Lymphatic collection and cell isolation from mouse models for multiomic profiling

Abstract

Premetastatic cancer cells often spread from the primary lesion through the lymphatic vasculature and, clinically, the presence or absence of lymph node metastases impacts treatment decisions. However, little is known about cancer progression via the lymphatic system or of the effect that the lymphatic environment has on cancer progression. This is due, in part, to the technical challenge of studying lymphatic vessels and collecting lymph fluid. Here we provide a step-by-step procedure to collect both lymph and tumor-draining lymph in mouse models of cancer metastasis. This protocol has been adapted from established methods of lymph collection and was developed specifically for the collection of lymph from tumors. The approach involves the use of mice bearing melanoma or breast cancer orthotopic tumors. After euthanasia, the cisterna chyli and the tumor are exposed and viewed using a stereo microscope. Then, a glass cannula connected to a 1 mL syringe is inserted directly into the cisterna chyli or the tumor-draining lymphatics for collection of pure lymph. These lymph samples can be used to analyze the lymph-derived cancer cells using highly sensitive multiomics approaches to investigate the impact of the lymph environment during cancer metastasis. The procedure requires 2 h per mouse to complete and is suitable for users with minimal expertise in small animal handling and use of microsurgical tools under a stereo microscope.

Introduction

Tumor dissemination through lymphatic vessels represents an important route for metastasis in nearly all solid cancers and is a notable route of metastasis in melanoma, breast, pancreatic and prostate cancers^1–5. In the early stages of the premetastatic process, the tumor cells detach from the primary tumor through a process called intravasation and either enter the lymphatic vasculature or invade blood vessels by disrupting the endothelial barrier^4,6. Tumor cells and their derivatives, including proangiogenic factors, cytokines and extracellular vesicles, are transported via the lymph flow to the lymph nodes (LNs) that drain the tumor, where they can form LN metastasis and alter the LN microenvironment^7–9. The high interstitial pressure inside the primary tumor is thought to have a contributory role in the dissemination of tumor cells and tumor-associated factors to the tumor-draining LNs^7,10–14.

Once colonized, cancer cells from the LN can give rise to distant metastases through both lymph and blood vasculature^2,15–17. Cancer cell dissemination to distant tissue can occur independently from the lymphatic system by hematogenous metastasis. Unfortunately, we lack the biological knowledge of how the lymphatic environment influences cancer progression and, thus, have been unable to successfully target LNs to halt cancer progression using nonsurgical interventions. Undergoing surgical resection of LNs often leads to adverse side effects, including lymphedema, which impacts the quality of life, morbidity, financial burden and treatment-related distress of patients with cancer¹⁸. Moreover, it is increasingly recognized that LNs have a key role in eliciting the efficacy of immunotherapy treatments and that sparing LNs has the potential to improve patient prognosis^15,19–23. This has been supported by recent clinical trials demonstrating that neoadjuvant checkpoint blockade induces a tumor-specific immune activation in tumor-draining LNs, and maintenance of LNs is associated with an improvement in progression-free survival^24,25. Indeed, LNs serve as sites of immune activation, orchestrating the adaptive immune response against the tumor^26–28. Therefore, a better understanding of the role of the lymphatic system in cancer metastasis is crucial for developing novel therapeutic approaches that prevent metastasis, while sparing the adverse effects of surgical LN resection.

This research area has been limited, in part, because of the technical challenge of studying the lymphatic system. Here, we provide step-by-step instructions for collecting lymph from the cisterna chyli, as well as from tumor-draining lymph in mice, for the study of both cancer cells in lymph and of the lymphatic microenvironment. This approach enables the study of key functions of the lymphatic system, including antigen transportation, immune cell trafficking, lipid absorption and cellular/nutrient transportation. The approach can also be used to investigate the impact of the lymph environment on cancer cells, as well as the effects of cancer cells on the local-regional and distant lymph environment. Specifically, we discuss the use of this protocol for subsequent characterization of the metabolic and lipidomic profiles of lymph using high-sensitivity mass spectrometry approaches, as well as for lymph-derived tumor cell characterization by flow cytometry. With the use of this protocol, researchers can expect to successfully collect lymph fluid from tumor-bearing mice in their studies and obtain reliable data.

Overview of the lymphatic system

The lymphatic vasculature has substantial distinctions from the blood circulation. Lymphatic vasculature is a unidirectional conduit that collects and filters excess fluid and metabolites from tissue to return them to the blood circulation⁷. Beyond its function in immune surveillance, lipid absorption and tissue–fluid balance, the lymphatic system has essential roles during tumor metastatic dissemination^1–7. The lymphatic system consists of LNs and lymphatic vessels, which have unidirectional valves that transport lymph from local tissues toward the heart and systemic circulation via the thoracic duct and the right lymphatic duct⁷. While initial lymphatic vessels are made with loosely connected lymphatic endothelial cells with button-like junctions, the collecting lymphatic vessels consist of lymphatic endothelial cells, which are tightly connected with continuous zipper-like junctions and are covered with specialized lymphatic muscle cells^7,29. The contractile activity of the specialized lymphatic muscle cells and the valves of the collecting lymphatics are responsible for the unidirectional flow of the lymph returning to the thoracic duct and the right lymphatic duct⁷. Finally, the thoracic and right lymphatic ducts discharge lymph into the blood circulation⁷.

The average velocity of lymph in humans ranges from 0.008 ± 0.002 cm s⁻¹ to 2.3 ± 3 cm s⁻¹, while the pulsatile lymph velocity in mice ranges from 0.0053 cm s⁻¹ to 0.135 cm s⁻¹ (refs. 30–33). The lymphatic fluid is a colorless fluid made of water that contains proteins, white blood cells (mostly lymphocytes), salts, glucose and lipids³⁴. Our knowledge of lymph composition is limited due to the challenge to collect lymph and the heterogeneity of lymph composition, which is highly dependent on its locality and the physiological state of the body³⁵. More than 200 LNs in humans and more than 30 LNs in mice are integrated into the lymphatic vasculature by draining the lymph flow from the afferent collecting lymphatics^7,36,37. LNs have a bean-shaped structure that is highly organized and receives lymph through multiple afferent collecting lymphatics⁷. After filtration, lymph leaves the LN through efferent collecting lymphatics. LNs are found at the convergence of blood vessels allowing cellular entry into nodes via the blood. Spatially, the lymphatic vasculature is asymmetrical with the lymphatics from the bottom and left side of the body being drained into the thoracic duct and the lymphatics from the top right of the body being drained into the right lymphatic duct⁷ (Fig. 1).

