key: cord-0782438-1tuboh8n
authors: Meghnem, Dihia; Oldford, Sharon A.; Haidl, Ian D.; Barrett, Lisa; Marshall, Jean S.
title: Histamine receptor 2 blockade selectively impacts B and T cells in healthy subjects
date: 2021-04-30
journal: Sci Rep
DOI: 10.1038/s41598-021-88829-w
sha: 4b59dcaf77fe4a96facdc4d1899b583a035ffcdc
doc_id: 782438
cord_uid: 1tuboh8n

Histamine receptor 2 (H2R) blockade is commonly used in patients with gastric, duodenal ulcers or gastroesophageal reflux disease. Beyond the gastrointestinal tract, H2R is expressed by multiple immune cells, yet little is known about the immunomodulatory effects of such treatment. Clinical reports have associated H2R blockade with leukopenia, neutropenia, and myelosuppression, and has been shown to provide clinical benefit in certain cancer settings. To systematically assess effects of H2R blockade on key immune parameters, a single-center, single-arm clinical study was conducted in 29 healthy subjects. Subjects received daily high dose ranitidine for 6 weeks. Peripheral blood immunophenotyping and mediator analysis were performed at baseline, 3 and 6 weeks into treatment, and 12 weeks after treatment cessation. Ranitidine was well-tolerated, and no drug related adverse events were observed. Ranitidine had no effect on number of neutrophils, basophils or eosinophils. However, ranitidine decreased numbers of B cells and IL-2Rα (CD25) expressing T cells that remained lower even after treatment cessation. Reduced serum levels of IL-2 were also observed and remained low after treatment. These observations highlight a previously unrecognised immunomodulatory sustained impact of H2R blockade. Therefore, the immune impacts of H2R blockade may require greater consideration in the context of vaccination and immunotherapy.

Immune phenotyping. Immune phenotyping for MDSC and lymphocyte populations was performed on whole blood cells collected in BD Vacutainer Lithium Heparin tubes (BD) by flow cytometry. Antibodies to CD11b, CD14, CD15, CD33, CD66b, HLA-DR, CD3, CD19, CD56, CD4, CD8, CD16, CD25, CD57, CD314 (NKG2D), CD159a (NKG2A), Fixable Viability Stain 780 and Brilliant Stain Buffer were obtained from BD Biosciences (San Jose, CA). Antibody clones are indicated in Supplementary table S1. CD1d tetramers loaded with synthetic glycolipid PBS57 were obtained from the NIH tetramer core facility (Emory University, Atlanta, GA). For each donor, 150 µL of whole blood cells were incubated with fixable viability stain 780 (BD Bioscience) for 10 min on ice then washed in PBS supplemented with 2% FCS. The Fc receptors were blocked with 3% human immunoglobulin (Hizantra, CSL Behring Canada, Ottawa, Canada) for 10 min. Combinations of fluorescently tagged antibodies (Supplementary Table S1 ) were prepared in brilliant violet buffer and 50 µl were added to each sample for 30 min. After staining, cells were washed twice with PBS supplemented with 2% FCS (Gibco) and the RBC were lysed by adding 5 ml of ammonium chloride for 30-60 s then washed with PBS supplemented with 2% FCS. Cells were fixed with PBS containing 1% paraformaldehyde overnight at 4 °C then washed and resuspended in PBS supplemented with 2% FCS before analysis on BD LSRFortessa™ (BD). The instrument was calibrated with BD Rainbow Calibration Particles (BD) before each acquisition to ensure consistent readings over time. Data were analyzed using FlowJo Version 10 software (BD). Sample gating strategies are depicted in Supplementary Figure S1 . Gating was based on fluorescence minus one control. Absolute numbers of immune cell subsets were calculated from automated total white blood cell counts.

