Luo Tong formula attenuates retinal inflammation in diabetic rats via inhibition of the p38MAPK/NF-κB pathway
- 120 Downloads
Diabetic retinopathy (DR) is a serious microvascular complication of diabetes and remains the leading cause of blindness in adults. Retinal inflammation is playing a crucial role in the development of DR, and targeting inflammatory mediators is a promising strategy for controlling DR. Here, we investigated compound Chinese medicine Luo Tong formula (LTF) alleviated retinal inflammatory responses in a STZ-induced diabetic rat model.
Sprague–Dawley rats were divided into four groups: control, streptozotocin-induced diabetic, LTF-treated diabetic, and calcium dobesilate (CaD)-treated diabetic rats. Blood samples were collected for blood glucose examination. Hematoxylin–eosin and periodic acid-Schiff staining were conducted for light microscopy observations. Retinal cell apoptosis was detected using the TUNEL assay. Proteins expression was quantified by Western blotting and/or immunohistochemistry, and gene expression was assessed by real-time PCR.
Diabetic rats showed significant increases in the expression of tumor necrosis factor α (TNF-α), interleukin-1β (IL-1β), monocyte chemotactic protein-1 (MCP-1), intercellular adhesion molecule-1 (ICAM-1), nuclear factor-κB (NF-κB), and the phospho-p38 mitogen-activated protein kinase (p-p38-MAPK)/p38 MAPK ratio compared to control rats. LTF treatment significantly improved both retinal and pancreatic pathological injury, LTF treatment also inhibited inducible the p-p38 MAPK/p38 MAPK ratio and NF-κB activation and decreased the subsequent induction of the retinal expression of proinflammatory mediators TNF-α, IL-1β, MCP-1 and ICAM-1 compared to diabetic rats. LTF also exhibited a protective effect on islet function.
LTF before the onset of DR can alleviate retinal pathological injury, LTF may play an anti-inflammatory role by inhibiting p38-MAPK and then inhibiting NF-κB pathway. But further studies are needed to confirm this conclusion.
Trial registration This is an animal experiment, trial registration is not necessary.
KeywordsDiabetic retinopathy Prevention Luo Tong formula Pro-inflammatory factors NF-κB pathway p38-MAPK pathway
one-way analysis of variance
Association for Research in Vision and Ophthalmology
bovine serum albumin
high-performance liquid chromatography
intercellular adhesion molecule-1
Luo Tong formula
monocyte chemotactic protein-1
p38 mitogen-activated protein kinase
sodium dodecylsulfate polyacrylamide gel electrophoresis
tumor necrosis factor α
vascular cell adhesion molecule-1
Diabetic retinopathy (DR) is one of the common and severe microvascular complications of diabetes mellitus (DM) and one of the leading causes of preventable blindness worldwide. The prevalence of DR increases with the duration of DM, and nearly all type 1 diabetic patients and more than 60% type 2 diabetic patients suffer from some degree of retinopathy in the 20 years after the oncet of DM [1, 2, 3]. Proliferative diabetic retinopathy is closely associated with vision loss, which is irreversible and seriously affects the quality of life. Early intervention is of great significance. Inflammation response contributes to the early development of DR [4, 5, 6]. Both an increase of inflammatory factors and leukocyte adhesion contribute to blood-retinal barrier (BRB) breakdown. In diabetic patients and diabetic animals, many proinflammatory factors are up-regulated in retinas, including tumor necrosis factor α (TNF-α), interleukin-1β (IL-1β), interleukin-6 (IL-6) and monocyte chemotactic protein-1 (MCP-1), etc., which are associated with vascular leakage. Additionally, studies have shown that hyperglycemia damages vasculature via inducing endothelial activation and pro-inflammatory phenotype of endothelial cells [7, 8], which is characterized by up regulation of cell surface adhesion molecules such as intercellular cell adhesion molecule-1(ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1). These molecules cause the adhesion and results in the recruitment of leukocyte across endothelium, leading to retinal vascular inflammation and endothelial cell death, which ultimately induces DR [7, 8, 9]. Controlling inflammation can help reduce