Characterization Of Immune Cell Populations Research Articles

Characterization of Immune Cell Populations and Acid-Sensing Receptors in the Human Esophagus

Characterization of Immune Cell Populations and Acid-Sensing Receptors in the Human Esophagus

Abstract 6333: A high-parameter mass cytometry panel for the functional characterization of CAR T cells

Abstract High-parameter flow cytometry provides a powerful tool for in-depth analysis of CAR T cell therapy, from characterizing CAR T cell populations to understanding their interactions within the tumor microenvironment. This information is crucial for advancing CAR T cell therapies and improving their effectiveness in treating cancer. To address the need for deep immunoprofiling at all stages of CAR T cell therapy, we created a robust and comprehensive mass cytometry panel of antibodies against 44 cell surface and intracellular markers. This assay allows for comprehensive characterization of CAR T cell features including surface marker expression, activation state, cytokine production, and differentiation status. These data will provide researchers with a better understanding of the phenotype of the CAR T cells used in therapy and can guide optimization efforts to enhance expansion, persistence, and self-renewal. For example, newer-generation CAR T cells (T cells redirected for antigen-unrestricted cytokine-initiated killing, or TRUCKs) that are engineered to produce cytokines such as IL-2 to enhance their own survival can be evaluated in vitro to select CAR constructs with the most potent anti-tumor activity. The lyophilized and validated 30-marker Maxpar® Direct™ Immune Profiling Assay™ enables comprehensive characterization of immune cell populations in both whole blood and PBMC. The 7-marker Maxpar Direct T Cell Expansion Panel 3 (OX40, TIGIT, CD69, PD-1, Tim-3, ICOS, and 4-1BB) and the Maxpar Direct Basic Activation Expansion Panel (CD107a, IL-2, TNFα, IFNγ, perforin, and granzyme B) are add-on modules that can further characterize CAR T cell exhaustion, degranulation, cytokine production, and cytotoxicity against target cells. We also compared CAR T detection options for sensitivity and specificity using a CD19 CAR-transduced cell line spiked into healthy donor PBMC. Indirect staining with Miltenyi Biotec biotinylated CD19 CAR Detection Reagent followed by anti-biotin- or streptavidin-conjugated metal tags was compared against a directly conjugated or biotinylated anti-G4S linker antibody. We included sample multiplexing using 6 metal-tagged CD45 antibodies to address the need to minimize batch effects, crucial to multi-site and longitudinal studies. Samples were acquired on a CyTOF® XT™ instrument, and preliminary data analysis was performed using Maxpar Pathsetter™ for automated enumeration of immune cells. By combining the lyophilized single-tube Maxpar Immune Profiling Assay and Expansion Panels with specific detection of CAR T cells, this assay enables researchers to analyze multiple key parameters simultaneously at the single-cell level at all stages of CAR T therapy development. For Research Use Only. Not for use in diagnostic procedures. Citation Format: Deeqa Mahamed, Geneve Awong, Thiru Selvanatham. A high-parameter mass cytometry panel for the functional characterization of CAR T cells [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6333.

Characterization of Immune Cell Populations of Cutaneous Neurofibromas in Neurofibromatosis 1

Cutaneous neurofibromas (cNFs) are characteristic of neurofibromatosis 1 (NF1), yet their immune microenvironment is incompletely known. A total of 61 cNFs from 10 patients with NF1 were immunolabeled for different types of T cells and macrophages, and the cell densities were correlated with clinical characteristics. Eight cNFs and their overlying skin were analyzed for T cell receptor CDR domain sequences, and mass spectrometry of 15 cNFs and the overlying skin was performed to study immune-related processes. Intratumoral T cells were detected in all cNFs. Tumors from individuals younger than the median age of the study participants (33 years), growing tumors, and tumors smaller than the data set median showed increased T cell density. Most samples displayed intratumoral or peritumoral aggregations of CD3-positive cells. T cell receptor sequencing demonstrated that the skin and cNFs host distinct T cell populations, whereas no dominant cNF-specific T cell clones were detected. Unique T cell clones were fewer in cNFs than in skin, and mass spectrometry suggested lower expression of proteins related to T cell-mediated immunity in cNFs than in skin. CD163-positive cells, suggestive of M2 macrophages, were abundant in cNFs. Human cNFs have substantial T cell and macrophage populations that may be tumor-specific.