Multiplex immunoassays and ELISA. Soluble mediator levels were assessed in platelet poor plasma isolated from whole blood samples collected in BD Vacutainer EDTA tubes (BD). Plasma samples were assessed for IL-1β, IL-1RA, IL-2, IL-6, IL-7, IL-8, IL-10, IL-15, M-CSF, MMP-9, CCL2, CCL5, CXCL1, IFN-γ, TNF, VEGF-A, IgA, IgE, IgG1, IgG2, IgG4 and IgM via magnetic Luminex performance assays (ThermoFisher, Oakville, Canada) and G-CSF, (R&D Systems, Oakville, Canada). Samples were read using a Bio-Plex 200 system (Bio-Rad, Mississauga, Canada) and analyzed with Bio-Plex Manager 6.0 software (Bio-Rad). Plasma levels of B cell 

Study population. A total of 34 subjects were screened and 30 healthy volunteers were enrolled. One patient withdrew 2.6 weeks after study start. Data are presented on the remaining 29 healthy volunteers who completed the study. Participants received ranitidine twice daily at a dosage between 7.58 and 8.5 mg/kg/day, a high dose but lower than the one used in patients with Zollinger Ellison syndrome. The median age of the cohort was 31 years (range 23-47 years), all were Caucasian, and 17% were male (Supplementary Table S2 ). Self-reported ranitidine adherence was on average 97% adherence (range 89-100%). There were no study drugrelated safety issues.

Ranitidine treatment did not affect the overall leukocyte profile or MDSC numbers. Ranitidine use has been associated with neutropenia, anemia, granulocytopenia and thrombocytopenia in various clinical contexts 13, 33 . Complete blood counts and differential white blood cell counts were assessed using automated hematology at baseline, during and following ranitidine treatment. Ranitidine treatment did not significantly alter whole blood hematologic counts (Table 1) . Minor changes in the number of reticulocytes (RET), mean capsular hemoglobin concentration (MCHC), and Hematocrit (HCT) were observed over the trial but were within normal clinical ranges. Since ranitidine or famotidine treatment has been shown to regulate MDSCs in murine models 21, 34 , MDSC subsets were examined in trial subjects by flow cytometry 35 to identify early stage MDSC or where data distribution was appropriate using repeated measures, one-way ANOVA followed by Dunnett's multiple comparison, using T0 as control. *P < 0.05, **P < 0.01 compared to T0. a CBC with WBC differential counts were assessed before ranitidine treatment (T0), after 3-and 6-weeks of treatment (T3 and T6), and after 12 weeks treatment cessation (T18). Data are presented as median (interquartile range). Figure S1A ). The level of MDSCs are normally low in healthy individuals but increase in cancer patients 36 . As expected, low numbers of circulating MDSCs were observed in these healthy subjects, and their overall numbers changed between some sampling times during the study (Table 1) . However, ranitidine treatment did not affect the absolute numbers of MDSC subsets at T3, T6 and T18 when compared to T0 (Supplementary figure S2D-F). Analysis of the percentage of the three different populations showed that the percentage of PMN-MDSC were decreased at T18 (Supplementary Figure S2 ).

Ranitidine treatment was associated with a decrease in B lymphocytes but not immunoglobulin levels. H2R is known to regulate B cell activation, antibody production and class switch in experimental models 33 . We examined the effects of ranitidine on B cells and T cells by flow cytometry (Supplementary Figure S1B ). Ranitidine treatment decreased the absolute numbers of CD4 + and CD8 + T cells at T6 and T18, respectively (Fig. 1A ,B right panels). However, it did not significantly affect their percentages (Fig. 1A ,B left panels). Interestingly, both the numbers and percentage of CD19 + B cells were profoundly altered by ranitidine treatment. The absolute number of B cells decreased after 3 and 6 weeks of ranitidine treatment and this decrease was maintained after treatment cessation (Fig. 1C) with a post-treatment T18 trend towards T0 baseline levels.

Given the notable decrease in B cells observed following ranitidine treatment serum immunoglobulins were also assessed. No significant changes in immunoglobulin levels were observed within the time frame of this study (Supplementary Figure S3 ).

