery. Scale bars = 10 µm. Please click here to view a larger version of this figure.
Supplementary Figure S1: Custom-built 400 nm LED Array lamp.(A) Off and (B) on. CF = Cooling fan, R = Reflector, LED = High-power 400 nm LED. Please click here to download this File.
In this paper, we describe a surgical protocol for the selective photoconversion of tumor cells in the lung. This technique enables researchers to selectively mark tumor cells in the lung and track their fate by reidentifying them throughout the body at a later time point, facilitating the study of metastasis from lung metastases. Using this protocol, it was possible to visualize photoconverted cells in the brain, liver, and non-photoconverted right lung of mice that had undergone the surgery with photoconversion, indicating that these cells redisseminated from the photoconverted lung metastases to these tertiary sites. We did not find any red cells in the organs of mice that did not undergo photoconversion, indicating that natural autofluorescence in cells is not bright enough to be confused with photoconverted tumor cells.
Careful attention to certain aspects of the experiment can increase the success rate of this procedure. First, when generating lung metastases, the protocol for metastasis generation and photoconvertible/photo-switchable protein must be considered. It has been reported that some fluorescent proteins are immunogenic33,34,35,36. Thus, cancer cells expressing such proteins may not spontaneously metastasize in immunocompetent mice. Indeed, we observed that primary tumors form in both syngeneic, immunocompetent (FVB) mice and immunocompromised (Rag2-/-) mice upon orthotopic injection of 6DT1-Dendra2 cells. However, metastases only developed in Rag2-/- mice, indicating that Dendra2 may be somewhat immunogenic. Ways to overcome the potential immunogenicity of fluorescent proteins to generate lung metastases include using an immunocompromised mouse, generating metastases via tail vein injection, using a different photoconvertible/photo-switchable protein, or using a transgenic animal in which the photoconvertible/photo-switchable protein is genetically encoded so that the mice are tolerized to the protein37. The strength and stability of the fluorescent signal in the photoconverted channel must also be considered. For many photoconvertible/photoswitchable proteins, the signal in the photoconverted channel declines over time as the protein gets degraded and diluted with cell division. We and others have previously shown that photoconverted H2B-linked Dendra2 can be reliably detected for 7 days in dividing cells and 16 days in non-dividing cells38,39. Linking the photoconvertible protein to a protein with a low turnover rate can extend the half-life of the photoconverted signal. Methods for troubleshooting the lung metastasis model must depend upon the research aim and cancer cell type being used.
Second, it is important to time the surgery properly based on the disease model of interest. Those who work with disease models with fast-growing lung metastases may have a limited timeframe in which to perform the surgery before the mice become too overwhelmed by tumor burden to recover from surgery and/or survive the desired amount of time after surgery. A thorough understanding of the disease model timeline, as well as the use of non-invasive imaging strategies such as micro computed tomography, may help in timing this procedure.
Third, inadvertent cutting of major vessels during soft tissue removal can cause excessive bleeding, leading to poor visibility of the surgical field and/or exsanguination, as previously described40. Avoiding major vessels and using a cauterizing pen during surgery can prevent this. Fourth, when inserting and removing the retractor, it is crucial to keep the retractor blades parallel to the chest wall at all times. Angling the blades towards the lung increases the risk of lung damage, and angling the blades towards the chest wall increases the risk of breaching the chest wall in other areas. If inserting or removing the retractor in a parallel fashion does not seem possible, the blunt edge micro-dissecting scissors can be used to slightly increase the length of the intercostal incision before trying to insert/remove the retractor again. Finally, care should be taken to ensure that the ribcage is completely resealed at the end of the surgery. When suturing the ribs back together, stitching should be begun ~1 mm before the start of the incision and ended ~1 mm after. This helps to avoid residual openings on the ends of the incision. Additionally, take care to pull and tie the suture tight enough to eliminate space between the ribs, but not so tight that the sutures tear the adjacent intercostal muscles and create additional breaches. If resealing the thoracic cavity cannot be achieved, the mouse should be humanely sacrificed during the surgery, as it will have great difficulty or be unable to breathe on its own when mechanical ventilation is stopped.
The light source and exact methodology for photoconversion may vary between laboratories. Any light source and methodologies that are safe, consistent, and appropriate for the model may be used. For instance, photoconversion may be performed by transferring the mouse to a stereoscope or other microscope capable of shining light of the wavelength and intensity needed for photoconversion on the exposed lung. We use a custom-built 400 nm light-emitting diode array lamp set to its highest intensity to photoconvert Dendra2-expressing lung metastases (Supplementary Figure S1). To support the lamp above the mouse and ensure that the lamp is placed at the same distance above each mouse, we cut a 6.5 x 6.5 cm opening in the bottom of an open-topped cardboard box (18.4 cm (L) x 13.2 cm (W) x 7.3 cm (H)). We place the box upside down over the mouse and rest the 8.5 x 8.5 cm lamp over the 6.5 x 6.5 cm cutout (Figure 1F), so the lamp is supported, and the blue light can reach the exposed lung tissue.
It is important to note that some aspects of this protocol, including surgery and the use of anesthetics, may alter the dissemination kinetics and capacity of tumor cells41,42,43,44. As such, proper controls must be used to ensure that results are not confounded by the procedure.