the overall severity of retinopathy and macular edema and restore the BRB . Additionally, persistent hyperglycemia impairs the function of pancreatic beta cells and eventually leads to apoptosis. The function and quantity of pancreatic beta cells are essential for blood sugar homeostasis [11, 12]. The Chinese medicine compound, Luo Tong formula (LTF), composed of Huangqi (Radix Astragali seu Hedysari), Danshen (Radix Salviae Miltiorrhizae), Sanqi (Radix Notoginseng), Shuizhi powder (Hirudo), and Dahuang (Rhubarb), possesses blood invigorating properties and collaterals unblocking properties. We have previously shown that preventive treatment with LTF before the onset of DR can alleviate or delay the process of DR in diabetic rats. LTF can also ameliorate injury due to oxygen free radicals and improve endothelial cell function [13, 14]. All herbs in LTF have anti-inflammatory properties and have been found to inhibit inflammation by downregulating nuclear factor-κB (NF-κB) and p38 mitogen-activated protein kinase (p38-MAPK) signaling pathways [15, 16, 17, 18]. However, the anti-inflammatory effects of LTF on DR are yet to be elucidated. In this study, we aimed to explore the mechanisms underlying the LTF modulation of retinal inflammation through p38-MAPK/NF-κB signaling pathways, providing new scientific evidence for complementary and alternative therapies.
Materials and methods
Standardization of Luo Tong formula
Medicine material crude slices of LTF adopted for this study were provided by Kangmei Pharmaceutical Co., Ltd. after quality assessment, and all herbs were from the same batch. Decoctions were prepared by the Guang`anmen Hospital, China Academy of Chinese Medical Sciences, according to standard operating procedures. LTFwas composed of Huangqi (Radix Astragali seu Hedysari), Danshen (Radix Salviae Miltiorrhizae), Sanqi (Radix Notoginseng), Shuizhi powder (Hirudo), and Dahuang (Rhubarb), in the ratio of 15:10:3:2:1. The major compounds in LTF were identified using high-performance liquid chromatography (HPLC; Waters 2695 HPLC system; Waters, CA, USA). A Luna® Omega Polar C18 analytical column (250 × 4.6 mm, 3.0 μm; Phenomenex, CA, USA) with a mobile phase that contained acetonitrile (A) and -0.2% phosphoric acid acid in water (B) was used. The mobile phase gradient elution was programmed as follows: 23% A (0–10 min), 23–34% A (10–20 min), 34% A (20–31 min), 34–75% A (31–36 min) and 75% (36–60 min); 77% B (0–10 min), 77–66% B (10–20 min), 66% B (20–31 min), 66–25% B (31–36 min) and 25% (36–60 min). The column temperature was maintained at 35 ℃, the flow rate was set at 0.5 mL/min, and a detection wavelength of 203 nm was used. LTF was dissolved in double distilled water containing 0.05% dimethylsulfoxide (DMSO). The solution was centrifuged, filtered and disinfected using a syringe filter (specification: 13 mm nylon filter, 0.45 µm, 100 pcs/pack), and preserved at − 20 ℃ for further experimentation.
Animal model and treatments
The study was approved by the Ethics Committee on Laboratory Animal Management of Guang’ anmen Hospital of China Academy of Chinese Medical Sciences. We abided by guidelines about animal use from the Association for Research in Vision and Ophthalmology (ARVO) throughout the study. Male 6-week-old Sprague–Dawley (SD) rats were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) and acclimated for 1 week. Animals were maintained under a 12-h light/12-h dark cycle at 22 °C in cages, and the litter was kept dry. The animals had free access to food and water. Diabetes was induced by an intraperitoneal injection of streptozotocin (65 mg/kg, in 10 mmol/l citrate buffer, pH 4.5) in SD rats. Age-matched control rats received an equal volume of vehicle. Two days after the streptozotocin (STZ) injection, rats with a blood glucose level higher than 250 mg/dl were considered as diabetic . The animals were divided into four groups as follows: (1) normal control rats (NC; n = 40), (2) STZ-induced diabetic rats (DM; n = 40); (3) STZ-induced diabetic rats treated with LTF (DM + LTF; 42 ml/kg, n = 40) and (4) STZ-induced diabetic rats treated with calcium dobesilate (DM + CaD; 104 mg/kg, n = 40). The doses of LTF and CaD were based on the daily dose commonly prescribed in humans.