Ranitidine treatment did not initially affect the numbers of CD3 -CD56 + NK cells but a modest but statistically significant decrease at T6 and T18 was observed compared to the baseline T0 (T6 P = 0.033; T18 P = 0.0034, Fig. 2A 

Ranitidine has been shown to act synergistically with IL-2 to increase NK activity in vitro 39 and increased the efficacy of IL-2 anti-tumor therapy in a clinical study 27 . Therefore, we analyzed the effect of ranitidine treatment on IL-2Rα (CD25) expression in T lymphocytes (Fig. 3) . Interestingly, the percentage of CD25 + CD4 + T cells initially increased at T3 before a decline at T6 and T18 (Fig. 3A left panel) . Numbers of CD25 expressing CD4 + cells were decreased at the end of ranitidine treatment and this decrease was sustained following treatment cessation (Fig. 3B left panel) . Numbers of CD25 expressing CD8 + cells were also decreased at the end of ranitidine treatment (T6; P = 0.0330) and this decrease was sustained at T18 (P = 0.0001, Fig. 3B right panel). Similar changes were observed for CD25 + CD8 + percentages (Fig. 3A right panel) .

Selective decrease in serum IL-2 levels associated with ranitidine treatment. Levels of selected soluble immune mediators associated with lymphocyte development and activation or hematopoiesis were assessed in plasma before, during and following ranitidine treatment ( Table 2) . Levels of immune mediators in subjects following ranitidine treatment were compared to baseline measures in individual subjects. Notably, ranitidine treatment was associated with a selective decrease in plasma IL-2 levels ( Table 2) . IL-2 was detectable in 25 of the 29 individuals at baseline but was decreased following ranitidine treatment (T3 and T6; P < 0.0001) and remained significantly lower following treatment cessation (T18; P = 0.0132) ( Table 2) . Interestingly, G-CSF was present at very low levels and detectable above the limit of detection in only 2 of the 29 individuals at T0. G-CSF levels increased following treatment up to 3.61 ± 1.03 pg/ml at T6 (end of treatment timepoint), at which time 31% (n = 9) of subjects had detectable G-CSF levels. At T18 62% (n = 18) of individuals had detectable G-CSF, which was further increased to 6.35 ± 1.02 pg/ml (T18; P < 0. 01) compared to T6 or baseline.

The effect of ranitidine on human immune cell populations has not been well defined, despite its very common clinical use, especially in the elderly. The present study demonstrates that ranitidine treatment was associated with sustained decreases in CD19 + B cells (Fig. 1) and CD25 expressing CD4 + and CD8 + T cells (Fig. 3) . However, even at a relatively high dose, ranitidine treatment did not alter total peripheral white blood cell, red cell and platelet parameters ( Table 1 ). The percentage and the absolute number of circulating basophils, neutrophils and monocytes were also not altered by ranitidine (Table 1) although H2R blockade was associated with decreased percentages of PMN-MDSC (Supplementary Figure S1 ).

To our knowledge, this study defines, for the first-time, the effects of H2R blockade on immune cells in healthy individuals. While ranitidine has been associated with neutropenia in some clinical reports 11, 12 , it did not induce neutropenia in healthy individuals over a six week time course (Table 1) even when used at a relatively high clinical dose. Ranitidine has also been associated clinically with thrombocytopenia 40,41 but we did not observe such impact (Table 1) . A sixfold increase in plasma G-CSF levels was observed after cessation of ranitidine treatment ( Table 2 ). While we do not have a clear mechanistic explanation for this G-CSF increase, this change might suggest a compensatory effect for changes in granulocyte populations. We recognise that the dose of ranitidine used in this study was above the one used in a clinical setting and there remain some possibility for off target impacts This study made the novel observation that ranitidine use was associated with decreased numbers of peripheral CD19 + B cells (Fig. 1C) . Few studies have analysed the effects of ranitidine on B cells. Ranitidine was shown to modify T lymphocyte function when administered acutely to patients with head injury while having no effect on B cell frequencies 42 . Conversely, in B cell chronic lymphocytic leukemia, ranitidine treatment increased the antibody response to tetanus-toxoid conjugated or unconjugated Haemophilus influenzae type-B vaccine 51, 52 . Ranitidine also increased antibody responses in a murine tumor model 23 . In these studies, an increase in plasma cells following ranitidine treatment has been suggested. Differentiation of B cells during their developmental stages is supported by IL-7 43 while their differentiation to plasma cells can be regulated by alternate mediators were assessed before ranitidine treatment (T0), after 3-and 6-weeks treatment (T3 and T6), and 12 weeks after treatment cessation (T18) by flow cytometry. Statistical analysis was performed using repeated measures Friedman's test with Dunn's multiple comparison using T0 as control. Graphs depict median and IQR, n = 29. *P < .05; **P < .01; ***P < .001; ****P < .0001. 44 and is accompanied by a down regulation of CD19 surface marker 45, 46 . The analysis of soluble BAFF and IL-7 in our study (Table 2 ) did not support an elevation of these mediators in ranitidine treated subjects. Regardless of mechanism, ranitidine impacts on B cells are substantial and might have implications for the humoral immune responses. H2R blockade has been demonstrated to have some beneficial impacts clinically in the context of multiple myeloma. Although the bulk of multiple myeloma cells do not express CD19 it has been suggested that a critical population of myeloma propagating cells is CD19 positive 47, 48 . The notable impact of ranitidine treatment in decreasing CD19 positive cells in the blood could therefore be worthy of further investigation in the context of this disease, where ranitidine treatment has already been suggested to be beneficial 39 .

