Correlation Between Immune Lymphoid Cells and Plasmacytoid Dendritic Cells in Human Colon Cancer

doi:10.21203/rs.3.rs-46781/v1

Download PDF

Research

Correlation Between Immune Lymphoid Cells and Plasmacytoid Dendritic Cells in Human Colon Cancer

https://doi.org/10.21203/rs.3.rs-46781/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Innate lymphoid cells (ILCs), so far studied mostly in mouse models, are important tissue-resident innate immune cells that play important roles in the colorectal cancer microenvironment and maintain the mucosal tissue homeostasis. Plasmacytoid dendritic cells (pDCs) present complexity in various tumour types and are correlated with poor prognosis. pDCs can promote HIV-1–induced group 3 ILC (ILC3) depletion through the CD95 pathway. However, the role of ILC3s in human colon cancer and their correlation with other immune cells, especially pDCs, remain unclear.

Methods: We characterised ILCs and pDCs in the tumour microenvironment of 58 colon cancer patients by flow cytometry and selected three patients for RNA sequencing.

Results: ILC3s were negatively correlated, and pDCs were positively correlated, with cancer pathological grade. There was a negative correlation between the numbers of ILC3s and pDCs in tumour tissues. RNA sequencing confirmed the correlations between ILC3s and pDCs and highlighted the potential function of many ILC- and pDC-associated differentially expressed genes in the regulation of tumour immunity. pDCs can induce apoptosis of ILC3s through the CD95 pathway in the tumour microenvironment.

Conclusions: One of the interactions between ILC3s and pDCs is via the CD95 pathway, which may help explain the role of ILC3s in colon cancer.

Cancer Biology

ILC

pDC

colon cancer

RNA-Seq

Colon cancer is the third most commonly diagnosed cancer worldwide [1–3]; furthermore, its incidence has significantly increased recently [4] and is expected to further rise by 50% in the next five years [5]. Epidemiological studies show that the causes of colon cancer are related to environmental, lifestyle, and genetic factors, and that age, intestinal polyps, and ulcerative colitis also represent high-risk factors [6, 7]; however, the specific pathogenesis of colon cancer remains unclear.

Currently, treatment of primary colon cancer is mainly surgical; however, postoperatively there is still a risk of recurrence and metastases [8]. Therefore, it is important to fully understand the causes of colon cancer to promote the discovery of novel and effective therapeutic targets; this reflects an urgent clinical need from both a theoretical and a practical point of view.

With the rapid rate of discoveries in the field of immunology, cancer immunotherapy has attracted increasing attention [9]. The immune system plays an important role in the protection of the host against tumour onset (i.e., tumour immunosurveillance) [10]. In addition to tumour cells, stromal and immune cells are also present in the tumour microenvironment, where tumour cells often either recruit or locally induce their proliferation or differentiation to release an array of cytokines that participate in the immune response [10–12]. In colon cancer, T and B lymphocytes have been found in proximal colon tumour tissue [5], and natural killer (NK) cells, monocytes/macrophages, dendritic cells (DCs), mast cells, and neutrophils have been detected in the colon tumour microenvironment [12]. Notably, the differential distribution of these cells in the tumour microenvironment reflects the diversity of tumour biology [13].

Innate lymphoid cells (ILCs) are a characterised subset of innate lymphocytes [14] that includes three groups: Group 1 ILCs (ILC1s) consist of NK cells that express the transcription factor T-bet and secrete interferon (IFN)-γ; group 2 ILCs (ILC2s) express the transcription factor GATA-binding protein 3 and secrete interleukin (IL)-5 and IL-13; and group 3 ILCs (ILC3s) express the transcription factor RAR-related orphan receptor-γt and secrete IL-17 and IL-22 [15–17]. ILCs lack an antigen-specific receptor; however, they can still be activated by danger signals from injured mucosal tissue and quickly produce an array of effective cytokines to repel pathogens and tumour cells, thereby sustaining mucosal integrity [18, 19]. A previous study has suggested that ILCs may exert both pro- and antitumour functions depending on the phase of cancer and environmental context [20]. Until now, most studies on ILCs have focused on mouse models, and very few studies on human colon cancer. Recently, Salimi et al. investigated 13 patients with gastrointestinal (including oesophageal, gastric, colon, and rectal) tumours and found a significantly higher frequency of group 1 ILCs (p value: 0.001) in malignant gastrointestinal tumours than in benign tissues [21]. Ikeda et al. collected 28 samples from colon cancer patients and reported that the number of NKp44⁺ ILC3s from colorectal cancer tissue was decreased in T3/T4 tumours, with associated decreases in tertiary lymphoid structure induction [22]. In this study, we expanded the number of research samples to further study the distribution characteristics of ILCs in colon cancer and their correlation with other immune cells. We investigated the role of ILCs in the colon tumour microenvironment to identify potential strategies for the induction of antitumour immune responses in colon cancer.

Patient tissue specimens

Fresh tumour specimens, including those from tumour-proximal and distal regions, were collected from 58 patients with colon cancer who did not receive radiotherapy or chemotherapy prior to surgery at the Department of Gastric Colorectal Anal Surgery, First Affiliated Hospital, Jilin University (Changchun, China). Patient clinicopathological characteristics were determined according to the National Comprehensive Cancer Network (NCCN) guidelines for colon cancer (Version 2.2018). Proximal tissue was defined as a 2-cm to 5-cm zone bordering the tumour margin. Distal tissue was located > 5 cm from the tumour and was considered to be a normal tissue sample and used as a control. There were no restrictions on cancer subtype, age, or sex, and tumour types were identified by histological analysis.

Patients were divided based on TNM stage and tumour histological grade. The TNM staging system is based on the extent of the tumour (T), the extent of spread to the lymph nodes (N), and the presence of metastasis (M). Tumours were classified as stage I, II, III, or IV based on TNM stage and prognosis, where a higher number indicated a more advanced cancer and, likely, a worse outcome. Among them, stage III patients are the most common type. Additionally, the patients were classified as having either glandular or mucous carcinoma.

Tables 1 and 2 show the clinical characteristics of the patients included in this study. The percentage of ILC3s and pDCs in tumour versus distal tissue was separated into “high” and “low” categories according to ILC3 and pDC content, with “high” representing a ratio > 1, and “low” representing a ratio < 1.