Single-Cell RNA Sequencing Reveals Immuno-Oncology Characteristics of Tumor-Infiltrating T Lymphocytes in Photodynamic Therapy-Treated Colorectal Cancer Mouse Model.

Photodynamic therapy (PDT) has shown promise in reducing metastatic colorectal cancer (CRC); however, the underlying mechanisms remain unclear. Modulating tumor-infiltrating immune cells by PDT may be achieved, which requires the characterization of immune cell populations in the tumor microenvironment by single-cell RNA sequencing (scRNA-seq). Here, we determined the effect of Chlorin e6 (Ce6)-mediated PDT on tumor-infiltrating T cells using scRNA-seq analysis. We used a humanized programmed death-1/programmed death ligand 1 (PD-1/PD-L1) MC38 cell allograft mouse model, considering its potential as an immunogenic cancer model and in combination with PD-1/PD-L1 immune checkpoint blockade. PDT treatment significantly reduced tumor growth in mice containing hPD-1/PD-L1 MC38 tumors. scRNA-seq analysis revealed that the PDT group had increased levels of CD8+ activated T cells and CD8+ cytotoxic T cells, but decreased levels of exhausted CD8+ T cells. PDT treatment also enhanced the infiltration of CD8+ T cells into tumors and increased the production of key effector molecules, including granzyme B and perforin 1. These findings provide insight into immune-therapeutic modulation for CRC patients and highlight the potential of PDT in overcoming immune evasion and enhancing antitumor immunity.

Characterization of immune cell populations in syngeneic murine tumor models.

Immunocompetent murine models are important tools for preclinical evaluation of immunotherapies. Here, six different immunocompetent tumor models based on four different cell lines were characterized, including metastatic lung cancer (CMT 167), triple-negative breast cancer (4T1), pancreatic cancer (KPCY), and colon cancer (MC38). The tumors were implanted subcutaneously or orthotopically before the animals were treated with anti-PD1 checkpoint inhibitor. A range of innate and adaptive immune cells were then quantified by flow cytometry of single-cell suspensions from the tumors. Furthermore, confocal laser scanning microscopy was used to quantify the density and distribution of T-cells in frozen sections. A model-dependent cellular immune landscape was observed, with variable responsiveness toward anti-PD1, ranging from the most responsive MC38 colon cancer model to the least responsive 4T1 breast cancer model. The study provides an overview of the immune landscape of these tumor models, and a foundation for further elucidation of pro-tumor and anti-tumor mechanisms behind heterogeneous responses towards immunotherapies.

Automated Gating and Interpretation of Clinical Flow Cytometry Data: A Computational Approach using Artificial Intelligence and Deep Learning

Abstract Flow cytometry (FCM) allows pathologists to accurately immunophenotype hematopoietic cells by detecting the expression of surface proteins with fluorochrome-conjugated antibodies. Accordingly, FCM plays a pivotal role in the diagnostic workup of many hematologic malignancies. However, post-analytic processing and the interpretation of FCM data are primarily manual processes that impact the consistency and limit the throughput of the method. The post-analytic processing workflow is colloquially referred to as ‘gating’ and involves the identification and characterization of immune cell populations by hand through successions of biaxial plots. The gated FCM data is then used by the pathologist to make a clinical interpretation and to then determine potential diagnoses from the patterns and cell frequencies seen in bivariate plots of the data. Since gating and interpretation are manually performed, post-analytic analyses of FCM data are not only laborious from a workflow perspective but remain subjective and prone to variability based on the experience and skill of both the medical laboratory scientist and pathologist. The objective of this study was to develop a computational pipeline that leverages machine learning-based (ML) solutions to automate gating and clinical interpretation of FCM data to increase the throughput and improve the repeatability of FCM analysis. Raw FCS files from clinical samples being evaluated for the presence of T-cell lymphoproliferative disease were exported from the on-instrument database. Automated gating was performed using open-source, supervised-ML packages for flow cytometry data (flowCore and flowDensity; R). These packages implement all processing steps that would typically be done manually (e.g. applying compensation, quality control (QC), and gating). Plots of interest were then generated from the gated data and classified as normal or abnormal using the clinical interpretations that were applied during the normal clinical workflow. These binary labels were used to train an ML-based classifier (VGG-19; Python). To evaluate pipeline performance, we collected FCS files from 1,188 samples that were analyzed by our flow cytometry lab. The automated gating pipeline was used to gate for CD4+CD3+ cells and to create bivariate plots for CD7/CD26 expression. 1150 (96.8%) passed QC and were all gated correctly by visual inspection. Of the 38 (3.2%) samples that failed QC, 13 (1.1%) were lymphopenic and were gated correctly by visual inspection, and the remaining 25 (2.1%) were gated manually. Using the CD7/CD26 plots, the classifier demonstrated promising predictive performance and achieved a precision of 0.85 and recall of 0.83 (weighted average across classes). Our findings represent a novel effort to automate both the gating and interpretation of FCM data using artificial intelligence. These results suggest that ML-based tools have potential utility in aiding the processing and interpretation of FCM data and can augment the efficiency and consistency of workflows in the clinical flow cytometry laboratory.