Fig. 1 |. Protocol overview for lymph collection in tumor-bearing mice.

Illustration demonstrating lymphatic collection sites in tumor-bearing mice. After collection, lymph fluid is ready to use for subsequent analysis of its metabolites and lipid content by mass spectrometry. Lymph-derived tumor cells can be purified by cell sorting for metabolic, lipidomic and transcriptomic analyses. ALN, axillary lymph node; ILN, inguinal lymph node. The figure was created with BioRender.com.

Lymph collection in mice

Mice represent the most frequently used animal model in cancer research as they share similarities with human genetics and physiology, including the presence of the lymphatic system. Tumors derived from highly metastatic cancer cells can form metastases in tumor-draining and distant LNs, as well as distant organs in mice^16,17,38. While the process of hematogenous metastasis to distant organs has been extensively studied in tumor-bearing mice, our biological knowledge of lymphatic dissemination remains limited. This is due to the technical challenge of studying cancer cells in the lymphatic system, including difficulties in identifying lymphatic vessels and tumor-draining LNs, limited accessibility of the lymphatic system in vivo, limited sample quantities for analyses and the heterogeneous and dynamic changes of lymph. Gaining foundational knowledge of cancer cell characteristics in the lymphatics in mouse models will allow us to study lymphatic metastases in the context of previously established and well-characterized in vivo cancer models.

On the basis of our current knowledge of lymphatic vasculature and tumor lymphatics, we adapted a protocol to collect lymph and tumor-draining lymph in mouse models. To obtain a sample of lymph that represents the composition of this fluid, lymph samples are procured as close as possible to the point where the thoracic duct releases into the bloodstream. The cisterna chyli is the largest site distal to this location that allows for the efficient collection of sufficient lymph fluid for analytical analyses. Using a glass cannula, we can collect up to 2 μL of lymph from the cisterna chyli per mouse. To obtain lymph that drains the primary tumor, we collect the fluid from the collecting lymphatics on top of the tumor. Owing to the increased lymphatic pressure within the tumor’s surrounding tissue, we can collect up to 0.5 μL of lymph that drains from the tumor per mouse.

Applications of the method

We have applied this protocol to characterize the metabolic and lipidomic profiles of cancer cells metastasizing through lymph^38,39. Owing to the limited amount of lymph that can be collected, highly sensitive techniques are also needed to characterize very small amounts of tissue or cells. The Morrison laboratory developed a method that is an improvement over other approaches as it facilitates routine metabolomic/lipidomic profiling of flow cytometrically isolated low numbers of cancer cells using hydrophilic liquid interaction chromatography and high-sensitivity orbitrap mass spectrometry^38,39. These tools allowed for the isolation of melanoma cells in the lymph of mice transplanted with human or mouse melanomas, in which it was discovered that melanoma cells that metastasize through lymph are metabolically different from those that metastasize through blood³⁸. This recent work provided the first evidence that metabolic changes in cancer cells in LNs facilitate cancer cell survival³⁸. In this work, oleic acid was identified as one of the lipid adaptations of cancer cells in the LNs that promotes metastasis, and it is anticipated more metabolic and lipidomic adaptations that facilitate cancer cell survival in LNs remain unidentified³⁸. This protocol can be applied to downstream analyses of bulk and single-cell DNA or RNA sequencing, as well as cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) of flow-cytometrically isolated cancer cells and immune cells in lymph. Moreover, metabolites and lipids can be directly extracted from the collected lymph for metabolomic and lipidomic analyses of the lymph fluid.

Comparison with other methods

Before the emergence of the lymphatic metastasis field, lymph collection methods had been studied in the context of intestinal absorption, pharmacokinetic studies and immunological characterizations for decades. The lymph collection technique in mice was pioneered by cannulating the thoracic duct^40,41. This method has several advantages; it allows for the continuous collection of clear lymph on live animals with a low risk of contamination from other body fluids. However, this method often leads to the collapse of the vessel. Moreover, it is technically challenging because it requires an invasive surgery necessitating advanced microsurgical skills and extensive equipment.

Also, this method has been limited to the thoracic duct cannulation through the cisterna chyli and has not yet been able to be applied to the collection of tumor-draining lymph. Other approaches for lymph collection involve using filter paper absorption of lymph from the cisterna chyli and LN crushing⁴². However, LN crushing could lead to the enzymatic degradation of metabolites by enzymes released from crushed cells. Zawieja et al.⁴³ developed an advanced protocol for terminal lymphatic cannulation and lymph collection from different lymphatic vessels in mice for subsequent metabolic and flow cytometry profiling. Their approach allows for the acquisition of lymph not only from the thoracic duct but also from collecting lymphatics in various regions of the mouse. Briefly, this protocol involves securing a ~1 cm lymphatic vessel at both proximal and distal ends using nylon thread before cutting it at both ends above the sutures and transferring the ligated vessel into a custom-made vessel chamber or into a Petri dish that both contain a solution (extraction buffer or media). Then, the suture is removed, and the lymph fluid is flushed out of the vessels using a cannula or pressed out using forceps⁴³. This flushing method of lymph collection maximizes the cell yield that can be recovered from lymphatics, and this technique could also be applied to this current protocol after the cisterna chyli has been isolated to maximize cell yields in instances where this is the objective of the experimental design.

While this technique efficiently leads to the collection of pure lymph, the vessel—and therefore the collected lymph—are resuspended directly into a solution. This leads to the direct dilution of the lymph fluid and does not allow for obtaining a pure and undiluted lymph sample with a controlled volume. To our knowledge, currently, there is no study demonstrating how to collect lymph that drains the tumor in mice. In this context, we adapted these previously established methods of lymph collection from the cisterna chyli and developed a protocol for tumor-draining lymph collection. Our approach enables the collection of a pure lymph sample from both cisterna chyli and tumor-draining lymphatics.

