In the present protocol, we explain how to easily process and culture tonsillar mononuclear cells from healthy humans undergoing partial surgical tonsillectomy to study innate immune responses upon activation, mimicking viral infection in mucosal tissues.
Studying isolated cells from mucosa-associated lymphoid tissues (MALT) allows understanding of immune cells response in pathologies involving mucosal immunity, because they can model host-pathogen interactions in the tissue. While isolated cells derived from tissues were the first cell culture model, their use has been neglected because tissue can be hard to obtain. In the present protocol, we explain how to easily process and culture tonsillar mononuclear cells (TMCs) from healthy human tonsils to study innate immune responses upon activation, mimicking viral infection in mucosal tissues. Isolation of TMCs from the tonsils is quick, because the tonsils barely have any epithelium and yield up to billions of all major immune cell types. This method allows detection of cytokine production using several techniques, including immunoassays, qPCR, microscopy, flow cytometry, etc., similar to the use of peripheral mononuclear cells (PBMCs) from blood. Furthermore, TMCs show a higher sensitivity to drug testing than PBMCs, which needs to be considered for future toxicity assays. Thus, ex vivo TMCs cultures are an easy and accessible mucosal model.
Studies on human organs are restricted due to accessibility as well as obvious ethical reasons. However, they are essential to fully understand the complexity of human biology. Cultures of isolated cells (primary cultures or cell lines) are a standard system in cell biology studies due to their availability. While isolated cell cultures have allowed outstanding discoveries, the use of cell lines has come upon closer scrutiny because they do not fully mimic in vivo organ biology. However, the culture of three-dimensional cells or tissue explants is highly complex4,5,6. Indeed, a piece of tissue or organ is highly heterogenous because its cell composition differs depending on the localization in the tissue. Thus, using tissue blocks requires the analysis of many technical and biological replicates, leading to the need of a large number of donors or patients.
The mucosa-associated lymphoid tissues (MALT) are structurally similar to the lymph nodes but have unique functions, because their main role is to regulate mucosal immunity7. Unlike the lymph nodes, which are usually located at some distance from the tissues, MALT are generally located immediately below the epithelium of the mucosal tissue. Histologically, they are mainly composed of high concentrations of B and T cells, but also antigen-presenting cells such as macrophages and dendritic cells. MALT constitute about 50% of the lymphoid tissue in the human body. MALT are subdivided into nine groups depending on their location: GALT (gut-), BALT (bronchus-), NALT (nasal-), CALT (conjunctival), LALT (larynx-), SALT (skin-), VALT (vulvo-), O-MALT (organized), and D-MALT (diffused). The O-MALT is mainly composed of the tonsils of Waldeyer's tonsillar ring and is the most accessible MALT8,9. Indeed, tonsils located in the oropharynx constitute the major barrier protecting the digestive and respiratory tracts from (potential) invasive microorganisms10. In addition, the tonsils are covered by a fine stratified squamous non-keratinizing epithelium, supported by a capsule of connective tissue containing blood vessels, nerves, and lymphatics, providing easy access to the immune cells11,12. Furthermore, tonsillectomy, the surgical act of removing tonsils, is a common procedure performed on children having sleep-disordered breathing, making tonsils an easily available tissue13 in physiological settings.
Tonsils allow the study of immune cell response in pathologies involving mucosal immunity. Indeed, in HIV infection, because tonsils are composed of a high concentration of immune cells, they are the main target of viral replication but also produce a large amount of cytokines that are not detected in the circulation14,15. At steady states, rare populations of innate-like cells are present in various mucosal tissues, including the tonsils, but are essentially absent from blood.
Thus, mononuclear cells from tonsils (TMCs) are a more relevant and complex model than PBMCs and can answer more profound questions. On the other hand, the use of tissue explants can be complex and not always relevant to innate immune studies. Thus, we established a model to study mucosal immune activation using TMCs16. Here, we describe a method for efficient isolation of TMCs from fresh human tonsils. This method allows the recovery of a large number of immune cells while keeping their integrity for ex vivo studies.