48P T-helper cell-driven immune response as an effect for seroma formation (SF) after mastectomy (ME) in breast cancer (BC) (SerMa pilot EUBREAST 5)

Seroma formation (SF) is one of the most frequent complications after breast cancer (BC) surgery. It occurs most frequently in patients after mastectomy (ME). The aim of this trial was to study (auto)immunological processes as an effect for SF. Identification and characterization of immune cell populations as well as specific immune reactions found in seroma and peripheral blood in patients after ME were therefore the main points of interest. Sera of healthy controls (without seroma and breast cancer) were chosen for comparison.

Technological advancements in multiomic single cell immune profiling and analysis

Abstract Continued advances in single cell gene detection sensitivity and the ability to analyze data quickly with biological context are critical for discoveries in therapeutic research. The second version of the Chromium Single Cell Immune Profiling Solution by 10x Genomics enables highly sensitive detection of gene expression, full-length paired T-cell alpha- and beta-chain and immunoglobulin sequences, T-cell antigen specificity, and cell surface protein expression from the same single cells, allowing a comprehensive view of the immune response at the cellular level. The new workflow provides a 45% increase in the number of genes detected per cell and up to a 25% increase in the cells detected with paired full-length V(D)J receptor sequences in melanoma tumor derived cells. Using a new version of Cell Ranger (v5.0), clonotypes were grouped from >30,000 cells, increasing the power to detect small clonotype expansions with fewer total cells. In addition, the new software enabled comparison of pre- and post-influenza vaccination B-cell receptor sequences from a single donor, identifying post-vaccination specific clonotypes. These technological and informatics advancements enhance a researcher’s ability to perform a broad characterization of immune cell populations at unprecedented throughput and resolution.

Myeloid arginase-1 controls excessive inflammation and modulates T cell responses in Pseudomonas aeruginosa pneumonia

Regulatory properties of macrophages associated with alternative activation serve to limit the exaggerated inflammatory response during pneumonia caused by Pseudomonas aeruginosa infection. Arginase-1 is an important effector of these macrophages believed to play an essential role in decreasing injury and promoting repair. We investigated the role of arginase-1 in the control of inflammatory immune responses to P. aeruginosa pneumonia in mice that exhibit different immunologic phenotypes. C57BL/6 mice with conditional knockout of the arginase-1 (Arg1) gene from myeloid cells (Arg1ΔM) or BALB/c mice treated with small molecule inhibitors of arginase were infected intratracheally with P. aeruginosa. Weight loss, mortality, bacterial clearance, and lung injury were assessed and compared, as were the characterization of immune cell populations over time post-infection. Myeloid arginase-1 deletion resulted in greater morbidity along with more severe inflammatory responses compared to littermate control mice. Arg1ΔM mice had greater numbers of neutrophils, macrophages, and lymphocytes in their airways and lymph nodes compared to littermate controls. Additionally, Arg1ΔM mice recovered from inflammatory lung injury at a significantly slower rate. Conversely, treatment of BALB/c mice with the arginase inhibitor S-(2-boronoethyl)-l-cysteine hydrochloride (BEC) did not change morbidity as defined by weight loss, but mice at day 10 post-infection treated with BEC had gained significantly more weight back than controls. Neutrophil and macrophage infiltration were similar between groups in the lung parenchyma, and neutrophil migration into the airways was reduced by BEC treatment. Differences seem to lie in the impact on T cell subset disposition. Arg1ΔM mice had increased total CD4+ T cell expansion in the lymph nodes, and increased T cell activation, IFNγ production, and IL-17 production in the lymph nodes, lung interstitium, and airways, while treatment with BEC had no impact on T cell activation or IL-17 production, but reduced the number of T cells producing IFNγ in the lungs. Lung injury scores were increased in the Arg1ΔM mice, but no differences were observed in the mice treated with pharmacologic arginase inhibitors. Overall, myeloid arginase production was demonstrated to be essential for control of damaging inflammatory responses associated with P. aeruginosa pneumonia in C57BL/6 mice, in contrast to a protective effect in the Th2-dominant BALB/c mice when arginase activity is globally inhibited.