Experimental design

The proposed experimental design has been established to adhere with the ARRIVE guidelines for the reporting of animal experiments, the PREPARE guidelines for planning animal experiments and the OBSERVE guidelines for the refinement of rodent cancer model^44–46. The number of mice necessary for each experimental group should be predefined using a statistical approach. We recommend using a power analysis, which is commonly used for determining the sample size for animal research^47,48. An online sample size calculator based on power analysis approaches can be used (such as https://clincalc.com/stats/samplesize.aspx).

Aging is known to impair the anatomy of the lymphatic vasculature and fluid homeostasis of lymph⁴⁹. Therefore, collecting lymph in mice younger than 4 months old is recommended to facilitate efficient lymph sampling. The composition of lymph is notably affected by the absorption of nutrients during food intake, and it remains to be determined the extent to which lymphatic composition is impacted by circadian rhythm^50–54. Indeed, dietary lipids are absorbed by lacteals in the gastrointestinal system and are packaged into chylomicrons, which transit via lymph⁵³. To limit variability due to differential dietary intake of the mice for lymph isolation, we recommend fasting the mice for 5 h before collection. Additionally, it is crucial to maintain the same diet across independent cohorts of animals and collect lymph samples at the same time of day to ensure consistent kinetics across independent cohorts.

We caution that the ability to detect low abundance but potentially biologically impactful differences is limited by the experimental variability and the sensitivity/specificity of the downstream analysis techniques. The differential abundance detection thresholds should be considered on an experimental basis to sufficiently power the study for the necessary lymph volume required to detect changes of interest. For low-abundance factors, we recommend pooling samples from independent animals within matched cohorts to obtain sufficient volume for analysis. Furthermore, to ensure robustness and reproducibility of the technique used to analyze the collected lymph for the factors of interest, we advise including positive and negative controls and applying data randomization techniques for all experiments. Lastly, both male and female mice can be used, as no key differences in LN number and distribution have previously been observed³⁶. However, sex difference in immune response have been described in mice transplanted with melanoma cells and thus considerations for sex differences should be given based on each individual experimental design⁵⁵. In this protocol, both male and female mice were used for melanoma models, but only female mice were used for breast cancer models. If little is known about the influence of sex on a scientific question, it is recommended to include both male and female mice to consider sex as a potential biological variable⁵⁶.

This protocol can be performed for most of the available strains. However, it is important to note that mouse strain-dependent effects are present, and this should be considered when conducting experiments in mice with different backgrounds (such as NOD-SCID-Il2rg^−/− (NSG) compared with C57BL/6J). First, Evan’s blue dye does not efficiently stain the LN and lymph vessels of NSG mice. Second, LN and lymphatic vessels, including cisterna chyli, are surrounded with adipose tissue in NSG mice, making lymphatic visualization difficult in this mouse model. Also, it is conceivable that a large LN metastasis could alter lymph flow from a primary tumor and limit tumor draining material collected. We recommend cautious interpretation of samples resulting from mice where there is evidence (visually or by Evan’s blue) that could suggest that the LN may be physically occluded.

This protocol has been optimized to be performed on mice killed by cervical dislocation. Terminal lymph collection has been successfully performed before on anesthetized mice⁴³. However, collecting lymph on euthanized mice have several benefits that need to be considered. First, it enables a higher yield and purity of lymph because postmortem collection facilitates uninterrupted access to the lymphatic vessels and a reduce risk of contamination by other fluids (especially blood). It is technically easier and more precise to collect lymph when there is no movement or breathing. Second, euthanasia avoids prolonged stress and pain that could occur during the procedure, preventing distress and potential alteration of the collected sample. Prolonged exposure to anesthetics might not only change the physiological composition of lymph but also decrease the lymphatic contractility^57,58. Finally, euthanasia reduces the risk of complications associated with prolonged anesthesia, including respiratory and cardiovascular instability, that might affect the mouse’s condition while collecting lymph. It is important to mention that cervical dislocation can damage the thoracic duct. However, damage to the thoracic duct often occurs in the neck region of the animal, which does not affect the efficiency of lymph collection from the cisterna chyli located close to the left kidney, far from the neck. Overall, we recommend considering all the aspects mentioned above when choosing between euthanasia and terminal anesthesia before collecting lymph to obtain high quality and consistent samples. Finally, the range of cancer and noncancer cell populations in the lymph differs greatly based on the experimental design (that is, metastatic capacity varies between models) and needs to be considered to efficiently isolate the cell population of interest. The experiment must be performed at least twice and in blind fashion, when possible, to assure reproducibility and robustness of the results.

Expertise needed to implement the protocol

The use of animals requires a proper license and approval from the Institutional Animal Care and Use Committee (IACUC) or the appropriate governmental department for nonacademic studies. Adherence to ethical standards and regulations is necessary for the proper conduct of experiments on animals. This method requires a researcher to have small animal-handling skills and expertise in anesthesia and terminal procedures. Specifically, conducting this protocol necessitates expertise in handling microsurgical tools when working under a stereo microscope.

Limitations

The major limitation of mouse lymph collection is the small volume of lymph that can be collected from one puncture, which ranges from 0.5 to 2 μL. Therefore, only highly sensitive analytic profiling can be efficiently performed on lymph. and it may be necessary to pool samples from independent animals. Secondly, this protocol is a terminal procedure and does not permit continuous lymph collection on live animals over time. Finally, isoflurane anesthesia can alter mouse metabolism, which might affect lymph composition. Alternatively, euthanasia by carbon dioxide inhalation can be used when terminal perfusion is not required.

Regulatory approvals

Regulatory approval is needed from the IACUC of the home institution or the appropriate governmental department for nonacademic studies before starting the procedures from animal euthanasia to lymph collection (from Steps 1 to 24). The duration of this process can vary depending on how frequently the committee meets and on revisions required after the initial submission. This phase can last anywhere from 2 to 6 months.