Examination of blood glucose
Rat tail-vein blood glucose at the different time points was measured using OneTouch Ultra test strips (Shen Zhen, China). At 12 weeks, all rats were anesthetized and euthanized, and blood samples were collected. Blood serum glucose (n = 6/group) was measured using an automatic biochemical analyzer (OLYMPUS-640, Tokyo, Japan).
Light microscopic observations
Freshly removed eyeballs were fixed in 4% paraformaldehyde and embedded in paraffin. Sections were stained with hematoxylin–eosin (HE) and periodic acid-Schiff (PAS) staining for light microscopic observation. HE staining was used to analyze pathological changes in retinas and retinal digest preparations, and PAS staining was used to analyze pathological changes in retinal capillaries and for morphological quantification. Ten tangent retinal areas were randomly selected for observation, and images were taken and analyzed using Image-Pro Plus software (Media Cybernetics, America) to quantify endothelial cells and pericytes.
Rat retinal cell apoptosis was detected by TUNEL assay. Paraffin sections of retinas were dewaxed, diluted with distilled water, fixed in acetone for 8 min and washed three times with phosphate-buffered saline (PBS). The sections were blocked with 3% bovine serum albumin (BSA) for 30 min at room temperature. TUNEL reaction mixture (Roche, China) was added and incubated for 1 h at room temperature. Next, sections were washed three times with PBS, stained with DAPI for 1 min, and washed again. The mean optical densities of the sections were analyzed semiquantitatively with Image-Pro Plus analysis software (Media Cybernetics, America).
Paraffin-embedded sections of retinas were dewaxed, diluted with distilled water, and blocked with 3% hydrogen peroxide for 15 min. Sections were then fixed in citrate under high pressure for 3 min, then incubated with primary antibodies against insulin (1:100 dilution) and glucagon (1:200 dilution) for 24 h at 4 °C. The sections were washed three times with PBS and incubated with secondary antibodies (goat anti-rabbit, 1:200; goat anti-mouse, 1:200) for 2 h at room temperature. Color Development Kits were used, with hematoxylin as a mild counterstain, the sections were dehydrated in a graded series of ethanol concentrations and sealed. The mean optical density of the sections was analyzed semiquantitatively with Image-Pro Plus analysis software (Media Cybernetics).
After 12 weeks of treatment, rat retinas were harvested and washed with cold PBS, followed by incubation in RIPA buffer (Solarbio, China) on ice. Retinas were extracted and protein content was quantified using a BCA assay kit (Cwbiotech, China). After loading buffer was added, cell lysates were separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene fluoride (PVDF) membranes (Millipore, USA). Membranes were blocked using Super Block T20 (TBS) buffer and incubated with primary antibodies (Cell Signaling Technology, USA) overnight at 4 °C, then washed 3 times with TBST [tris-buffered saline (TBS) and Tween 20]. Dilution ratios: mouse monoclonal antibodies against ICAM-1 (1:1000), rabbit monoclonal antibodies against TNF-α (1:1000), rabbit polyclonal antibodies against NF-κB (1:1000), rabbit polyclonal antibodies against p38-MAPK (1:1000). The membrane was then incubated with HRP-conjugated secondary antibodies (Jackson Immuno Research, USA) for 1 h, at the following dilution ratios: goat anti-rabbit, 1: 5000; goat anti-mouse, 1: 5000. Enhanced chemiluminescence (Millipore, USA) was used to visualize the bands and an image analyzer (Bio-Rad, USA) was employed for densitometric analysis. Protein expression was normalized to that of GAPDH (Abcam, USA).