Tissue digestion for single-cell suspension

Freshly resected colon tissues from patients with colon cancer and tonsil tissues from children with tonsillar hypertrophy were minced into small pieces in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) supplemented with 1% foetal calf serum (Lonza, Basel, Switzerland) and were sequentially digested with collagenase D (1 mg/ml; Sigma-Aldrich, St. Louis, MO, USA) and DNase I (50 µg/ml; Sigma-Aldrich) at 37 °C for 40 min and 30 min, respectively. The cell suspensions were then passed through 100-µm and 40-µm cell strainers (BD Biosciences, Franklin Lakes, NJ, USA) to remove debris. The cell suspensions were resuspended in RPMI 1640 medium (Invitrogen) supplemented with 10% foetal calf serum (FCS; Lonza) and 1% penicillin/streptomycin (Sigma-Aldrich) before isolation of mononuclear cells (MNCs) by centrifugation over a Ficoll–Hypaque density gradient centrifugation for 30 minutes at 24 °C for further analysis.

Isolation of ILC3s and pDCs

For further purification, MNCs from freshly resected patient tissue specimens were subjected to Ficoll-Hypaque gradient centrifugation for 30 min at 24 °C. Next, the MNC layer was transferred to a new tube, washed twice with phosphate-buffered saline (PBS), and suspended in PBS. ILCs were sorted using a BD FACSAria system (BD Bioscience) as Lin⁻-enriched MNCs as Lin cocktail⁻ (CD3, CD19, CD20, and CD14), CD94⁻ CD34⁻ CD1a⁻ TCRα/β⁻ TCRγ/δ⁻ CD45⁺ CD127⁺ CRTH2^+/− CD117^+/− cells using FITC anti-Lin (643510; BD Bioscience), CD94 (305504; Biolegend, San Diego, CA, USA), CD34 (343504; Biolegend), CD1a (300104; Biolegend), T cell receptor (TCR)α/β (306706; Biolegend), TCRγ/δ (331208; Biolegend), allophycocyanin (APC)-H7 anti-CD45 (56017; Biolegend), Percp-cy5.5 anti-CD127 (351322; Biolegend), phycoerythrin (PE)-Cy7 anti-CRTH2 (350118; Biolegend), and BV605 anti-CD117 (562687; Biolegend). pDCs were sorted as Lin⁻ CD94⁻ CD34⁻ CD1a⁻ TCRα/β⁻ TCRγ/δ⁻ CD45⁺ BDCA2⁺ cells using FITC anti-Lin, CD94, CD34, CD1a, TCRα/β, TCRγ/δ, APC-H7 anti-CD45, and APC anti-BDCA2 (17-9818-42; Biolegend). Purity was routinely > 99%. Cell viability was determined by trypan blue staining and was > 99% after isolation.

Flow cytometric analysis

ILCs and pDCs were identified as described. ILC3s were further divided into NKp44⁺ ILC3s, and NKp44⁻ ILC3s were identified as Lin⁻ CD94⁻ CD34⁻ CD1a⁻ TCRα/β⁻ TCRγ/δ⁻ CD45⁺ CD127⁺ CRTH2⁻ CD117⁺ NKp44^+/− cells using AF647 anti-NKp44 (558564; BD Bioscience). Myeloid DCs (mDCs) were identified as Lin⁻ CD45⁺ CD11b⁺ CD11c⁺ cells using FITC anti-Lin, APC-H7 anti-CD45, PE-conjugated anti-CD11b (555388; BD Bioscience), and AF700 anti-CD11c (561352; BD Bioscience). Treg cells were identified as CD45⁺ CD4⁺ CD25⁺ forkhead box (Fox) P3⁺ cells utilising APC-H7 anti-CD45, PE-Cy7 anti-CD4 (557852; BD Bioscience), FITC anti-CD25 (555431; BD Bioscience), and Percp-cy5.5 anti-FoxP3 (561493; BD Bioscience). B regulatory (Breg) cells were identified as CD45⁺ CD19⁺ CD24⁺ CD38⁺ cells using APC-H7 anti-CD45, APC anti-CD19 (555415; BD Bioscience), BV605 anti-CD24 (562788; BD Bioscience), and PerCP-Cy5.5 anti-CD38 (551400; BD Bioscience). T cells were identified as CD45⁺ CD3⁺ cells utilising APC-H7 anti-CD45 and PerCP-Cy5.5 anti-CD3 (560835; BD Bioscience). B cells were identified as CD45⁺ CD19⁺ cells using APC-H7 anti-CD45 and APC anti-CD19. NK cells were identified as CD45⁺ CD56⁺ cells using APC-H7 anti-CD45 and AF700 anti-CD56 (557919; BD Bioscience). Monocytes were identified as CD45⁺ CD14⁺ cells utilising APC-H7 anti-CD45 and FITC anti-CD14 (555397; BD Bioscience). pDCs from tissue and blood were stained with the following monoclonal antibodies: PE-conjugated anti-HLA-DR, PE-Cy7 anti-CD86, Percp-cy5.5 anti-OX40L, and APC anti-PD-L1 (BD PharMingen, San Jose, CA, USA). Viability was assessed with an Aqua system (BD PharMingen).

For surface marker staining, MNCs were incubated with antibodies at 4 °C for 30 minutes and then washed twice before flow cytometric analysis. For the staining of apoptotic marker active caspase-3, cells were stained with a surface marker first and then permeabilised using a Cytofix/Cytoperm kit (BD Bioscience) and stained for intracellular protein caspase-3 with PE-conjugated anti-caspase 3 monoclonal antibodies (BD PharMingen). Fluorescence-associated cell sorting (FACS) plots depict the mean fluorescence intensity values of Ab staining after subtracting the mean fluorescence intensity of the respective isotype control (BD Bioscience).

Immunohistochemistry

Standard H&E staining was used for colon tissue localisation. Paraffin-embedded, 4-µm-thick tumour sections and tumour-proximal and distal tissue specimens from five patients with colon cancer were selected for immunohistochemistry analysis. Tissue sections were dewaxed and subjected to heat-induced epitope retrieval with preheated antigen-retrieval buffer (pH 9.0; Dako; Agilent Technologies, Santa Clara, CA, USA). Endogenous peroxidase activity was then blocked, and the sections were incubated overnight at 4 °C with anti-human BDCA2 (10 µg/ml; clone 124B3.13; Dendritics, Lyon, France). After incubation with a horseradish peroxidase-conjugated secondary antibody (Invitrogen) and development with diaminobenzidine, sections were counterstained with haematoxylin. PBS was used in place of the primary antibody for the negative controls. Images of tissue slides were acquired with a light microscope (BX51N-34-FL-1-D; Olympus, Tokyo, Japan) and analysed with CellSens Dimension software (Universal Imaging, Bedford Hills, NY, USA).