Materials

Biological materials

Eight-week-old mice are ideal for efficient lymph collection experiments. In this manuscript, we have performed this procedure on the BALB/cJ mouse strain (RRID: IMSR_ JAX:000651) and the C57BL/6J mouse strain (RRID: IMSR_ JAX:000664) purchased from The Jackson Laboratory. This protocol can be performed for most of the available strains. To perform tumor-draining lymph collection experiments, we used mice orthotopically engrafted with both human patientderived xenografts (into NSG mice) and murine melanoma cell lines Yale University melanoma models, as described in our previous study³⁸. Briefly, for initial tumor implantation, the cells were resuspended in 50 μL of 1× PBS:Matrigel (1:1) and transplanted into mice by subcutaneous injection on the right flank of the animal. The tumor-draining lymph collection procedure was performed after tumor growth (tumor volume 2.0 cm³, ~4 weeks after implantation).

▲ CAUTION Any experiments involving live mice must be performed in accordance with the relevant institutional and national regulations. All animal studies were approved by the Harvard Medical Area Standing Committee on Animals IACUC and the US Department of Agriculture-accredited facility. All procedures concerning animal care and animal studies were consistent with the Guide for Care and Use of Laboratory Animals (IS00003460). Mice were group housed (three to five mice per cage), with autoclaved food and water provided ad libitum in a room temperature (~20 °C), pathogen-free facility with a 12–12 h light–dark cycle in standard static microisolator top cages. Maximum tumor diameter and body-weight loss permitted on this protocol were 2.0 cm and 20%, respectively. Tumor engraftment was initiated in 6–10-week-old female C57BL/6J or BALB/cJ mice.

Reagents

• Sterile PBS without calcium and magnesium, 1× (Corning, cat. no. 21-040-CV)

• Normal saline (Quality Biological, cat. no. 114-055-101)

• Ethanol, 70% (Decon Labs, cat. no. 2401)

• Sterile fetal bovine serum (FBS, Corning, cat. no. 35-010-CV)

• EDTA (0.5 M), pH 8.0, RNase-free (Invitrogen, cat. no. AM9260G)

• Anti-mouse CD326 (EpCAM) clone G8.8 BV421 (BD, cat. no. 563214; https://scicrunch.org/resolver/RRID:AB_2738073)

• Anti-mouse CD31 clone MEC 13.3 allophycocyanin (APC) (BD, cat. no. 551262; https://scicrunch.org/resolver/RRID: AB_398497)

• Anti-mouse CD45 clone 30-F11 APC (BD, cat. no. 559864; https://scicrunch.org/resolver/RRID: AB_398672)

• Anti-mouse Ter119/erythroid cells clone TER-119 APC (BD, cat. no. 557909; https://scicrunch.org/resolver/RRID: AB_398635)

• Anti-mouse CD44 clone IM7 BV786 (BD, cat. no. 563736; https://scicrunch.org/resolver/RRID: AB_2738395)

• Anti-mouse CD146 clone ME-9F1 (Biolegend, cat. no. 134708; https://scicrunch.org/resolver/RRID: AB_10640123)

• 7-AAD viability staining solution (Biolegend, cat. no. 420403)

• FcR blocking reagent, mouse (Miltenyi Biotec, cat. no. 130-092-575)

• Corn oil (optional, Sigma, cat. no. C8267)

• Evans blue (optional, Sigma, cat. no. E2129)

• Isoflurane, USP (Piramal Critical Care Inc, cat. no. 66794-013-25)

▲ CAUTION Isoflurane inhalation can cause drowsiness, damage the cardiovascular system and is considered dangerous to pregnant women. Wear protective equipment and handle in a fume hood and a highly ventilated environment.

Equipment

• Digital microscope (Andonstar, cat. no. AD249S-M) or stereo microscope (AmScop, cat. no. SM-3TZ-54S-5M or equivalent)

• Dual Gooseneck Lights Microscope Illuminator (AmScop, cat. no. LED-20WY)

• Cork board, 30 cm × 30 cm × 6 mm (Amazon, cat. no. B0BZP1CJ2R or equivalent)

• Surgical tape (3 M, cat. no. 2862)

• Extra-fine bone scissors (FST, cat. no. 14084–08)

• Thumb-dressing forceps, serrated, delicate 4.5-inch length; 1.3 mm tip width (Roboz, cat. no. RS-8120)

• Vannas micro dissecting spring scissors (Roboz, cat. no. RS-5640)

• Ceramic-coated Dumont no. 5 forceps Inox tip size 0.05 mm × 0.01 mm (Roboz, cat. no. RS-4925)

• Glass cannula 40–60 μM, 1.2 mm outer diameter, 0.68 mm (Living Systems Instrumentation, cat. no. GCP-40–60)

• Polyurethane (PU) tubing 0.040 inch ID, 0.055 inch outer diameter (SAI Infusion Technologies, cat. no. PU-065–100)

• Female Luer 1/16 inch, 1.6 mm polypropylene hose barb (Amazon, cat. no. B08414K3PP or equivalent)

• 1 mL syringe (BD, cat. no. 309659)

• 18G needle (BD, cat. no. 305195)

• Transfer pipet 3 mL (Cole-Palmer, cat. no. EW-06226–29)

• Gauze pads (Covidien, cat. no. 8044)

• Laboratory cotton swabs, small head (Growing Lab, cat. no. 394345)

• 0.2 mL microcentrifuge tubes (Bio-Rad, cat. no. TFI0201)

• 1.5 mL microcentrifuge tubes (Eppendrof, cat. no. 022363212)

• PCR Cold Prep Rack (Globe Scientific, cat. no. PCR-COLDPSBS)

• Liquid nitrogen (Med-tech, cat. no. 13XLC32 or equivalent)

• Ice

• Container with a secure lid for euthanasia using open-drop exposure to isoflurane (Thermo Scientific, cat. no. 2118–032)

  ▲ CAUTION The container must be constructed of nonporous material that is sanitizable or disposable and allows for constant visualization of the animal. The container should not be so large as to require excessive anesthetic.