Real-time reverse transcription PCR assays (RT-PCR)
Primers sequences used for real-time PCR
The experiments were performed at least three times, and all values are expressed as mean ± standard deviation (SD). Data were evaluated using one-way analysis of variance (ANOVA), the post hoc Test (Bonferroni methods) was performed following ANOVA. SPSS program (version 20.0; IBM Corp., Armonk, NY, USA) was used for the analyses. Values of p < 0.05 were considered statistically significant.
HPLC analysis of LTF
Effects of LTF on body weight and blood glucose
At each time point, blood glucose levels in the DM group were dramatically higher than in the NC group (4 weeks: 30.18 ± 4.49 versus 5.20 ± 0.50; 8 weeks: 30.61 ± 3.03 versus 5.20 ± 0.58; 12 weeks: 31.38 ± 2.20 versus 5.23 ± 0.51, P < 0.05, respectively), and compared with the DM group, the blood glucose levels of LTF-treated rats was significantly decreased to 28.58 ± 4.53 at 4 week (P < 0.05), significant difference existed between LTF-treated and CaD-treated groups (28.58 ± 4.53 versus 31.03 ± 2.35, P < 0.05) at 4 weeks. In LTF and CaD group, they were comparable to those of DM rats at 8 and 12 weeks (P > 0.05), and no significant difference existed between the two groups (P > 0.05) (Fig. 2b).
Effects of LTF on the development of DR
In consistent with the diabetes-mediated alterations of retinal structure, the results of TUNEL staining indicated that retinal apoptosis-positive staining was much stronger in the DM than in the NC group, and the apoptosis-positive staining mainly assembled in the layers of ganglion cell and nerve fiber. After treatment with LTF and CaD, apoptotic staining was less pronounced than in DM retinas (Fig. 3c). Moreover, there are significantly reduced cell number of retinal ganglion cell layer (RGCL) and increased the numbers of TUNEL-positive cells in the RGCL of the diabetic retina (Retinal cell counts: 3.57 ± 1.50 versus 8.06 ± 1.32; TUNEL-positive cells: 36.46 ± 8.55 versus 0; P < 0.05, respectively) compared with to NC retina. Meantime, treatment with LTF was able to significantly prevent the reduction of retinal cells (5.45 ± 1.25 versus 3.57 ± 1.50, P < 0.05) and decrease retinal cell apoptosis (16.75 ± 6.84 versus 36.46 ± 8.55, P < 0.05) compared with that of the diabetic retina. CaD treatments also markedly attenuated the reduction of retinal cells and the increase of TUNEL-positive cells compared to that of diabetic (Retinal cell counts: 5.29 ± 1.63 versus 3.57 ± 1.50; TUNEL-positive cells: 15.44 ± 9.90 versus 36.46 ± 8.55; P < 0.05, respectively) (Fig. 3e, f). The proportion of retinal cell counts and TUNEL-positive cells in LTF-treated and CaD-treated groups indicated to be no significantly difference (P > 0.05). Taken together, the results suggested that administration of LTF to rats can correct the diabetes-induced retinal structural abnormalities and inhibit the development of DR.
Effects of LTF on the expression of inflammatory mediators
Effects of LTF on NF-κB (p65) and p38-MAPK pathways
Effects of LTF on panceratic insulin and glucagon
LTF, modified based on the Di Dang Decoction (Synopsis Golden Chamber), is a commonly used traditional Chinese medicine prescribed for treating diabetic microvascular complications [20, 21]. In this study, we showed that STZ-induced diabetes resulted in DR, and 12 weeks of LTF treatment ameliorated retinal damage, delayed DR occurrence, and significantly decreased the expression of TNF-α, IL-1β, MCP-1, and ICAM-1. The latter effect was probably associated with decreased retinal inflammation in the diabetic rats through modulation of the NF-κB(p65) and p38-MAPK pathways. In addition, 12 weeks of LTF treatment helped improve the pathological damage to pancreas, increased insulin expression in pancreatic beta cells, and downregulated the expression of glucagon in pancreatic α cells.