RNA-Seq and analysis

We used freshly sorted ILC3s and pDCs from tumours, proximal and distal regions, and peripheral blood from three patients with stage III colon cancer for RNA-Seq analysis. A total of 200 sorted cells (ILCs or pDCs) were utilised. Cells were sorted into an Eppendorf tube containing 4 µl of lysis buffer (Beijing Genomics Institute, Shenzhen, China) and quickly transferred to liquid nitrogen. RNA-Seq analysis was performed by the Beijing Genomics Institute. The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus [23] and are accessible through GEO Series accession number GSE127934 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc =GSE127934).

Preparation of colon tumour-derived supernatant

Single-cell suspensions from colon tumours were incubated at a final concentration of 2.5 × 10⁶ cells/ml in complete RPMI in a 6-well tissue culture plate. Tumour supernatant (TS) was collected after 24 hours, filtered at 0.2 mm, and frozen at -80 ℃ until use.

Co-culture of ILC3s and pDCs

ILC3s and pDCs from normal tonsil tissue were prepared and cultured separately at 1 × 10⁶ cells/ml in complete RPMI (RPMI 1640 containing 10% heat-inactivated foetal bovine serum, 100 U/ml penicillin, 100 mg/ml streptomycin sulphate, 100 U/ml IL-2, and 50 ng/ml IL-7, Cellgro) in the presence or absence of 25% TS, IFNα (1000 IU/ml, Millipore), and anti-IFNα (10 ug/ml, Millipore) for 72 hours. Cells and culture supernatant were then harvested for subsequent experiments. Flow cytometric analysis was used for the expression of apoptosis-related genes and the survival rate of ILC3s. ELISA was used to detect the secretion of IL-22 in co-culture supernatant. Giemsa staining was used to detect the morphology of ILC3s and pDCs.

ELISA

ELISA kits for hIFN-α and hIL-22 (R&D Systems, USA) were used according to the manufacturer's instructions. For hIFN-α, pDCs were cultured at 2.5 × 10⁶ cells/ml with TS, TLR7 ligand IMQ (1.5 µM, Invivogen) or anti-IFNα (10 ug/ml, Millipore) in RPMI medium supplemented with 10% FCS, 1% Pen/Strep, nonessential amino acids, sodium pyruvate, and β-mercaptoethanol. Supernatants were collected after 2 days and analysed with ELISA. For hIL-22, culture supernatant was collected from co-cultured ILC3s and pDCs. All ELISA results are expressed in pg/ml.

Giemsa staining

For Giemsa staining, ILC3s and pDCs were seeded on glass coverslips and co-cultured in the presence of 25% TS for 72 hours. Coverslips were air-dried, fixed in methanol, and stained with modified Giemsa stain GS500 (Sigma Diagnostics, USA). Each slide specimen was observed under a light microscope (BX51N-34-FL-1-D, Olympus Corporation, Tokyo, Japan).

Statistics

Independent samples were compared using Student's t-test. Correlation analysis was performed using the Spearman test. The following clinical variables were collected prospectively: age, sex, region, and tumour stage. All statistical analyses were performed using GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA). P < 0.05 was considered statistically significant.

Number of ILC3s in colon cancer tissue specimens is negatively correlated with disease progression

In our study, following collection of tissue samples from 58 patients with colon cancer, flow cytometric analysis of MNCs isolated from the tissue showed that nearly 1% of CD45⁺ colon lymphocytes exhibited an ILC phenotype (Lin⁻CD127⁺) (Fig. 1A). Further subtyping revealed that these ILCs comprised chemoattractant receptor-homologous molecules expressed on Th2 cells (CRTH2) ⁺ ILC2s, CRTH2⁻ CD117⁺ NKp44^+/− ILC3s, and CRTH2⁻ CD117⁻ NKp44⁻ ILC1s (Fig. 1A). Additionally, we found that the percentage of total ILCs among the CD45⁺ lymphocytes in the tumour tissue was lower than that in regions proximal and distal to the tumour (Fig. 1B), with the ILC3 percentage lower than that in the proximal and distal regions (Fig. 1C) and higher in the proximal region than in the distal region (Fig. 1C). However, there was no difference in the percentage of ILC1s and ILC2s in the investigated tissue regions (Fig. 1C). These findings are not consistent with the results of Salimi et al. [21], but consistent with the results of Ikeda et al., which may be due to differences in the investigated tissue samples. Additionally, variations in NKp44^+/− ILC3 levels, especially those of NKp44⁺ ILC3s, among the CD45⁺ lymphocytes followed a similar pattern (Fig. 1D). Because ILC3s can also be classified as CCR6^+/−, NKp30^+/−, and NKp46^+/− [24], we investigated other subtypes and observed no significant differences in the percentages of the other ILC3 subtypes in the examined regions (Figure S1A).

A major prognostic factor for the survival of patients with colon cancer is the pathological tumour grade; therefore, we analysed possible correlations between ILC3s or NKp44^+/− ILC3s and the pathological cancer grade, and observed a negative correlation with ILC3s, especially NKp44⁺ ILC3s, and grade (Fig. 1E and F; Table 1). These data were consistent with previously reported results and showed that natural cytotoxicity receptor (NCR)⁺ ILC3s are more prevalent in stage I/II non-small cell lung cancer than in more advanced-stage tumours, and that they contribute to the formation of protective tumour-associated tertiary lymphoid structures [24]. However, in the present study, we found no correlation between NKp44⁻ ILC3s and the pathological grade (Fig. 1F).

Su et al. [25] reported the clinical significance of circulating immune cells at different colon tumour locations. In the present study, our results showed no correlation between ILC3 percentage and the different tumour regions (Figure S1B and Table 1). Glandular and mucous carcinomas are common forms of colon cancer. We found similar variations in ILC3 levels among CD45⁺ lymphocytes in colon glandular cancer and mucous carcinoma with the percentage of tumour ILC3s lower than that of proximal and distal ILC3s and the percentage of proximal ILC3s higher than that of distal ILC3s (Fig. 1G).