• Approved exhaust hood or exhaust Snorkel

• Dewar Flask, high density polyethylene (Nalgene, cat. no. 63380–085-EA)

• 5 mL sterile FACS tube with cap (Falcon, cat. no. 352054)

• 5 mL sterile FACS tube with cell-strainer cap (Falcon, cat. no. 352235)

Equiment setup

The equipment set up is shown in Fig. 2.

Fig. 2 |. Experiment set up.

a, Photo of the dissection area. b, Photo of the glass cannula linked to the collecting syringe via the Luer lock adaptor.

Dissection area

Clear the dissection area and disinfect surfaces with 70% ethanol. Ensure that the gas scavenging unit is functioning properly before procedure initiation. Clean all instruments and autoclave all surgical tools before use. Prepare lymph collection supplies such as syringe, microcentrifuge tubes, gauze, 1× PBS and glass cannula for use. Obtain temperature control supplies such as PCR Cold Prep Rack, liquid nitrogen and Dewar flask immediately before beginning procedure. Prepare a 1 mL syringe linked to a female Luer lock syringe, which is followed by ~2 cm tubing (Fig. 2b). Prepare 0.2 mL microcentrifuge tube placed on the previously cooled PCR Cold Prep Rack.

Microscope

Prepare the digital microscope area, ensuring it is clear of all supplies. Turn on the microscope and set up the Lens L (60–240×) and the dual gooseneck lights illuminator so that both are focused onto the plane of view where the procedure will be performed. For image capture and video recording, ensure there is proper memory space on the microSD. A stereo microscope can be used instead of the digital microscope (‘Equipment’ section).

FACS buffer

Add 1% heat-inactivated FBS to 1× PBS without calcium and magnesium. The final solution should be stored at 4 °C.

▲ CRITICAL The FBS must be heat inactivated at 56 °C for 25 min and filtered.

To reduce cell clumping, it is advised to add 0.5–5 mM EDTA to the FACS buffer to prevent cell-to-cell adhesion (optional).

▲ CRITICAL The concentration of EDTA must be optimized for each cell type of interest as high levels can cause cell death.

Evans blue dye 2% solution (optional)

Resuspend Evans blue powder into sterile 1× PBS without calcium and magnesium at 2% final concentration. Vortex thoroughly until total dissolution of the powder. The final solution should be stored at room temperature for several weeks. Note that Evans blue dye should be used as a tracer for learning the protocol but cannot be used with sensitive metabolomic and lipidomic techniques, as it can interfere with analytical detection methods. Evans blue dye will be absorbed into the mouse’s skin and will induce a blue-tinged appearance (readily visible on ears and paws of mouse) for several weeks post injection with the dye.

Procedure

Animal euthanasia

● TIMING 5 min

▲ CAUTION The euthanasia procedure by open-drop exposure to isoflurane followed by cervical dislocation used in this protocol is approved by the Harvard Medical Area Standing Committee on Animals IACUC and US Department of Agriculture-accredited facility and was consistent with the Guide for Care and Use of Laboratory Animals (IS00003460). As an alternative euthanasia procedure, we recommend using CO₂ inhalation followed by cervical dislocation. The policies for mice euthanasia may vary depending on the institutions and those policies must always be followed.

• 1Prepare the container for open-drop exposure to isoflurane for mouse euthanasia. Turn on the exhaust Snorkel. Place a gauze imbibed with isoflurane into the container (1 mL isoflurane per 500 mL container volume). Place a physical separator on top of the imbibed gauze to avoid direct contact with the animal with isoflurane.

  ▲ CAUTION The use of inhalant anesthetics should not be done without proper ventilation. Do not use isoflurane on bench tops without approved anesthesia exhaust vents or in ducted-approved chemical fume hoods. The policies on the use of inhaled anesthetics may vary depending on the institutions, and those policies must always be followed.

• 2Transfer the animal to the anesthesia jar and close the lid tightly. Monitor the animal closely. Within ~1 min, the animal will become anesthetized. Initially, the respiratory rate will increase and then decrease. Clinical indications of a deep plane of anesthesia in rodents include the lack of a righting reflex (upon tipping the jar gently) and a 50% reduction in respiratory rate compared with pre-anesthesia levels (that is, to 80–100 breaths/min).
• 3Allow the animal to remain at a deep anesthetic plane for ~10 s before proceeding. Quickly, yet carefully, remove the animal from the jar and place it on a clean work surface. Replace the lid on the jar immediately.
• 4The anesthetic plane is determined by nonresponsiveness to toe pinch. Immediately afterward, perform a cervical dislocation by jerking the tail base of the animal while keeping the neck immobilized, either with the opposite hand or a cervical dislocation device.

  ▲ CAUTION Air dry used gauze inside an open anesthesia jar in a fume hood for 15 min, then wrap it in a plastic bag and dispose as biomedical waste.

Organ dissection

● TIMING 5 min

• 5Place the dorsal side of the euthanized mouse on the cork board and secure the extremities using 18G needles, dissection pins or surgical tape to position the ventral side of the mouse for dissection (Fig. 3a).
• 6Spray ethanol 70% on the ventral side of the euthanized mouse.
• 7Using the extrafine bone scissors and thumb-dressing forceps, create a ventral incision beginning from the caudal (hip level) to cranial (cervical level) along the central axis of the mouse. Then, make a secondary incision from the central caudal axis downward toward each of the lower hind limbs to allow for removal of the surrounding cutaneous tissue to expose the viscera.
• 8Using thumb-dressing forceps, remove the cutaneous tissue from both sides and secure laterally using 18G needles or dissecting pins. When dissecting a tumor-bearing mouse, the subcutaneous tumor should remain attached to the skin and will now be exposed on the lateral position of the euthanized mouse.
• 9Make an incision through the viscera caudally following the central axis up to the diaphragm to expose the internal organs using the extrafine bone scissors and the thumb-dressing forceps (Fig. 3a).

  ▲ CRITICAL To avoid blood leakage, which might contaminate the lymph sampling, it is recommended to leave the organs of the euthanized mouse in position while performing the cannulation technique.