Abstract 4551: Expression profiling-based characterization of immune cell populations in pediatric brain cancers

Abstract Cancer treatment strategies targeting the tumor microenvironment and corresponding immune cell infiltrate (i.e., immunotherapy) have recently proven effective against many solid tumors. In pediatric brain cancers, there is little known about the immune tumor microenvironment. We aim to characterize immune cell populations of pediatric brain tumors, which will better inform us about immune cell infiltration in different brain cancer subtypes and may indicate potential for patient response to immunotherapy. The Institute for Genomic Medicine at Nationwide Children’s Hospital has recently initiated a translational protocol to evaluate tumor specimens from pediatric patients with rare or refractory brain cancers in a focused N-of-1 manner. We performed RNA-seq and NanoString™ PanCancer Immune expression profiling on extracted tumor RNA from 22 patients comprising eight distinct brain cancer subtypes. We used CIBERSORT algorithm for deconvolution of RNA-seq global gene expression data and NanoString™ expression profiling to investigate methods of quantifying tumor-infiltrating leukocytes (TIL) and expression of immune genes. Our analyses identified a subset of pediatric brain tumors with evidence of immune infiltration. Using CIBERSORT, four gliomas were predicted to harbor significant TIL populations. Overall, we observed high correlation (Pearson R2 = 0.884) between expression values derived from RNA-seq and NanoString™ methods. Similarly, the two methods produced comparable predictions for TIL composition in the glioma tumors. Further genomic analysis identified one patient, with a diagnosis of glioblastoma and Neurofibromatosis type 1 (NF1), with marked infiltration of macrophages and neutrophils. Overexpression of tumor-associated macrophage/microglia genes (CD14, CD163, CD68, ITGAM, PTPRC) and other immune markers (PD-L1, CCL2/MCP-1, IL1B) was observed relative to all other samples. More specifically, overexpression of CD204, CD206, and CD163 suggested the infiltrating macrophages represent the M2/protumor phenotype. CIBERSORT allows more “granularity” in macrophage population prediction and indicated an equal representation of M2 and M0/uncommitted macrophages. Given the M2 phenotype, it is likely that the infiltrating macrophages contributed to tumor progression and radiation resistance observed in this patient. If confirmed, the presence of M0 macrophages and/or overexpression of PD-L1 could indicate potential for immunotherapy in this patient (e.g., oncolytic viral therapy, strategies focusing on M0 to M1/antitumor polarization, or combination targeted therapy with PD-L1 inhibitor). In summary, we identified a patient with progressive glioblastoma tumor growth and significant macrophage tumor infiltration as a likely candidate for immunotherapy. Citation Format: Katherine E. Miller, Kathleen Schieffer, James Fitch, Vincent Magrini, Amy Wetzel, Anthony R. Miller, Daniel R. Boue, Jeffrey Leonard, Jonathan L. Finlay, Diana S. Osorio, Mohamed S. AbdelBaki, Christopher R. Pierson, Annie Drapeau, Jonathan Pindrik, Kristen Leraas, Elizabeth Varga, Devon Dishman, Lauren Shoemaker, Nicole Ross, Jeremy Pitts, Julie Gastier-Foster, Peter White, Catherine E. Cottrell, Richard K. Wilson, Elaine R. Mardis. Expression profiling-based characterization of immune cell populations in pediatric brain cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4551.