Decreased numbers of ILC3s in colon cancer tissues are correlated with higher pDC levels

pDCs have clinical importance in different tumour types [26] and play a critical role in the tumour microenvironment to promote cancer progression through stimulation of Th2 and regulatory immunity [27]. Su et al. [28] showed that the percentage of ILC3s was negatively correlated with pDC levels in lymphoid organs of NRG humanised mice with a persistent HIV-1 infection. Because ILC3 depletion by HIV-1 infection is dependent upon pDCs and IFN-I activity, we hypothesised that the frequency of the two cell types is also correlated in colon cancer. To test this hypothesis, we detected the incidence of pDCs in colon cancer tissues. Flow cytometric analysis of MNCs isolated from colon cancer tissue specimens showed that nearly 0.5% of CD45⁺ colon leukocytes exhibited a pDC phenotype (Lin⁻ CD45⁺ blood DC antigen 2, BDCA2⁺) (Fig. 2B) with the percentage of pDCs higher than that in proximal and distal regions (Fig. 2C); furthermore, the pDC percentage was higher in the proximal region than in the distal region (Fig. 2C). Immunohistochemistry staining of BDCA2 revealed that pDCs were present in colon tumour tissue specimens in similar proportions to those obtained by flow cytometry (tumour > proximal > distal; Fig. 2A).

Additionally, we found a positive correlation between the number of pDCs and pathological tumour grade (Fig. 2D and Table 2). Interestingly, there was also a positive correlation between the percentage of pDCs and the examined tumour region (Fig. 2E and Table 2). Moreover, the variations in pDC levels among CD45⁺ cells in colon glandular cancer tissue and mucous carcinoma were similar, as the percentage of tumour pDCs was higher than that of pDCs in the proximal and distal regions, and the percentage of proximal pDCs was higher than that of distal pDCs (Fig. 2F).

Next, we determined the correlation between ILC3 frequency and pDCs among tumour infiltrating CD45⁺ cells and observed a negative correlation between the number of ILC3s and pDCs in colon cancer tissues (Fig. 2G). We then detected the percentage of other immune cells in colon tissues (tumour, proximal, and distal) and found that the percentage of Treg cells among the CD45⁺ cells was higher than that in the proximal and distal regions. However, there was no difference in the percentages of mDCs, CD4⁺ T cells, CD8⁺ T cells, Breg cells, B cells, NK cells, or monocytes among the CD45⁺ cells in the investigated regions (Figure S2). Additionally, there was no correlation between the change in the number of ILC3s and that of other immune cells (data not shown).

Correlation between ILC3s and pDCs at the level of differentially expressed genes (DEGs)

To investigate gene expression in ILC3s and further assess the correlation between ILC3s and pDCs in colon cancer tissues, we performed RNA-Seq analysis using tumour-derived (T) ILC3s or pDCs and controls comprising distal (D) ILC3s or pDCs from three patients with stage III colon cancer.

For the ILC3s, > 60 million clean reads were obtained from each sample group with a Q20 score > 98% and a mapping rate to the reference genome of each sample varying from 73.89–91.40% (Table S1), indicating that the data were reliable and could be used for further analysis. A total of 14,943 and 4213 genes were upregulated and downregulated, respectively, in tumour ILC3s relative to distant ILC3s (Fig. 3A); among them, 7352 genes were related to cancer (Fig. 3B). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis confirmed a significant enrichment of genes involved in cancer-associated and RNA-degradation pathways (Fig. 3C), with more upregulated than downregulated genes. These results show that the tumour environment altered the expression of many ILC3 genes.

After removing genes which could not be confidently mapped to existing entries in any public sequence database, we calculated the log value and fragments per kilobase of transcript per million (FPKM) reads value for each sample. We identified tumour-related genes with significant differences, including 29 upregulated genes and 2 downregulated genes identified from the comparison of tumour and distal regions in four patients (Fig. 3D). The upregulated genes included those associated with tumour development (PBX3, ARID3B, NID2, PRR11, COL23A1, TGIF2, SEMA4A, COL23A1, and SLC25A29) and inhibition of tumour development (LTBP4, KANK2, RTEL1, ANGPTL4, and SCIN) (Fig. 3D), whereas the downregulated genes included one associated with tumour development (CSE1L) (Fig. 3D). These data suggest that ILC3s in tumour tissue might play dual roles during tumour development, in agreement with a previous report [20].

In addition, we analysed the expression of inflammatory factors and chemokines on ILC3s. The analysis showed that 22 inflammatory genes and 13 chemokines were significantly expressed on ILC3s (Fig. 3D). Of the inflammatory factors, 19 upregulated genes and three downregulated genes were identified from the comparison of tumour and distal regions in three patients (Fig. 3D). Of the chemokines, eight upregulated genes and five downregulated genes were identified from the comparison of tumour and distal regions in four patients (Fig. 3D).

For the pDCs, we obtained > 60 million clean reads with a Q20 score > 97% and a mapping rate to the reference genome of each sample varying from 75.74–88.42% (Table S2). A total of 10,840 and 11,549 genes were upregulated and downregulated, respectively, in tumour pDCs as compared with distal pDCs (Fig. 4A); among them, 6446 genes were related to cancer (Fig. 4B). KEGG pathway analysis confirmed the enrichment of cancer-associated genes (Fig. 4C), and as with the ILC3 results, the data show that the tumour environment altered the expression of multiple genes. We identified tumour-related genes with significant differences, including 14 upregulated and 10 downregulated genes from tumour versus distal pDCs in three patients (Fig. 4D). We found that the upregulated genes were associated with tumour development (ARHGAP4, HSPD1, HNRNPA2B1, UBAP2L, STAG1, TUBB, GPX2, CD44, PEBP4, and CD274) (Fig. 4D), and the downregulated genes were associated with tumour inhibition (SNAP23, PTPRE, RPS13, and OGT) (Fig. 4D). These findings were consistent with previously reported results, showing that pDCs in the tumour microenvironment are associated with the development and maintenance of immunosuppression [27, 29–31].