NK cells are known to express H4R and H2R 49, 50 and their activity can be regulated by histamine receptors antagonists such as ranitidine [50] [51] [52] . In this study, NK cells were decreased by ranitidine treatment (Fig. 2A) , which was reflected in a decrease in CD56 dim CD16 bright , CD56 bright CD16 neg and CD56 bright CD16 dim (Fig. 2B) sub-populations. It was further observed that ranitidine treatment decreased NK cell surface expression of NKG2D (Supplementary Figure S4B) . NKG2D is an activating receptor expressed by T cells and NK cells and is important in anti-tumor as well as anti-viral immune responses 53 . The observed effects on NKG2D expression could potentially lead to a decrease in NK and CD8 + T cell activation. T0 T3 T6 T18  T0 T3 T6 T18  T0 T3 T6 T18   0 Ranitidine treatment is associated with a decrease of NK cells: The percentage and number of peripheral blood total CD3-CD56 + NK cells (A) and their subsets CD56 bright CD16 neg , CD56 bright CD16 dim cells and CD56 dim CD16 bright NK cells subsets (B) were assessed before ranitidine treatment (T0), after 3-and 6-weeks treatment (T3 and T6), and 12 weeks after treatment cessation (T18) by flow cytometry. Statistical analysis was performed using repeated measures Friedman's test with Dunn's multiple comparison using T0 as control (A and B) and t-test to compare immunoglobulin level at T0 and T6. Graphs depict median and IQR, n = 29. *P < .05; **P < .01. Figure S2) , but not the absolute number of PMN-MDSC was decreased in the peripheral blood after ranitidine treatment (Supplementary figure S2) . This finding is similar in nature to the observed impact of ranitidine treatment on MDSC populations in mice 34 and could contribute to the reported positive impact of ranitidine in some cancer settings.

Ranitidine treatment modulated CD25 expression on T lymphocytes and NK cells ( Fig. 3 and Supplementary Figure S4A) . A previous study of the postoperative effect of ranitidine on abdominal surgery patients showed an increased plasma level of CD25 in treated patients 56 , indicating a potential role of ranitidine in activation of CD4 + T cells and subsequent shedding of CD25. Ranitidine, in combination with IL-2, was also shown to increase NK cell cytotoxicity in vitro 9 as well as having synergistic effects with anti-tumoral IL-2 therapy 27 . Indeed, frequencies of CD25 expressing cells were increased after 3 weeks of ranitidine treatment (Fig. 3A left panel) which can potentiate IL-2-induced T and NK cell activation. IL-2 itself is a key regulator of CD25 expression and was found to be decreased by three-fold in the plasma after ranitidine treatment. Overall, the effects of ranitidine on B cells as well as CD25 expressing cells and IL-2 plasma levels indicate that H2R blockade profoundly impacts several aspects of acquired immunity.