The main limitation of this technique is that only metastases on the exposed portion of the lung will undergo photoconversion. Therefore, not every tumor cell that redisseminates from the lung will be photoconverted and any quantification of redisseminated cells will be relative. Despite this limitation, this technique can still provide valuable insight into tumor cell redissemination, including determining whether and when it happens, what promotes or prevents it, and the organotropism of redisseminated cells. To our knowledge, this is the first technique that allows selective marking and fate-tracking of tumor cells in the lung. In conclusion, this protocol describes a novel technique that can be used to facilitate and enhance the study of tumor cell redissemination and metastasis-to-metastasis seeding without genomic analysis. Targeting lung metastases makes the procedure useful and relevant to metastasis researchers studying most major cancer types.
The authors have nothing to disclose.
The authors would like to thank Wade Koba for his assistance with micro computed tomography (S10RR029545), Vera DesMarais and Hillary Guzik of the Analytical Imaging Facility for their training and assistance with microscopy, the Einstein Montefiore Cancer Center, the National Cancer Institute (P30CA013330, R01CA21248, R01CA255153), the Gruss Lipper Biophotonics Center, the Integrated Imaging Program for Cancer Research, a Sir Henry Wellcome Postdoctoral Fellowship (221647/Z/20/Z), and a METAvivor Career Development Award.
0-30 V, 0-3 A Power Supply | MPJA | 9616 PS | |
12 VDC, 1.2 A Unregulated Plug Supply | MPJA | 17563 PD | |
28 G 1 mL BD Insulin Syringe | BD | 329410 | |
400 nm light emitting diode array lamp | LedEngin Inc. | 897-LZPD0UA00 | Photoconversion lamp, custom-built (individual parts included below) |
5-0 braided silk suture with RB-1 cutting needle | Ethicon, Inc. | 774B | |
9 cm 2-0 silk tie | Ethicon, Inc. | LA55G | |
Baytril 100 (enrofloxacin) | Bayer (Santa Cruz Biotechnology) | sc-362890Rx | Antibiotic used in drinking water |
Buprenorphine | Hospira | 0409-2012-32 | Analgesic |
Cables (Cable Assemblies) 2.1 DC JACK-STRAIGHT 72" BLACK/ZIP CORD | Mouser | 172-7426-E | |
Cables (Cable Assemblies) 2.5 JK-ST 72" ZIP CD | Mouser | 172-0250 | |
Chlorhexidine solution | Durvet | 7-45801-10258-3 | Chlorhexidine Disinfectant Solution |
Compressed air canister | Falcon | DPSJB-12 | |
Extra Fine Micro Dissecting Scissors 4" Straight Sharp/Sharp 24 mm | Roboz Surgical | RS-5912 | Sharp Micro Dissecting Scissors |
Fiber-optic illuminator | O.C. White Company | FL3000 | Used during mouse intubation |
Gemini Cautery Kit | Harvard Apparatus | 726067 | Cautery pen |
Germinator 500 | CellPoint Scientific | GER 5287-120V | Bead Sterilizer |
Graefe forceps | Roboz | RS-5135 | |
High power LEDs – single color ultraviolet 90 watts | Mouser | LZP-D0UA00 | |
Infrared heat lamp | Braintree Scientific | HL-1 | |
Isoflurane SOL 250 mL PVL | Covetrus | 29405 | Anesthetic |
Isoflurane vaporizer | SurgiVet | VCT302 | |
Jacobson needle holder with lock | Kalson Surgical | T1-140 | |
Labeling tape | Fisher Scientific | S68702 | |
LED Lighting Reflectors CREE MP-L SNGL LENS REFLECTOR & LOC PIN | Mouser | 928-C11395TM | |
Long cotton tip applicators | Medline Industries | MDS202055 | |
Masscool / Soccket 478 / Intel Pentium 4/Celeron up to 3.4GHz / Ball Bearing / Copper Core / CPU Cooling Fan | CompUSA | #S457-1023 | |
Micro Dissecting Scissors 4" Straight Blunt/Blunt | Roboz Surgical | RS-5980 | Blunt Micro Dissecting Scissors |
Murine ventilator | Kent Scientific | PS-02 | PhysioSuite |
Nair Hair Removal Lotion | Amazon | B001RVMR7K | Depilatory cream |
Personnet mini retractor | Roboz | RS-6504 | Retractor |
Phosphate Buffered Saline 1x | Fisher Scientific | 14190144 | PBS |
pLenti.CAG.H2B-Dendra2.W | Addgene | 51005 | Dendra2 lentivirus |
Puralube | Henry Schein Animal Health | 008897 | Eye Lubricant |
Rodent intubation stand | Braintree Scientific | RIS 100 | |
Small animal lung inflation bulb | Harvard Apparatus | 72-9083 | |
SurgiSuite Multi-Functional Surgical Platform for Mice, with Warming | Kent Scientific | SURGI-M02 | Heated surgical platform |
Test Leads 48" TEST LEAD BANANA – Black | Mouser | 565-1440-48-0 | |
Test Leads 48" TEST LEAD BANANA – Red | Mouser | 565-1440-48-2 | |
Tracheal catheter | Exelint International | 26746 | 22 G catheter |
Wound closing system veterinary kit | Clay Adams | IN015 | Veterinary surgical stapling kit |