In addition, we analysed the expression of inflammatory factors and chemokines on pDCs. Nine inflammatory genes and one chemokine were significantly expressed on pDCs (Fig. 4D). Of the inflammatory factors, seven upregulated genes and two downregulated genes were identified from the comparison of tumour and distal regions in three patients (Fig. 4D). Of the chemokines, one downregulated gene was identified from the comparison of tumour and distal regions in three patients (Fig. 4D).

To assess correlations between ILC3s and pDCs in colon cancer tissue specimens, we analysed the RNA-Seq data and found that 3408 upregulated and 962 downregulated genes were co-expressed in ILC3s and pDCs (Fig. 5A and B). Moreover, calculation of the Pearson correlation coefficient from RNA-Seq data from all tumour samples revealed an obvious correlation between ILC3s and pDCs in each tissue sample (Fig. 5C). The analysis of DEGs in ILC3s (Fig. 3C) versus pDCs (Fig. 4C) revealed that most upregulated and downregulated ILC3 genes were associated with RNA degradation, metabolic, and apoptotic pathways, whereas most upregulated and downregulated pDC genes were associated with tumour development or inhibition (Fig. 5D) (Figure S3). In addition, some inflammatory factors and chemokines were highly expressed on ILCs or pDCs (Fig. 5D), particularly ILC3s. These findings were consistent with our flow cytometry results, which showed a negative correlation between the numbers of ILC3s and pDCs in colon cancer tissues (Fig. 2G).

pDCs can induce apoptosis of ILC3s in the tumour microenvironment

Our results showed a negative correlation between the numbers of ILC3s and pDCs in colon cancer tissues; the number was low for ILCs and high for pDCs (Fig. 2G). The RNA-Seq results showed that ILC3s showed high expression of apoptosis-related genes such as CD95, TNFRSF21, caspase 8, and caspase 3, and pDCs showed high expression of IFNα-related genes such as IFNA1, IFNAR2, and IRF6 in tumours (Fig. 5D). Su et al. [28] showed that ILC3 depletion by HIV-1 infection is dependent upon pDCs and IFN-I activity. Therefore, we tested whether the low number of ILC3s in the tumour microenvironment is related to pDCs.

First, we used flow cytometry to further verify the expression of apoptosis-related genes in ILC3s in the tumour tissues of colon cancer patients. The results showed that ILC3s overexpressed CD95 and caspase 3 in the tumour tissue compared with that in the distal tumour control group (Fig. 6A and B). Secretion of cytokine IFNα was detected in the TS (Fig. 6A and D).

Previous studies have reported that IFNα is mainly secreted by pDCs [32]. As such, we wanted to detect whether pDCs in the tumour microenvironment secrete IFNα. Owing to the limited number of pDCs in colon tissue and peripheral blood, we isolated pDCs from normal tonsil tissue for subsequent in vitro culture experiments. We added TS to the pDC culture system and used IMQ as a positive control to detect the secretion of cytokine IFNα. TS was able to promote the secretion of IFNα from pDCs compared with the TS itself (Fig. 6C and D). After added anti-IFNα to the pDC culture system with TS, the release of IFNα is significantly reduced (Fig. 6C and E).

To test whether pDCs can affect the survival of ILC3s through IFNα in the tumour microenvironment, we co-cultured pDCs and ILC3s from normal tonsil tissue in the presence or absence of TS, IFNα, and anti-IFNα to detect the expression of apoptosis-related genes and the survival rate of ILC3s. After co-culturing ILC3s and pDCs with TS or IFNα, the expression of apoptosis-related genes caspase 3 and CD95 on ILC3s was significantly upregulated (Fig. 6F and G); the survival rate of ILC3s was significantly reduced (Fig. 6F and H); and the main factor secreted by ILC3s (IL-22) was also significantly downregulated (Fig. 6F and I). After added anti-IFNα to neutralize IFNα, the above effect is obviously weakened (Fig. 6F, G, H and I). Giemsa staining results revealed that the ILC3 cell membrane was incomplete and there were scattered apoptotic bodies (Fig. 6J), this result needs further verification. The above results indicate that pDCs can induce apoptosis of ILC3s through the CD95 pathway by releasing IFNα in the tumour microenvironment.

ILCs are important tissue-resident innate immune cells; the numbers and relative percentages of the three subtypes (ILC1, ILC2, and ILC3) vary in different organs [33, 34]. In response to acute environmental challenges and as tissue-resident cells, ILCs can renew and expand in both lymphoid and non-lymphoid organs [34]. A change in the ILC population in human tissues is associated with the pathogenesis and progression of chronic infections and inflammatory diseases [18, 28, 35]. Recently, Ikeda et al. reported that the number of NKp44⁺ ILC3s from colorectal cancer tissue is associated with tumour-associated tertiary lymphoid structures [22]. Our group collected 58 samples from colon cancer patients to further study the distribution characteristics of ILCs in colon cancer and their correlation with other immune cells.

Flow cytometry showed that the numbers of ILC3s and NKp44⁺ ILC3s in colon tumour tissues were lower than those in distal regions and negatively correlated with the pathological grade of cancer; however, there was no correlation between the number of ILC3s and patient age, sex, tumour location, tumour size, lymphatic metastases, or distant metastases. RNA-Seq showed that among the DEGs in tumour versus distal ILC3s, many were associated with tumour development or inhibition (Table S3); however, most of the DEGs were involved in tumour suppression, especially SCIN, which was upregulated in tumour ILC3s. These data concur with previous studies suggesting that ILC3s in the tumour microenvironment might have dual functions depending on the cancer phase and environmental context [20, 36–38].