Ranitidine and other H2R antagonists such as cimetidine and famotidine have previously been shown to affect immune responses to viral infections or vaccination. Cimetidine had beneficial effects during several viral infections [57] [58] [59] and increased immune responses during vaccination 29, 30 . Recently, famotidine was associated with beneficial effects in COVID-19 31 and its use correlated with a two-fold decrease in need for intubation or death 32 . The authors suggested a potential effect of famotidine via decreased cytokine levels 32 . Our results indicate that treatment with the related H2 antagonist ranitidine significantly decreased IL-2 compared to baseline and decreased IL-15 and IL-1β in individuals with higher baseline levels ( Table 2) . Furthermore, ranitidine has an immunosuppressive effect on NK and T cells demonstrated by a decreased in CD25 and NKG2D expressing . Ranitidine treatment is associated with a substantial decrease in CD25 expressing T cells: The percentage (A) and number (B) of peripheral blood CD25 expressing CD4 + and CD8 + T cells were assessed before ranitidine treatment (T0), after 3-and 6-weeks treatment (T3 and T6), and 12 weeks after treatment cessation (T18) by flow cytometry. Statistical analysis was performed using repeated measures Friedman's test with Dunn's multiple comparison using T0 as control. Graphs depict median and IQR, n = 29. *P < .05; **P < .01; ***P < .001; ****P < .0001. Figure S4 ). These two phenomena might dampen the immune or inflammatory response to infection and help explain the effect of H2R blockade in COVID-19 infection.

Ranitidine is known to inhibit the constitutive activity of H2R 67 Such blockade will have a direct effect on H2R expressing cells. Ranitidine can also act on bone marrow cells 14 . For example, hematopoietic stem cells treated with histamine exhibited anti-apoptotic effects on neutrophils and granulocytes 60 . Therefore, inhibiting histamine receptors can lead to altered immune populations in the periphery. The half-life of ranitidine is estimated to be three hours while some observed effects were maintained up to 12 weeks after treatment cessation and these impacts are more likely a result of a long-term effect on the bone marrow microenvironment.

Overall, this study demonstrates that H2R inhibition using ranitidine in healthy individuals induced substantial, prolonged, and multifaceted changes in immune cells to an extent that is likely to be important in immune response regulation. These changes were most notable in the CD19 positive B cell and CD25 positive T cell populations and were associated with decreased plasma IL-2. Our findings indicate that when clinically evaluating immunotherapeutic strategies and immunization responses, greater consideration should be given to histamine receptor antagonists as contributors to variable or modified immune responses.

Received: 14 December 2020; Accepted: 12 April 2021 Table 2 . Effects of ranitidine treatment on plasma soluble mediator levels. Plasma soluble mediators were measured before treatment (T0) after 3-and 6-weeks treatment (T3 and T6) and after 12 weeks treatment cessation (T18) by multiplex immunoassay or ELISA. The values depicted represent concentrations (pg/ ml) ± SEM. Statistical analysis was performed using a Friedman's test followed by Dunn's multiple comparison, or, where appropriate for data distribution, repeated measures one-way ANOVA test followed by Dunnett's multiple comparison, using T0 as control. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 compared to T0. ND, not detectable (analyte below the LOD in > 85% of individuals, across timepoints). LOD limit of detection. 

Neonatal Formulary

Effects of eight weeks' continuous treatment with oral ranitidine and cimetidine on gastric acid secretion, pepsin secretion, and fasting serum gastrin

The role of histamine and histamine receptors in mast cell-mediated allergy and inflammation: The hunt for new therapeutic targets

The guinea pig histamine H2 receptor: gene cloning, tissue expression and chromosomal localization of its human counterpart

Histamine H(2) receptors mediate the inhibitory effect of histamine on human eosinophil degranulation

Expression and function of histamine receptors 1 and 2 on human monocyte-derived dendritic cells

Expression and function of histamine receptors in human monocyte-derived dendritic cells

H1 and H2 histamine receptors are absent on Langerhans cells and present on dermal dendritic cells

7 Hermodsson, S. Histamine H2-receptor-mediated regulation of human natural killer cell activity

T lymphocytes as a target of histamine action

Haematological adverse effects of histamine H2-receptor antagonists

Ranitidine-induced granulocytopenia: Recurrence with cimetidine administration

Agranulocytosis possibly caused by ranitidine in a patient with renal failure

The use of ranitidine in bone marrow transplantation: A review of 223 cases

Bone marrow hypoplasia during intensive care: Bone marrow culture studies implicating ranitidine in the suppression of haemopoiesis

Histamine potently suppresses human IL-12 and stimulates IL-10 production via H2 receptors

histamine receptors and their role in immune pathology

Evidence for H2-receptor-mediated inhibition of histamine release from isolated rat mast cells