Fig. 3 |. Detailed stages of the lymph collection from cisterna chyli.

a, The mouse limbs are attached to the cork board with dissection pins and viscera is exposed to reveal the cisterna chyli (yellow circle). b, The mouse is positioned below a digital microscope (Adonstar, Lens L, 60–240×). The cisterna chyli is cleaned with PBS and dried using a gauze pad and a cotton swab. Then, a glass cannula (attached to the 1 mL syringe/Luer/tubing apparatus, Fig. 2b) is slowly inserted into cisterna chyli. Lymph fluid is collected within the glass cannula by capillary action and gentle aspiration. Finally, the collected lymph is transferred into a cold 0.2 mL microcentrifuge tube. (1) mesenteric lymph duct, (2) superior mesenteric duct, (3) vena cava, (4) renal vein, (5) left kidney.

Lymph collection from the cisterna chyli

● TIMING 10 min

• 10Place the cork board with the secured mouse below the microscope. Adjust the dual gooseneck illuminator to allow for adequate visibility of the internal organs and adjust focus accordingly.
• 11Using ceramic-coated Dumont forceps, gently adjust the gastrointestinal tract and the stomach laterally to allow for access to the cisterna chyli. The cisterna chyli is located posteriorly to the abdominal aorta and anterior to the left renal artery. It can be recognized by the milky aspect of the lymph (Figs. 3b and 4 and Supplementary Video 1).

  ▲ CRITICAL Manipulate the organs gently to avoid blood leakage, which can contaminate the lymph sampling.

  ◆ TROUBLESHOOTING

• 12Rinse the area of the cisterna chyli with 1× PBS loaded into a 3 mL transfer pipet.
• 13Dry the region of the cisterna chyli using a gauze pad.
• 14Repeat Steps 13 and 14 once or twice, if necessary. The cisterna chyli area must be cleared of blood and body fluid for an optimal lymph collection.
• 15Completely dry the cisterna chyli gently by small pressure using a cotton swab (Fig. 3b and Supplementary Video 1).
• 16Load a glass cannula onto the 1 mL syringe/Luer/PU tubing.

  ◆ TROUBLESHOOTING

  (Fig. 5)

• 17Insert the glass cannula into the cisterna chyli perpendicularly to the central axis, along the lateral-to-medial axis. The glass cannula must be inserted nearly parallel to the cisterna chyli with an angle not exceeding ~15° to avoid blood contamination of the lymph sampling. When the glass cannula is well inserted into the cisterna chyli, the lymph fluid will begin to be taken up into the cannula by capillary action (Fig. 3b and Supplementary Video 1).

  ◆ TROUBLESHOOTING

• 18Slowly aspirate the lymph fluid by small pulling of the syringe plunger (Fig. 3b and Supplementary Video 1).

  ▲ CRITICAL STEP Ensuring a small opening size of the cannula tip is crucial to prevent possible pressure-related collapse of the cisterna chyli when collecting lymph.

  ◆ TROUBLESHOOTING

• 19Transfer the lymph fluid into a precooled 0.2 mL microcentrifuge tube. The glass cannula can be inserted two to three times to collect more lymph to obtain sufficient volume for analysis.

  ◆ TROUBLESHOOTING

• 20
  1. For metabolomic or lipidomic analyses

     • I
       Close the lid of the 0.2 mL microcentrifuge tube and directly snap-freeze the lymph sample by putting it into the Dewar flask filled with liquid nitrogen. Store the lymph samples at −80 °C for long-term storage.
  2. For subsequent isolation of cancer cells or immune cells from lymph fluid by flow cytometry

     • II
       Add 100μL of FACS buffer into the 0.2 mL microcentrifuge tube, close the lid and keep the sample on ice before proceeding to Step 25.
       ◆ TROUBLESHOOTING

Fig. 4 |. Anatomical variation of mouse cisterna chyli.

a, Cisterna chyli demonstrating high volume of lymph (10-week-old female BALB/c mouse). b, Cisterna chyli demonstrating low volume of lymph (14-week-old female C57BL/6J mouse). c, Concealed cisterna chyli embedded in adipose tissue (14-week-old female C57BL/6J mouse). d, Visualization of mouse cisterna chyli after corn oil oral gavage. A total of 300 μL of corn oil has been administrated orally by gavage to a 14-week-old female C57BL/6J mouse. After 1 h, the mouse has been killed and dissected to expose the cisterna chyli. The arrows in a–d indicate the cisterna chyli localization.

Fig. 5 |. Detailed stages of the lymph collection from tumor-draining lymphatic vessel.

a, After gentle exposition of the subcutaneous tumor, 10 μL of Evan’s blue dye solution (1%) has been injected into the tumor to trace the tumor-draining lymphatics (blue). The mouse is positioned below a digital microscope (Adonstar, Lens L, 60–240×). A glass cannula (attached to the 1 mL syringe/Luer/tubing apparatus, Fig. 2b) is slowly inserted into the tumor-draining lymphatic. b, Lymph fluid is collected within the glass cannula by capillary action and gentle aspiration. Finally, the collected lymph is transferred into a cold 0.2 mL microcentrifuge tube. c, Snapshot from Supplementary Video 2 showing the tumor-draining lymphatic from a subcutaneous B16 melanoma tumor from a C57BL/6j female mouse. The tumor-draining lymphatic is delimitated by the dashed yellow lines on the right panel. d, Tumor-draining lymphatic vessel from a 4T1 breast cancer tumor from a NSG female mouse. Right: the tumor-draining lymphatic is delimitated by the yellow dashed lines and the 4T1 tumor by the purple dashed lines.

Tumor-draining lymph collection

● TIMING 30 min

▲ CRITICAL Using similar techniques as described above in Steps 10–20, lymph may also be collected from mice bearing spontaneously metastasizing models of cancer metastasis. For these models, mice should be injected with cancer cells subcutaneously (melanoma) or in the mammary fat pad (breast cancer), and tumors should grow up to 2.0 cm in diameter. At this point, the mice can be killed, and the tumor-draining lymph can be collected for subsequent analyses.