Human ILC3s are the most heterogeneous ILCs. In addition to conventional NK cells, the ILC3 population can also express NCRs and can be divided according to this expression into NKp44^+/− ILC3s, NKp30^+/− ILC3s, and NKp46^+/− ILC3s [24]. Additionally, ILC3s can be classified according to the C-C motif chemokine receptor (CCR) 6 expression into CCR6⁺ and CCR6⁻ ILC3s [18]. In the present study, changes in the NKp44^+/− ILC3 population in tumours and proximal and distal regions were similar to those in total ILC3s, especially the NKp44⁺ ILC3 population; however, changes in the number of ILC1s and ILC2s among the analysed locations were not significant. These results may be due to insufficient tissue sample size. In future investigations, we will expand the sample size and repeat this analysis.

pDCs are type-I IFN-producing cells that bridge the innate and adaptive immune systems [32] and are specialised in endosomal TLR7/9-mediated recognition of viral nucleic acids with their response involving massive secretion of type-I IFNs to promote virus removal [39]. pDCs in the tumour microenvironment mainly exist in a non-activated state and are associated with the development and maintenance of an immunosuppressive environment [27, 29–31]. Functional alterations of pDCs in the tumour microenvironment are associated with tumour immune-escape mechanisms [29, 40, 41]. In the present study, the number of pDCs in flow cytometric analysis of colon tumour tissues was higher than that in distal regions and positively correlated with tumour location, pathological grade, lymphatic metastases, and especially distant metastases of colon cancer. However, we did not find any correlation between the number of pDCs and patient age, sex, or tumour size. Our RNA-Seq results showed that, among the genes upregulated in tumour pDCs (versus distal pDCs), many were associated with tumour development, whereas many of the downregulated genes were associated with tumour inhibition (Table S4). These data suggest that pDCs might participate in tumour progression and immune escape.

Zhang et al. [28] reported that chronic HIV-1 infection induces ILC3 apoptosis via pDC activation, induction of type-I IFN expression, and CD95-mediated apoptosis. Additionally, Maazi et al. [42] showed that pDC activation alleviates airway hyperreactivity and inflammation by suppressing ILC2 function and survival. However, the relevance of ILCs and pDCs in the tumour microenvironment has not been reported. Our flow cytometric data showed a negative correlation between ILC3s and pathological grade and a positive correlation between pDCs and pathological grade. Additionally, we found a negative correlation between percentages of ILC3s and pDCs, with RNA-Seq analysis subsequently confirming this result. The analysis of ILC3 versus pDC DEGs showed that many tumour ILC3 DEGs were involved in RNA degradation, metabolic, and apoptotic pathways, whereas most tumour pDC DEGs were associated with tumour development or inhibition. Julieta et al. [43] reported mRNA degradation as an early apoptotic event in colon cancer, which is concordant with our findings.

In the in vitro experiments, after co-culturing ILC3s and pDCs with TS or IFNα, the expression of apoptosis-related genes caspase 3 and CD95 on ILC3s was significantly upregulated; the survival rate of ILC3s was significantly reduced. In addition to molecules caspase 3 and CD95, other apoptosis-related genes on ILC3s may play important roles in the way pDCs affect ILC3s; this needs further verification. For pDCs, KEGG pathway analysis showed that many of the DEGs were associated with cancer. This supports the results reported by Zhang et al. [28] that pDCs might induce ILC3s apoptosis during chronic HIV-1 infection. Additionally, in the colon cancer environment, pDCs may induce ILC3 apoptosis and promote tumour progression, which would explain the difference in percentage of ILC3s and pDCs in tumour tissues (ILC3s, low; pDCs, high). Moreover, we found multiple upregulated and downregulated genes with similar patterns between ILC3s and pDCs. Pearson correlation analysis of all samples showed obvious correlations between ILC3s and pDCs in colon cancer tissue samples. In our future work, we will confirm these results using in vivo experiments.

Su et al. [25] reported that different levels of circulating immune cells are associated with tumour location, stage, differentiation status, and lymphatic metastases in patients with colon cancer. Additionally, they found that the epidemiology, pathogenesis, genetic and epigenetic alterations, molecular pathways, and prognoses differed in patients with left-sided and right-sided colon cancers. In the present study, we found that the percentage of pDCs in the tumour tissue was correlated with the region of the colon with the tumour and that the number of pDCs progressively decreased in the sigmoid, descending, transverse, and ascending colon. However, there was no correlation between the percentage of ILC3s and the region of the colon with the tumour or between the pathological grade and the tumour region (Figures S1B and C).

Interestingly, our results showed that the number of ILC3s in the tumour was lower than that in distal and proximal regions, but the number of ILC3s in the proximal region was higher than that in the distal region. Additionally, we found a negative correlation between ILC3s from proximal regions and the pathological grade of cancer (data not shown). RNA-Seq analysis revealed thousands of DEGs between proximal and distal ILC3s (data not shown), including oncogenes. In our future work, we plan to investigate the role of ILC3s in tumour and proximal regions in colon cancer.

In summary, our data reveal that ILC3s and pDCs represent important cellular components in colon cancer that may participate in tumour progression or inhibition. Specifically, pDCs may induce immune tolerance and promote tumour metastasis in the tumour microenvironment. Furthermore, our findings suggest that ILC3s and pDCs may represent novel therapeutic targets for the modulation of the immune response against colon cancer. Moreover, the identification of ILC3s and pDCs in tumour specimens may represent a new immune score factor to aid in prognostic determination for patients undergoing surgery for colon cancer.

dendric cell; pDCs:plasmacytoid dendritic cells; ILCs:innate lymphoid cells; NK:natural killer; ILC1s:group 1 ILCs; ILC2s:group 2 ILCs; ILC3s:group 3 ILCs; IFN:interferon; IL:interleukin; TCR:T cell receptor; MNCs:mononuclear cells; NCR:natural cytotoxicity receptor; BDCA2:blood DC antigen 2; PD-L1:programmed death-ligand 1; DEGs:differentially expressed genes; RNA-Seq:RNA sequencing; KEGG:Kyoto Encyclopedia of Genes and Genomes; FPKM:fragments per kilobase of transcript per million; mDC:myeloid DC; Breg:B regulatory; CCR6:C-C motif chemokine receptor 6.

Acknowledgments

We thank the Stomach Colorectal Anal Surgery in the First Hospital of Jilin University for the collection and provision of clinical samples for this study. We are grateful to the patients for volunteering to participate in this study.

Author contributions

JTC and JW conceived the study. JW, HC, and GXZ performed the experiments and analysed the data. MYZ, HLW, JWC, and JS recruited and curated the patient samples and information for flow cytometric analysis. TTL and SZ contributed to the flow cytometry studies. JW and LLQ generated and analysed the RNA-Seq data. JS, YJL, and JTC conceived and designed the experiments and supervised the work with participation from JWC and HU. All authors contributed to data analysis and reviewed and edited the manuscript. All authors approved the final version of the article, including the authorship list.