Optimizing checkpoint inhibitors therapy for relapsed or progressive classic Hodgkin lymphoma by multiplex immunohistochemistry of the tumor microenvironment.

Immune checkpoint‐blocking antibodies have therapeutic activity against relapsed or progressive classic Hodgkin lymphoma (cHL), but Hodgkin Reed‐Sternberg cells can develop resistance to this therapy via multiple mechanisms. To improve the efficacy of immune checkpoint blockade, we need a more precise understanding of the immune escape mechanisms active in individual cHL patients, and this requires a detailed characterization of immune cell populations in the tumor microenvironment. These cell‐cell interactions can now be studied by multiplex immunohistochemistry coupled to digital image analysis. This method should allow the identification of actionable target molecules mediating resistance to immune checkpoint inhibitors in individual cHL patients, thereby favoring the implementation of personalized therapies.

Abstract B83: Intratumoral T cell receptor β repertoire as a biomarker in head and neck squamous cell carcinoma

Abstract In this study, we examine the T cell receptor (TCR) repertoire in a cohort of Head and Neck Squamous Cell Carcinoma (HNSCC) patients (n=50) at time of diagnosis. The TCR repertoire is indicative of the ability of the body to mount an immune response, with increased productive clonality within the tumor environment suggesting accumulation of tumor antigen specific T cells. Recent studies demonstrate that the intratumoral TCR repertoire is prognostic of response to anti-PD-1 (Tumeh et al., 2014) and anti-CTLA-4 (Page et al., 2016) therapies. Here we examine diversity and clonality of intratumoral T cell β repertoires in HNSCC patients independently and in conjunction with other biomarkers including tumor mutational burden/neoantigen load, expression of inhibitory checkpoints, and characterization of immune cell populations, as well as clinical metadata including HPV status, smoking history, and survival outcomes. The TCRβ repertoire was analyzed from FFPE primary tumor samples through targeted next generation DNA sequencing of the TCRβ loci and examination of the CDR3 sequences to determine clonality of the TCRβ repertoire (1 – Pielou’s evenness). Somatic genomic alterations, tumor mutational burden, and HLA typing were conducted through whole exome sequencing. Inhibitory checkpoint expression and characterization of immune cell populations were conducted through immunohistochemistry, RNA expression, or both methods. Preliminary results indicate productive clonality of the TCR repertoire in the range of 0.03-0.16. In examining HPV+ and HPV- patients, where we hypothesize the presentation of foreign peptides of viral etiology will produce an increase in T cell populations recognizing those neoantigens, we see no difference in either T cell infiltration or productive clonality. However, we observe that PD-L1 status is significantly correlated (p < 0.05) to productive clonality of the tumor infiltrating T cells, with a higher average productive clonality in PD-L1+ patients. Ongoing studies will characterize the landscape of somatic genomic alterations including tumor mutational burden, and gene expression profiles of the tumor microenvironment. Citation Format: Lara McGrath, Elizabeth Maloney, Maria Secrier, Helen Angell, Emma Jones, Brian Dougherty, Daniel Stetson, Patricia McCoon, Andrew Pierce, Craig Barker, Carl Barrett. Intratumoral T cell receptor β repertoire as a biomarker in head and neck squamous cell carcinoma [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2017 Oct 1-4; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2018;6(9 Suppl):Abstract nr B83.

Predicting NK cell subsets using gene expression levels in peripheral blood and endometrial biopsy specimens.