Histamine as an autocrine growth factor: An unusual role for a widespread mediator

Bone marrow myeloid cells regulate myeloid-biased hematopoietic stem cells via a histamine-dependent feedback loop

The impact of ranitidine on monocyte responses in the context of solid tumors

Ranitidine modifies myeloid cell populations and inhibits breast tumor development and spread in mice

Ranitidine inhibition of breast tumor growth is B cell dependent and associated with an enhanced antitumor antibody response

The role of cyclooxygenase-2 in mediating the effects of histamine on cell proliferation and vascular endothelial growth factor production in colorectal cancer

Super-resolution microscopy reveals ultra-low CD19 expression on myeloma cells that triggers elimination by CD19 CAR-T

The role of CD19 and CD27 in the diagnosis of multiple myeloma by flow cytometry: a new statistical model

Sustained indomethacin and ranitidine with intermittent continuous infusion interleukin-2 in advanced malignant melanoma: a phase II study

Proton pump inhibitors and histamine 2 blockers are associated with improved overall survival in patients with head and neck squamous carcinoma

Cimetidine augments Th1/Th2 dual polarized immune responses to recombinant HBV antigens

Cimetidine synergizes with Praziquantel to enhance the immune response of HBV DNA vaccine via activating cytotoxic CD8(+) T cell

Delivering benefits at speed through real-world repurposing of off-patent drugs: The COVID-19 pandemic as a case in point

Famotidine use is associated with improved clinical outcomes in hospitalized COVID-19 patients: A propensity score matched retrospective cohort study

Mechanism of ranitidine associated anemia

Ranitidine modifies myeloid cell populations and inhibits breast tumor development and spread in mice

Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards

Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards

CD56bright natural killer (NK) cells: an important NK cell subset

Role of distinct natural killer cell subsets in anticancer response

The effect of ranitidine on cellular immunity in patients with multiple myeloma

Ranitidine-induced thrombocytopenia in a neonate: A case report and review of literature

Ranitidine-induced thrombocytopenia: A rare drug reaction

Ranitidine improves lymphocyte function after severe head injury: Results of a randomized, double-blind study

The many roles of IL-7 in B cell development; mediator of survival, proliferation and differentiation

BAFF and BAFF-receptor in B cell selection and survival

Long-lived plasma cells are contained within the CD19(-)CD38(hi)CD138(+) subset in human bone marrow

Early emergence of CD19-negative human antibody-secreting cells at the plasmablast to plasma cell transition

Prognostic value of immunophenotyping in multiple myeloma: A study by the PETHEMA/GEM cooperative study groups on patients uniformly treated with high-dose therapy

Chimeric antigen receptor T cells against CD19 for multiple myeloma

Allergic inflammation is augmented via histamine H4 receptor activation: The role of natural killer cells in vitro and in vivo

Functional expression of H4 histamine receptor in human natural killer cells, monocytes, and dendritic cells

Differential effects of histamine receptor antagonists on human natural killer cell activity

NKp46 and NKG2D receptor expression in NK cells with CD56dim and CD56bright phenotype: Regulation by histamine and reactive oxygen species

Natural killer cells: their role in tumour immunosurveillance

Cutting edge: Mast cells critically augment myeloid-derived suppressor cell activity

Myeloid-derived suppressor cells as a therapeutic target for cancer

Effect of ranitidine on soluble interleukin 2 receptors and CD8 molecules in surgical patients

Efficacy of cimetidine in treatment of Herpes zoster in the first 5 days from the moment of disease manifestation

Cimetidine therapy of herpes simplex virus infections in immunocompromised patients

Cimetidine treatment for recurrent respiratory papillomatosis

Characterization and function of histamine receptors in human bone marrow stromal cells

We would like to thank Maria Vaci, Nong Xu and Alexander Edgar for their technical assistance in addition to the support of the staff of the Canadian Centre for Vaccinology, Halifax, NS, Canada. We also thank Dr. Thomas Issekutz for his constructive comments on the manuscript.

The authors declare no competing interests.

The online version contains supplementary material available at https:// doi. org/ 10. 1038/ s41598-021-88829-w.Correspondence and requests for materials should be addressed to J.S.M.

Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.