The specimens are not collected specifically for research purposes and the study is not considered invasive. However, human tonsils collection requires ethical approval by the local relevant authorities. In our case, it was approved by the Comité de Protection des Personnes (IDRCB/EUDRACT: 2018A0135847). Furthermore, consent of each patient or legal representative is requested to obtain donors' personal data (e.g., sex, age, history of ENT infections) that can help interpret experimental results.
1. Handling of the Human Tonsillar Tissue
2. Dissection of the Human Tonsillar Tissue and Isolation of the Tonsillar Mononuclear Cells (TMCs)
3. Isolation of the TMCs by Cell Density Gradient
NOTE: The TMCs can be used after step 2.10. However, to obtain a clearer cell solution and to remove other cell debris and any red cells, it is advised to perform a cell density gradient isolation of the mononuclear cells.
4. Phenotyping of TMCs by Flow Cytometry
Antibody | Clone | Fluorochrome | Company |
Live-Dead | BV405 | ThermoFisher Scientific | |
CD3 | SP34-2 | V500 | BD Pharmingen |
CD8 | SK1 | Amcyan | BD Pharmingen |
CD8 | BW135/80 | VioBlue | Miltenyi Biotec |
CD4 | RPA-T4 | PE-Cy7 | BD Pharmingen |
CD45 | HI30 | PerCP-cy5.5 | BD |
CD19 | HD237 | ECD | Beckman Coulter |
CD20 | 2H7 | Alexa Fluor 700 | BD Pharmingen |
CD14 | M5E2 | PE-cy7 | BD Pharmingen |
CD14 | M5E2 | APC | BD Pharmingen |
CD16 | 3G8 | APC-H7 | BD Pharmingen |
CD56 | MEM188 | PE | BD Pharmingen |
CD123 | 7G3 | PE | BD Pharmingen |
BDCA-1 | L161 | Pacific Blue | BD Pharmingen |
BDCA-3 | AD5-14H12 | FITC | Miltenyi Biotec |
BDCA-4 | AD5-17F6 | APC | Miltenyi Biotec |
HLA-DR | G646-6 | PerCP-cy5.5 | BD Pharmingen |
CD11c | 3.9 | Alexa Fluor 700 | ebiosciences |
Table 1: List of antibodies for cell characterization by flow cytometry.
5. Example of Activation of TMCs and Measurement of Cytokine Production
We first characterized the immune profile of cells present in the culture and analyzed the amount of TMCs. We phenotyped the TMCs from tonsils with flow cytometry. As shown in Figure 1, all major immune cell types present in PBMCs from blood were represent in the TMCs from tonsils. However, in TMCs the frequency of all cell types, except B cells, were lower than in PBMCs.
Figure 1: Phenotyping of TMCs in tonsils. TMCs from healthy donors were obtained after dissection of human tonsillar tissue and isolation of cells. (A) Cells were stained, and data were acquired by flow cytometry. Data represent a representative experiment. (B) The table summarizes the percentage of each cell type in live TMCs in tonsils from six different healthy donors and compares each to the percentage of each cell type in PBMCs from blood, as defined in the literature18,19. Data are shown as the percentage of live cells SD. Please click here to view a larger version of this figure.
We then tested the immunological responses of the TMCs. One way to reproduce a cell's activation against a pathogen is to stimulate the innate immune sensors from the Toll-Like Receptors (TLRs) family. TLRs are mainly present in monocytes/macrophages, all dendritic cell (DC) subtypes, plasmacytoid dendritic cells (pDCs), and also in B cells. In this example, we studied TLR7 and TLR8, because they specialize in antiviral responses and most cases of tonsillitis (i.e., tonsil infections) are caused by a viral infection. Thus, TMCs from tonsils (graph in blue) from seven healthy donors were stimulated overnight with the TLR7/8 ligand resiquimod (R848). As in PBMCs (graph in green), activation through TLR7/8 signaling triggered production of type I interferons (IFN) and pro-inflammatory cytokines. In fact, the cytokine most highly produced in TMCs upon activation was IFNα, a member of the type I IFN family that is mainly involved in innate immunity against viral infection and is mostly produced by the pDCs. All cytokines tested, except IFN2/3 and IL-8, were produced by TMCs, although at lower levels than in PBMCs. Interestingly, some cytokines (mainly IL-8, but also IL-6, IL-10, and TNFα) were present in larger amount at basal levels. Furthermore, we compared the cytokine produced by isolated TMCs or by tonsillar tissue blocs prepared as described previously6. We measured all types of IFN produced in the supernatant of both cell cultures using the STING-37 cell line reporter assay as described previously17 and showed that we could only measure IFN levels in the isolated TMCs.