Pro-inflammatory cytokines/chemokines play an essential role in its pathogenesis. The published researches indicated that anti-inflammatory medication could delay early DR [4, 5, 6]. Huangqi, Danshen, Sanqi, Shuizhi powder, and Dahuang have been shown to inhibit the expression of pro-inflammatory mediators or chemokines such as IL-1β, IL-6, TNF-α, ICAM-1 and MCP-1, as well as signaling cascades such as the NF-κB pathway [18, 22, 23]. TNF-α is an important pro-inflammatory cytokine that is involved in numerous inflammatory pathologies. Upregulation of TNF-α has been previously detected in the retinas of diabetic rats and DR patients . Inhibition of TNF-α was demonstrate to suppress NF-κB activation, leukostasis, and BRB breakdown . Interleukin-1β has effects of increasing of vascular permeability, promoting the retinal cell apoptosis and leukocyte adhesion. IL-1β is mainly mediated by NF-κB activation, and increased expression of IL-1β may contribute to increased production of reactive oxygen species and further NF-κB activation . MCP-1 is a potent chemotactic factor for monocytes, macrophages, and T lymphocytes, which can help to promote the expression of superoxide and other mediators. MCP-1 can chemotactically aggregate monocytes, which leads to capillary obstruction, vascular leakage and absence of perfusion. Increased expression of MCP-1 is also implicated in the pathogenesis of DR . Here, we confirmed that LTF suppressed the expression of TNF-α, IL-1β, and MCP-1, which may contribute to the beneficial effects of LTF treatment. Adhesion of leukocytes to the retinal endothelium is increased in diabetic rats (known as leukostasis) and contributes to early DR, inducing the formation of acellular capillaries, vascular endothelial dysfunction, BRB breakdown, and increased vascular permeability. Our results indicated that LTF suppressed the expression of ICAM-1.
NF-κB activation can induce the release of some pro-inflammatory cytokines, such as IL-1β, IL-6, MCP-1 and TNF-α, which have been considered as central mediators of inflammatory processes, and NF-κB inhibition in diabetic retinas prevents the enhancement of ICAM-1, leukocyte adhesion and BRB leakage . In this study, we found that hyperglycemia induced the increase of NF-κB (p65) expression levels in rat retina, indicating that hyperglycemia can activate NF-κB pathway. After LTF treatment, the expression of NF-κB (p65) was decreased, which suggested that LTF might exert a protective role in rat retina through NF-κB pathway. Under hyperglycemic condition, p38MAPK is activated in the retina and involved in inflammatory response, endoplasmic reticulum stress and apoptosis. Hyperglycemia induced activation of the MAPK family, phosphorylation of related proteins and mediation of inflammatory response [28, 29]. Blocking p38-MAPK may be involved in alleviating inflammation. Previous studies have shown that herbal medicine promoting blood circulation and preventing vascular stasis relieves retinal pathological damage in diabetic mice, possibly as a consequence of reduced p38-MAPK phosphorylation . Consistently with this previous study, result indicated that LTF markedly downregulated the expression of p38-MAPK, suggested that the protective effect of LTF was related to the inhibition of p38-MAPK activation. NF-κB is closely related to p38-MAPK, both of which gather together in the downstream of p38-MAPK. Activated p38-MAPK activation could activate NF-κB by promoting the expression of inflammatory cytokines, which is an important downstream site of p38-MAPK signal pathway. If the expression of p38-MAPK is inhibited, the activation of NF-κB will also be affected . This study found that both p38MAPK and NF-κB pathways could be inhibited by LTF, suggesting that LTF play an anti-inflammatory role by inhibiting p38-MAPK and then inhibiting NF-κB pathway. But further studies are needed to confirm this conclusion.