In molecular analysis of tissue biopsy specimens, one crucial aspect is characterization of immune cell populations. This is especially important for evaluation of uterine receptivity by assessing levels of lymphocyte populations including CD56bright CD16- uterine NK cells and CD56dim CD16+ conventional NK cells. Our objective was to investigate whether measuring total RNA transcripts from a tissue specimen would accurately reflect immune cell levels and be a new technique to assess immune cell subsets. Peripheral blood mononuclear cells (PBMCs) and endometrial tissues were used. Flow cytometry was utilized for the analysis of lymphocyte subsets in PBMCs, and RT-qPCR was applied to quantify RNA transcripts indicative of lymphocyte and granulocyte populations. In PBMC specimens, there were significant correlations between gene expression levels and cell subsets. NK cells correlated with CD16A, NKp46, and CD56 transcripts, B cells correlated with EBF1, and CD8+ T cells correlated with CD8β. Finally, endometrial tissues displayed high CD56 expression and very low CD3ε, CD16A, and NKp30, reflecting the characteristic endometrial NK cell subsets. Strong correlations between RT-qPCR data and levels of lymphocyte subsets indicate that gene expression analysis will be a useful technique for characterizing levels of CD56+ cells in endometrial tissues.

Semi-automated and standardized cytometric procedures for multi-panel and multi-parametric whole blood immunophenotyping

Immunophenotyping by multi-parametric flow cytometry is the cornerstone technology for enumeration and characterization of immune cell populations in health and disease. Standardized procedures are essential to allow for inter-individual comparisons in the context of population based or clinical studies. Herein we report the approach taken by the Milieu Intérieur Consortium, highlighting the standardized and automated procedures used for immunophenotyping of human whole blood samples. We optimized eight-color antibody panels and procedures for staining and lysis of whole blood samples, and implemented pre-analytic steps with a semi-automated workflow using a robotic system. We report on four panels that were designed to enumerate and phenotype major immune cell populations (PMN, T, B, NK cells, monocytes and DC). This work establishes a foundation for defining reference values in healthy donors. Our approach provides robust protocols for affordable, semi-automated eight-color cytometric immunophenotyping that can be used in population-based studies and clinical trial settings.

Role of NK cells in host defense against pulmonary type A Francisella tularensis infection

Pneumonic tularemia is a potentially fatal disease caused by the Category A bioterrorism agent Francisella tularensis. Understanding the pulmonary immune response to this bacterium is necessary for developing effective vaccines and therapeutics. In this study, characterization of immune cell populations in the lungs of mice infected with the type A strain Schu S4 revealed a significant loss in natural killer (NK) cells over time. Since this decline in NK cells correlated with morbidity and mortality, we hypothesized these cells contribute to host defense against Schu S4 infection. Depletion of NK cells prior to Schu S4 challenge significantly reduced IFN-γ and granzyme B in the lung but had no effect on bacterial burden or disease progression. Conversely, increasing NK cell numbers with the anti-apoptotic cytokine IL-15 and soluble receptor IL-15Rα had no significant impact on Schu S4 growth in vivo. A modest decrease in median time to death, however, was observed in live vaccine strain (LVS)-vaccinated mice depleted of NK1.1+ cells and challenged with Schu S4. Therefore, NK cells do not appear to contribute to host defense against acute respiratory infection with type A F. tularensis in vivo, but they play a minor role in protection elicited by LVS vaccination.

Characterization of Immune Cell Populations in Oral Mucosal Tissue of Healthy Adult Cats

The aim of this study was to characterize the leucocyte subsets present in the oral mucosa of healthy cats. Immunohistochemical labelling and computer-assisted morphometric analysis was used to identify expression of MHC class II, CD3, CD79a, IgG, IgM, IgA, and leucocyte antigen L1 (L1) by cells in sections from 19 cats, and expression of CD4 and CD8 by cells in sections from 17 cats. Mast cells were detected by toluidine blue staining. In the epithelial compartment, CD3 + intraepithelial lymphocytes were detected, and CD8 + cells were more common than CD4 + cells. MHC class II labelling revealed intraepithelial and subepithelial cells with a characteristic dendritic morphology. In some sections these dendritic cells were closely associated with subepithelial clusters of CD3 + T cells containing both CD4 + and CD8 + cells. In the lamina propria and submucosal compartments, the cells most commonly identified were mast cells. CD3 + T-lymphocytes were also observed, and CD4 + and CD8 + cells were detected in similar numbers. L1 + and CD79 + cells were detected least frequently. The few plasma cells present were generally found to be either IgG + or IgA +. Within the stroma surrounding the salivary glands, CD79a + and IgA + cells predominated. Slight epithelial labelling for L1 was seen in some sections. The normal feline oral mucosa clearly contains a range of immune cell populations.