Figure 2: TMCs produced cytokines upon stimulation. (A) Purified and isolated TMCs (blue) and PBMCs (green) were stimulated overnight with R848 (5 µg/mL). Supernatants were retrieved and frozen until use. A bead-based immunoassay was performed to quantify multiple soluble cytokines simultaneously using a flow cytometer. The graphs represent the concentration in picograms per milliliter of each measured cytokine. The fold increases are annotated in red on the graphs. Mann-Whitney U test. (B) Purified and isolated TMCs from tonsils (blue) and tonsillar tissue blocs (orange) were stimulated for 24 h with Influenza A virus (IAV). A STING-37 reporter cell line assay was used to measure IFN production in the supernatant. Box and whisker plots with median ± minimum to maximum. Mann-Whitney U test. ***P < 0.001, **P < 0.01, *P < 0.05. NS = non-stimulated. Please click here to view a larger version of this figure.
Finally, we tested the viability of the TMCs. Several techniques are available to measure cell viability. Here, we used a luminescent cell viability assay, an easy and quick two-step assay. As shown in Figure 3, we clearly detected the cytotoxicity induced by compound 1 in TMCs, which was not toxic to PBMCs. Thus, TMCs showed a higher sensitivity to the drug tested.
Figure 3: Compound 1 was toxic to TMCs. Purified and isolated TMCs and PBMCs were preincubated with compound 1 (C1) at various concentrations for 1 h and then stimulated overnight with R848 (5 µg/mL). Supernatants were retrieved, 60 µL of cell viability solution was added, and luminescence was measured. The * represents the statistical analysis comparing C1 to NS. Box and whisker plots with median ± minimum to maximum. Kruskal-Wallis test with Dunn's post hoc correction. ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05. NS = non-stimulated. Please click here to view a larger version of this figure.
Human tonsils represent an integrative and physiological ex vivo model to study innate immune responses at the mucosal interface, because they mimic the role of a secondary lymphoid organ. Interestingly, the cellular composition of the TMCs is similar to the PBMCs and includes all the major cell populations, although their percentage can be different from PBMCs from blood (Figure 1). Additional populations can also be found, as all immune responses are initiated in tissues (mucosa or secondary lymphoid organs) and not in blood.
The use of human tissue explants as well as the use of cell culture or blood cells each have their advantages. However, for the of study cytokine secretion and cell activation, tissue explants were not the best model. In fact, we could not detect any IFN production in the supernatant of tissue explants (Figure 2B). We speculate that it is directly consumed by the surrounding cells. On the other hand, stimulation of blood cells (PBMCs) can mimic the primary response to infection but do not mimic what is happening in the tissues, where most of the cytokines are produced and where viruses replicate. Therefore, we set up a protocol to isolate and purify cells arising out of a secondary lymphoid tissue, the tonsil. We purified TMCs from healthy human tonsils, which allows us to investigate immune cell activation upon stimulation. However, one of the limitations of this model is that TMCs do not produce as many pro-inflammatory cytokines as PBMCs ex vivo upon TLR7/8 stimulation, although levels of IFNα, the major antiviral cytokine, is similar in both cultures.
The study of compound toxicity on TMCs versus PBMCs revealed that TMCs are more sensitive to a toxic drug (Figure 3). Thus, the toxicity of new drugs and future treatments should be tested in cells from tissues and not only on cell lines, PBMCs, or tissue explants. Therefore, the use of TMCs for drug testing seems to be a future application of the method described.
Tonsils from adults or children can be obtained and processed as described. However, we recommend obtaining tonsils from children for two main reasons: 1) Children's tonsils contain more cells than the adults'20. Indeed, tonsillar hypertrophy is particularly common in children, because they fight more childhood viruses. Thus, these large tonsils can become obstructive tonsils and need to be removed by a partial tonsillectomy to avoid complications like sleep apnea13; 2) Because adults' tonsils are usually removed after several episodes of ear-nose-throat (ENT) infections, these tonsils are less naïve and contain fewer cells.