Funding

This work was supported by the National Natural Science Foundation of China (Grant Nos. 81571534, 81870152, 81901591, and 81800021), the Key Scientific Project of Jilin Province (20140204024YY), the Scientific and Technological Developing Plan of Jilin Province (20160520141JH and 20180101097JC), the 62^nd batch of the China Postdoctoral Science Foundation Fund (801171172842), the “13^th Five-Year” Science and Technology Research of the Education Department of Jilin Province (YYKH20190043KJ), the Jilin Provincial Key Laboratory of Biotherapy (20170622011JC), the Program for JLU Science and Technology Innovative Research Team (2017TD-08), and the Fundamental Research Funds for the Central Universities.

Availability of data and materials

The datasets used and or analysed during the current study are available from the corresponding authors on reasonable request.

Ethics approval and consent to participate

All experiments regarding human tissue slides in this study were performed in according to ethical standards. Informed consent was obtained from all patients.

Consent for publication

All authors involved in the study had given their consent for submitting this article for publication.

Competing interests

The authors have declared that no conflict of interest exists.

Author details

¹Institute of Translational Medicine, The First Hospital, Jilin University, Changchun 130061, China

²Department of Pediatrics, The First Hospital, Jilin University, Changchun 130021, China

³Department of Stomach Colorectal Anal Surgery, The First Hospital, Jilin University, Changchun 130021, China

⁴Department of Pediatric Gastroenterology, The First Hospital, Jilin University, Changchun 130021, China

⁵Department of Cardiovascular Center, The First Hospital, Jilin University, Changchun 130021, China

⁶Cancer Center, The First Hospital, Jilin University, Changchun 130021, China

⁷Department of Microbiology, Icahn School of Medicine at Mount Sinai, New York, New York, USA