• 21Complete Steps 1–9 as described above. Make an incision through the viscera caudally following the central axis up to the diaphragm using the extra-fine bone scissors and the thumb-dressing forceps. Take care to gently expose the subcutaneous tumor and the surrounding lymphatic vessels, which will be readily apparent by the stereoscope and can be distinguished from blood vessels by the lack of blood and the smaller vessel diameter.

  ◆ TROUBLESHOOTING

• 22Load a glass cannula onto the 1 mL syringe/Luer/PU tubing, as shown in Fig. 2b.
• 23Insert the glass cannula into the tumor draining lymphatic, inserting the tip of the cannula away from the tumor bed and taking care not to disrupt the tumor. The glass cannula must be inserted nearly parallel to the tumor-draining lymphatic with an angle not exceeding ~15° to avoid blood and tumor cell contamination of the lymph sampling. When the glass cannula is well inserted into the tumor-draining lymphatic, the lymph fluid will begin to be taken up into the cannula by capillary action (similar to cisterna chyli aspirations shown in Fig. 5). A representative image of the tumor-draining lymphatic vessel without Evan’s blue dye is represented in Fig. 5d (extracted from Supplementary Video 2).
• 24Continue with Steps 18–20 above to collect the tumor-draining lymph fluid. Pool tumor-draining lymph fluid from six to ten mice, and if desired, proceed with endpoint lymph fluid flow cytometry analyses as described in Steps 25–41.

Isolation of cancer cells from lymph fluid by flow cytometry

● TIMING 90 min

▲ CRITICAL Work on ice and using precooled centrifuge (4 °C) and buffers/mix for all the following steps.

▲ CRITICAL Do not vortex the cells to avoid stress and cell death for all the following steps.

• 25Transfer the whole sample (FACS buffer containing the lymph fluid with the cancer cells) into a 5 mL FACS tube.

  ◆ TROUBLESHOOTING

• 26Wash the cells: add 4 mL of cold FACS buffer and spin down the cells at 300g for 4 min at 4 °C. Discard the supernatant by reverting the FACS tube rapidly. FcR blocking with anti-CD16/CD32 if working with immune cells is optional.
• 27Add 100 μL of staining mix containing the conjugated antibodies for flow cytometry analysis.

  ▲ CRITICAL We recommend defining the antibody panel in advance. For cancer cell isolation, select antibodies that positively detect the cancer cells and antibodies that positively stain noncancer cells (including immune cells, endothelial cells and red blood cells). Finally, select a viability marker for the elimination of dead cells. For breast cancer cell and melanoma cell isolation, we proposed the flow cytometry panels shown in Tables 1 and 2, respectively.

• 28Resuspend the cells gently with a p200 pipet.
• 29Incubate in the dark for 15–20 min at 4 °C.
• 30Wash the cells: add 4 mL of cold FACS buffer and spin down the cells at 300g for 4 min at 4 °C. Discard the supernatant by reverting the FACS tube rapidly.
• 31Wash the cells a second time by repeating Step 28.
• 32Add 100 μL of viability staining mix (Tables 1 and 2).
• 33Resuspend the cells gently with a p200 pipet.
• 34Filter the cells through a 5 mL FACS tube with a 35 μM cell-strainer cap.
• 35Sort the cancer cells using a BD FACS Aria III cell sorter (or equivalent) into a 1.5 mL microcentrifuge tube containing 50 μL of FACS buffer. For melanoma and breast cancer cells, we recommend using a 100 μM nozzle.

  ▲ CRITICAL The flow cytometry panel needs to be adjusted depending on the type of cell to be sorted and the configuration of the sorter. As example, the cancer cell population correspond to the 7AAD⁻/CD31⁻CD45⁻Ter119⁻/EpCAM⁺CD44⁺ cells for the breast cancer cells and DAPI⁻/CD31⁻CD45⁻Ter119⁻/CD146⁺ cells for the melanoma cells.

• 36Spin down the sorted cells at 300g for 4 min at 4 °C.
• 37Discard the supernatant.
• 38Add 1 mL of cold normal saline to wash the sorted cells.
• 39Spin down the sorted cells at 300g for 4 min at 4 °C. Discard the supernatant.
• 40
  Select an option:

  1. For metabolomic, lipidomic or bulk RNA/DNA sequencing

     1. Directly perform the corresponding metabolite or lipids or RNA/DNA extractions, and sorting can be conducted directly into the extraction buffer using methods described previously³⁹.
     2. Alternatively, the cell pellet can be snap-frozen in liquid nitrogen and stored at −80 °C.
  2. For single-cell RNA/DNA sequencing or CITE-seq

     1. Resuspend the cells in 50 μL cold normal saline.
     2. Follow the procedure for scRNA/DNAseq or CITE-seq analyses according to the manufacturer protocols.

Table 1 |.

Example flow cytometry panel for breast cancer cell isolation from lymph fluid

Staining mix: the following antibodies are diluted in FACS buffer
Antibody/probe	Target cells	Conjugation	Dilution
Anti-EpCAM	Breast cancer cells	BV421	1:100
Anti-CD44	Breast cancer cells	BV786	1:200
Anti-CD31	Murine endothelial cells	APC	1:100
Anti-mCD45	Murine white blood cells (immune cells)	APC	1:200
Anti-Ter119	Murine red blood cells	APC	1:100
Viability staining mix: 7AAD is diluted in FACS buffer
7AAD	Dead cells	–	1:20

Table 2 |.

Flow cytometry panel for melanoma cell isolation from lymph fluid

Staining mix: the following antibodies are diluted in FACS buffer
Antibody/probe	Target cells	Conjugation	Dilution
Anti-CD146	Melanoma cancer cells	FITC	1:5
Anti-CD31	Murine endothelial cells	APC	1:100
Anti-mCD45	Murine white blood cells (immune cells)	APC	1:200
Anti-Ter119	Murine red blood cells	APC	1:100
Viability Staining Mix: DAPI is diluted in FACS buffer
DAPI	Dead cells	–	Final dose: 0.5 μg/mL

Troubleshooting

Troubleshooting advice can be found in Table 3.