It is critical to work on the tonsils as soon as possible after the removal surgery, ideally <3 h after collection. Indeed, right after removal and until dissection, the tonsils from each side are kept together in a 50 mL sterile vial containing around 20 mL of PBS, so that they are totally submerged.
Interdonor variability can represent a major issue for reproducibility. Indeed, variability in cellular composition between donors can be significant. Our protocol, using TMCs and not tissue explants, limits this variability because the cells from the whole tonsils are mixed before being plated and placed in culture, while tissue explants bring intradonor variability in cellular composition because the pieces come from different areas within the same specimen. Partial tonsillectomy can only be performed on noninflamed tonsils, which limits the inflammatory status variability between patients. We also recommend collecting tonsils surgically removed on different days. We have clearly noticed that inter-surgeon and inter-day variabilities are higher than interdonor variability (data not shown).
The authors have nothing to disclose.
This work was supported by the Agence National de la Recherche sur le SIDA et les Hépatites ANRS (J-P.H) for the experiments and N.B. fellowship (AAP 2017 166). N.S. acknowledges support from the ANRS for fellowship (AAP 2016 1), the European Molecular Biology Organization EMBO for Fellowship (LT 834 2017), the startup funding program "Baustein" of the Medical Faculty of Ulm University (LSBN.0147) and the Deutsche Forschungsgemeinschaft DFG (SM 544/1 1).
10 meshes steel grid – 1910 µm | Dutscher | 198586 | To put in the cell strainer Cellector |
60 meshes steel grid – 230 µm | Dutscher | 198591 | To put in the cell strainer Cellector |
70 µm white ClearLine cell strainers | Dutscher | 141379C | |
Anios Excell D detergent | Dutscher | 59852 | Detergent |
Antibiotic solution, 100x | Thermo Fisher | 15140122 | 100 U/mL Penicilium and 100 μg/mL Streptomycin – to add to culture media |
BD FalconTM Round-Bottom Tubes, 5 mL | BD Biosciences | 352063 | FACS Tubes |
Cell strainer Cellector, 85 mL and 37 mm diameter | Dutscher | 198585 | |
CellTiter-Glo (CTG) Luminescent Cell Viability Assay | Promega | G7572 | Viability assay |
Centrifuge 5810 R | Eppendorf | ||
Conical tubes Falcon 50 mL | Dutscher | 352070 | |
Curved tweezers | Dutscher | 711200 | |
Dimethyl sulfoxide (DMSO) | Sigma-Aldrich | D2650 | |
Dulbecco's Phosphate Buffered Saline (PBS) | Sigma-Aldrich | D8537 | Without calcium and magnesium |
EnVision | PerkinElmer | Measures the luminescence | |
Fetal Bovine Serum (FBS) | To add to culture media | ||
Fluorescence labeles antibodies | See Table 1 | ||
Glass Pestle | Dutscher | 198599 | |
Hepes (1M) | Thermo Fisher | 15630056 | Use at 20 mM |
Incubator | |||
LEGENDplex Human Anti-Virus Response Panel | BioLegend | 740390 | Bead-based immunoassay |
Lymphoprep | StemCell | 7801 | Density gradient medium |
Mr. Frosty container | Thermo Fisher | 5100-0001 | Slow freezing container |
Pierce 16% Formaldehyde (w/v), Methanol-free | Thermo Fisher | 28908 | |
Resiquimod (R848) | InvivoGen | tlrl-r848 | TLR7/8 agonist |
RPMI-1640 Medium | Sigma-Aldrich | R8758 | |
SPL Cell Culture Dish, 150 x 25 mm (SPL150) | Dutscher | 330009 | |
Surgical blade sterile N°23 | Dutscher | 132523 | |
UltraComp eBeads Compensation Beads | Thermo Fisher | 01-2222-41 | |
UltraPure 0.5M EDTA, pH 8.0 | Thermo Fisher | 15575020 | To make wash buffer in PBS |