⁸Sanofi Research and Development, Cambridge, MA 02139, USA

Li SK, Martin A. Mismatch Repair and Colon Cancer: Mechanisms and Therapies Explored. Trends Mol Med. 2016;22(4):274–89.
Kang M, Martin A. Microbiome and colorectal cancer: Unraveling host-microbiota interactions in colitis-associated colorectal cancer development. Seminars in immunology. 2017;32:3–13.
Hjerkind KV, Qureshi SA, Moller B, Weiderpass E, Deapen D, Kumar B, Ursin G. Ethnic differences in the incidence of cancer in Norway. International journal of cancer. 2017;140(8):1770–80.
Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global patterns and trends in colorectal cancer incidence and mortality. Gut. 2017;66(4):683–91.
Laskowski P, Klim B, Ostrowski K, Szkudlarek M, Litwiejko-Pietrynczak E, Kitlas K, Nienartowicz S, Dzieciol J. Local inflammatory response in colorectal cancer. Polish journal of pathology: official journal of the Polish Society of Pathologists. 2016;67(2):163–71.
Li QL, Ma XY, Yu LL, Xue F, Ma WL, Yao KY. [Age-specific detection rates of colorectal neoplasms by colonoscopic screening in high-incidence rural area]. Zhonghua zhong liu za zhi [Chinese journal of oncology]. 2013;35(2):154–7.
Subramanian V, Chatu S, Echterdiek F, Banerjee A, Finlayson C, Pollok RCG. Patients with Endoscopically Visible Polypoid Adenomatous Lesions Within the Extent of Ulcerative Colitis Have an Increased Risk of Colorectal Cancer Despite Endoscopic Resection. Digestive diseases sciences. 2016;61(10):3031–6.
Bellik L, Gerlini G, Parenti A, Ledda F, Pimpinelli N, Neri B, Pantalone D. Role of conventional treatments on circulating and monocyte-derived dendritic cells in colorectal cancer. Clinical immunology (Orlando Fla). 2006;121(1):74–80.
Menon S, Shin S, Dy G. Advances in Cancer Immunotherapy in Solid Tumors. Cancers 2016, 8(12).
Di Franco S, Turdo A, Todaro M, Stassi G. Role of Type I and II Interferons in Colorectal Cancer and Melanoma. Frontiers in immunology. 2017;8:878.
Candido J, Hagemann T. Cancer-related inflammation. J Clin Immunol. 2013;33(Suppl 1):79–84.
Legitimo A, Consolini R, Failli A, Orsini G, Spisni R. Dendritic cell defects in the colorectal cancer. Human vaccines immunotherapeutics. 2014;10(11):3224–35.
Galon J, Fridman WH, Pages F. The adaptive immunologic microenvironment in colorectal cancer: a novel perspective. Cancer research. 2007;67(5):1883–6.
Salome B, Jandus C. Innate lymphoid cells in antitumor immunity. J Leukoc Biol. 2018;103(3):479–83.
Simoni Y, Fehlings M, Kloverpris HN, McGovern N, Koo SL, Loh CY, Lim S, Kurioka A, Fergusson JR, Tang CL, et al. Human Innate Lymphoid Cell Subsets Possess Tissue-Type Based Heterogeneity in Phenotype and Frequency. Immunity. 2017;46(1):148–61.
Vivier E, Artis D, Colonna M, Diefenbach A, Di Santo JP, Eberl G, Koyasu S, Locksley RM, McKenzie ANJ, Mebius RE, et al. Innate Lymphoid Cells: 10 Years On. Cell. 2018;174(5):1054–66.
Krabbendam L, Nagasawa M, Spits H, Bal SM. Isolation of Human Innate Lymphoid Cells. Current protocols in immunology 2018:e55.
Cheng H, Jin C, Wu J, Zhu S, Liu YJ, Chen J. Guards at the gate: physiological and pathological roles of tissue-resident innate lymphoid cells in the lung. Protein cell. 2017;8(12):878–95.
Wang S, Xia P, Chen Y, Qu Y, Xiong Z, Ye B, Du Y, Tian Y, Yin Z, Xu Z, et al. Regulatory Innate Lymphoid Cells Control Innate Intestinal Inflammation. Cell. 2017;171(1):201–16.e218.
Carrega P, Campana S, Bonaccorsi I, Ferlazzo G. The Yin and Yang of Innate Lymphoid Cells in Cancer. Immunology letters. 2016;179:29–35.
Salimi M, Wang R, Yao X, Li X, Wang X, Hu Y, Chang X, Fan P, Dong T, Ogg G. Activated innate lymphoid cell populations accumulate in human tumour tissues. BMC Cancer. 2018;18(1):341.
Ikeda A, Ogino T, Kayama H, Okuzaki D, Nishimura J, Fujino S, Miyoshi N, Takahashi H, Uemura M, Matsuda C, et al: Human NKp44 + group 3 innate lymphoid cells associate with tumor-associated tertiary lymphoid structures in colorectal cancer. Cancer immunology research 2020:canimm.0775.2019.
Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic acids research. 2002;30(1):207–10.
Carrega P, Loiacono F, Di Carlo E, Scaramuccia A, Mora M, Conte R, Benelli R, Spaggiari GM, Cantoni C, Campana S, et al. NCR(+)ILC3 concentrate in human lung cancer and associate with intratumoral lymphoid structures. Nature communications. 2015;6:8280.
Di J, Zhuang M, Yang H, Jiang B, Wang Z, Su X. Clinical significance of circulating immune cells in left- and right-sided colon cancer. PeerJ. 2017;5:e4153.
Li S, Wu J, Zhu S, Liu YJ, Chen J. Disease-Associated Plasmacytoid Dendritic Cells. Frontiers in immunology. 2017;8:1268.
Aspord C, Leccia MT, Charles J, Plumas J. Plasmacytoid dendritic cells support melanoma progression by promoting Th2 and regulatory immunity through OX40L and ICOSL. Cancer immunology research. 2013;1(6):402–15.
Zhang Z, Cheng L, Zhao J, Li G, Zhang L, Chen W, Nie W, Reszka-Blanco NJ, Wang FS, Su L. Plasmacytoid dendritic cells promote HIV-1-induced group 3 innate lymphoid cell depletion. J Clin Investig. 2015;125(9):3692–703.
Sisirak V, Faget J, Gobert M, Goutagny N, Vey N, Treilleux I, Renaudineau S, Poyet G, Labidi-Galy SI, Goddard-Leon S, et al. Impaired IFN-alpha production by plasmacytoid dendritic cells favors regulatory T-cell expansion that may contribute to breast cancer progression. Cancer research. 2012;72(20):5188–97.
Dubrot J, Duraes FV, Potin L, Capotosti F, Brighouse D, Suter T, LeibundGut-Landmann S, Garbi N, Reith W, Swartz MA, et al. Lymph node stromal cells acquire peptide-MHCII complexes from dendritic cells and induce antigen-specific CD4(+) T cell tolerance. The Journal of experimental medicine. 2014;211(6):1153–66.
Wu J, Li S, Yang Y, Zhu S, Zhang M, Qiao Y, Liu YJ, Chen J. TLR-activated plasmacytoid dendritic cells inhibit breast cancer cell growth in vitro and in vivo. Oncotarget. 2017;8(7):11708–18.
Liu YJ. IPC: professional type 1 interferon-producing cells and plasmacytoid dendritic cell precursors. Annu Rev Immunol. 2005;23:275–306.
Riedel JH, Becker M, Kopp K, Duster M, Brix SR, Meyer-Schwesinger C, Kluth LA, Gnirck AC, Attar M, Krohn S, et al. IL-33-Mediated Expansion of Type 2 Innate Lymphoid Cells Protects from Progressive Glomerulosclerosis. Journal of the American Society of Nephrology: JASN. 2017;28(7):2068–80.
Gasteiger G, Fan X, Dikiy S, Lee SY, Rudensky AY. Tissue residency of innate lymphoid cells in lymphoid and nonlymphoid organs. Science. 2015;350(6263):981–5.
Hazenberg MD, Spits H. Human innate lymphoid cells. Blood. 2014;124(5):700–9.
Guillerey C. Roles of cytotoxic and helper innate lymphoid cells in cancer. Mammalian genome: official journal of the International Mammalian Genome Society 2018.
Fung KY, Nguyen PM, Putoczki T. The expanding role of innate lymphoid cells and their T-cell counterparts in gastrointestinal cancers. Molecular immunology 2017.
Wu J, Lv X, Zhu S, Li T, Cheng H, Chen J. Critical Roles of Balanced Innate Lymphoid Cell Subsets in Intestinal Homeostasis, Chronic Inflammation, and Cancer. Journal of immunology research. 2019;2019:1325181.
Arimura K, Takagi H, Uto T, Fukaya T, Nakamura T, Choijookhuu N, Hishikawa Y, Yamashita Y, Sato K. Crucial role of plasmacytoid dendritic cells in the development of acute colitis through the regulation of intestinal inflammation. Mucosal immunology. 2017;10(4):957–70.
Sawant A, Ponnazhagan S. Role of plasmacytoid dendritic cells in breast cancer bone dissemination. Oncoimmunology. 2013;2(2):e22983.
Mitchell D, Chintala S, Dey M. Plasmacytoid dendritic cell in immunity and cancer. J Neuroimmunol. 2018;322:63–73.
Maazi H, Banie H, Aleman Muench GR, Patel N, Wang B, Sankaranarayanan I, Bhargava V, Sato T, Lewis G, Cesaroni M, et al. Activated plasmacytoid dendritic cells regulate type 2 innate lymphoid cell-mediated airway hyperreactivity. J Allergy Clin Immunol. 2018;141(3):893–905.e896.
Del Prete MJ, Robles MS, Guao A, Martinez AC, Izquierdo M, Garcia-Sanz JA. Degradation of cellular mRNA is a general early apoptosis-induced event. FASEB journal: official publication of the Federation of American Societies for Experimental Biology. 2002;16(14):2003–5.

Download PDF

Version 1

posted

You are reading this latest preprint version

Correlation Between Immune Lymphoid Cells and Plasmacytoid Dendritic Cells in Human Colon Cancer

Status:

Version 1

Abstract

Figures

Background

Methods

Patient tissue specimens

Tissue digestion for single-cell suspension

Isolation of ILC3s and pDCs

Flow cytometric analysis

Immunohistochemistry

RNA-Seq and analysis

Preparation of colon tumour-derived supernatant

Co-culture of ILC3s and pDCs

ELISA

Giemsa staining

Statistics

Results

Number of ILC3s in colon cancer tissue specimens is negatively correlated with disease progression

Decreased numbers of ILC3s in colon cancer tissues are correlated with higher pDC levels

Correlation between ILC3s and pDCs at the level of differentially expressed genes (DEGs)

pDCs can induce apoptosis of ILC3s in the tumour microenvironment

Discussion

Conclusions

Abbreviations

Declarations

References

Supplementary Files

Status:

Version 1