Table 3 |.

Troubleshooting table

Step	Problem	Possible cause	Solution
11	Blood accumulation proximal to the cisterna chyli	A blood vessel has been damaged during the dissection procedure	Contain the blood leakage with gauze pads and cotton swabs
16	The cisterna chyli cannot be easily visualized	Physiological variability of the tumor and tumor-draining lymphatics per mouse (Fig. 4a–c)	Administer 300 μL of corn oil by oral gavage 1 h before the procedure (Fig. 4d). Note: if corn oil is used, this will interfere with metabolomic and lipidomic analyses, and this procedure can only be used with flow cytometric isolation of tumor cells for endpoint quantifications
18	Blood is aspirated into the glass cannula instead of lymph	The glass cannula has touched a blood vessel	If the blood leakage is contained, replace the cannula and reinsert into cisterna chyli to collect lymph. If the blood leakage is not contained, a new mouse sample may be required to collect a pure and undiluted lymph sample without blood contamination
19	Lymph is not aspirated when drawing back the syringe plunger	Lymph has obstructed flow in the cannula	Transfer the collected lymph into a 0.2 mL microcentrifuge tube and replace the cannula before recollecting lymph
20	Lymph cannot be flushed out from the cannula when depressing the syringe plunger	The plunger is depressed at its maximum, and no more pressure can be applied to flush out the lymph Lymph is clotted inside the glass cannula	Slowly detach the syringe from the Luer lock connector. Draw back the plunger to the maximum possible position (up to 1mL graduation). Reattach the syringe to the Luer lock connector and flush out the lymph into the 0.2 mL microcentrifuge tube. Prewet the glass cannula with anticoagulant (heparin or EDTA). This is applicable to nonmetabolic measurements
20.b	Lymph cannot be resuspended in FACS buffer	Lymph is clotted and/or cells aggregate	Add 0.5–5 mM EDTA into the FACS buffer. This is applicable to nonmetabolic measurements
21	Lymph vessels cannot be easily visualized in tumor draining bed	Physiological variability of the tumor and tumor-draining lymphatics per mouse	Inject ~20 μL of 1% Evan’s blue dye solution into the tumor to trace tumor-draining lymphatics. Note: if Evan’s blue dye is used, this will interfere with metabolomic and lipidomic analyses, and this procedure can only be used with flow cytometric isolation of tumor cells for endpoint quantifications
25	Cell recovery yields are unexpectedly low after flow sorting	The cell number in the collected lymph is low, and the cells are lost during the washing steps	Block the 5 mL FACS tubes with 2 mL of FACS buffer for 15–30 min to increase cell retention during washes. After incubation, pour out the FACS buffer. Do not rinse the tubes with water or PBS afterward and use the blocked tubes immediately

Timing

Steps 1–4, animal euthanasia: 5 min

Steps 5–9, organ dissection: 5 min

Steps 10–20, lymph collection from cisterna chyli: 10 min

Steps 21–24, tumor draining lymph collection: 30 min

Steps 25–41, isolation of cancer cells from lymph fluid by flow cytometry: 1 h 30 min

Anticipated results

Isolation of lymph fluid from the tumor-draining lymphatics as described above will yield ~0.5 μL of lymphatic fluid per mouse. Isolation of lymph fluid from the cisterna chyli as described above will yield ~2–4 μL of lymphatic fluid per mouse. When isolating lymph fluid from the cisterna chyli of tumor-bearing mice (for example, in melanoma in mice that bear a subcutaneous tumor that spontaneously metastasizes and that is ~2.0 cm in diameter), ~10,000 cancer cells can be isolated from 25 μL of lymphatic fluid, and this is a sufficient number of cells and fluid on which to conduct metabolomic and lipidomic profiling³⁷. Therefore, lymphatic fluid from six to ten mice per cohort can be pooled for these finite cancer cell analyses. The collected lymph can be directly used for metabolic and lipidomic profiling by mass spectrometry or for multiomic analyses of the cells isolated from lymph (Fig. 6a). We have successfully characterized the metabolic and lipidomic profile of melanoma cells isolated from lymph on as few as 10,000 cells^37,38. Additionally, we have been able to identify specific metabolic profiles of lymph compared with plasma (obtained by terminal intracardiac puncture) in C57BL/6J female mice using volumes of microliters of fluid (Fig. 6b,c). This result provides insight into the metabolic composition of lymph compared with plasma which is crucial for unravelling tissue-specific physiology in health and disease including cancer metastasis.

Fig. 6 |. Application and anticipated results obtained after lymph collection.

a, Illustration demonstrating the application for analyzing the collected lymph samples. The lymph sample can be directly prepared for metabolomics or lipidomics analyses by mass spectrometry. Alternatively, cells can be purified using a flow cytometry-based cell sorting. The purified cells can be analyzed using multiomics approaches. b,c, A principle component analysis (b) and heat map (c) of the metabolic profile of lymph and plasma from five C57BL/6J female mice (8–10 weeks) performed by liquid chromatography–mass spectrometry (LC–MS). One factor statistical analysis has been performed using MetaboAnalyst 6.0 using the peak intensities values measured by LC–MS. A heat map representing the top 25 dysregulated metabolites between lymph and plasma is shown in c (distance measure: Minkowski; clustering method: average).

Supplementary Material

Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41596-024-01081-0.

Key points.

• The elevated pressure in the lymphatics around the tumor is leveraged for the collection of lymph from each tumor and the subsequent metabolic and lipidomic characterization using mass spectrometry, as well as lymph-derived cell characterization by flow cytometry.

• Lymph can be alternatively collected by cannulating the thoracic duct; however, this requires surgery and specialized equipment, and has not been applied to tumor-draining lymphatics.

Data availability

The authors declare that the main data discussed in this protocol are available in the supporting primary research paper as ‘source data’ (https://doi.org/10.1038/s41586-020-2623-z). All photographic images shown in this manuscript are the original, unaltered photographs. The deidentified metabolomics heat map contains no source data, as it is provided as a representative example of the protocol application.