In addition, LTF apparently promoted the expression of insulin while suppressing that of glucagon, suggesting a function in islet protection. However, this study was not able to conclusively prove this. The efficacy of Chinese herbal medicines, such as extracts, monomers, and compounds, in protecting islet beta cells has been gradually recognized, and may contribute to the prevention and treatment of DM and related complications. This protective action towards islet beta cells may result in a plethora of effects including increased cell proliferation, reduced apoptosis, attenuation of oxidative stress, promotion of insulin resistance, improvement of islet microcirculation, and regulation of autophagy, among others .
This study has several limitations. First, the molecular targets of LTF may not dependent on the NF-κB and/or the p38-MAPK signalling pathway, other mechanism should not exclude. We will continue to further explore the mechanism of LTF. Second, additional experiments in human retinal microvascular endothelial cells are needed to ascertain DR prevention by LTF. Third, leukocyte adhesion was not addressed in vitro.
In summary, our findings show that preventive treatment with LTF before the onset of DR can alleviate retinal pathological injury and decrease the expression of pro-inflammatory factors. LTF effects were correlated with the inhibition of NF-κB and p38-MAPK pathways. Additionally, preliminary evidence regarding the potential of LTF as an islet protective agent was provided.
QN, X-lT and BP conceived the study, and participated in its design together. JS and Q-wL participate the animal administration and observation. BP, JW and SD participated in immunohistochemistry work, Western blotting and RT-PCR study. ML performed the statistical analysis. BP wrote the initial manuscript, ML helped to revise it critically, both of them are co-first authors. QN and X-lT are both corresponding authors. All authors read and approved the final manuscript.
This study is supported by grants from National Natural Science Foundation of China (81774296); from Special program for excellent scientific personnel training of Chinese Academy of traditional Chinese Medicine (ZZ13-YQ-032); from Institutional Research Foundation of Guang’anmen Hospital, China Academy of Chinese Medical Science (59957).
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 8.Morigi M, Angioletti S, Imberti B, Donadelli R, Micheletti G, Figliuzzi M, Remuzzi A, Zoja C, Remuzzi G. Leukocyte-endothelial interaction is augmented by high glucose concentrations and hyperglycemia in a NF-kB-dependent fashion. J Clin Investig. 1998;101(9):1905–15. https://doi.org/10.1172/JCI656.CrossRefPubMedGoogle Scholar
- 12.Lu ZQ, Ding W. The role of islet beta cell failure in the pathogenesis of type 2 diabetes and the choice of anti-hyperglycemic drugs. Yao Xue Yu Lin Chuang Yan Jiu. 2012;20(5):385–91.Google Scholar
- 13.Zhou SP, Tong XL, Pan L, Gao WY. Effects of Luotong on the morphological changes of retinal blood vessels in diabetic rats. Zhong Guo Zhong Yi Yao Xin Xi Za Zhi. 2006;13(3):25–7.Google Scholar
- 14.Zhou SP, Tong XL, Pan L, Gao WY. Effect of Luotong on apoptosis of retinal microvascular cells in diabetic rats. Zhong Hua Zhong Yi Yao Za Zhi. 2006;21(5):273–5.Google Scholar
- 19.Leal EC, Martins J, Voabil P, Liberal J, Chiavaroli C, Bauer J, Cunha-Vaz J, Ambrósio AF. Calcium dobesilate inhibits the alterations in tight junction proteins and leukocyte adhesion to retinal endothelial cells induced by diabetes. Diabetes. 2010;59(10):2637–45. https://doi.org/10.2337/db09-1421.CrossRefPubMedPubMedCentralGoogle Scholar
- 22.Gao D, Guo Y, Li X, Li X, Li Z, Xue M, Ou Z, Liu M, Yang M, Liu S, Yang S. An aqueous extract of Radix Astragali, Angelica sinensis, and Panax notoginseng is effective in preventing diabetic retinopathy. Evid Based Complement. Altern Med. 2013;2013:578165. https://doi.org/10.1155/2013/578165.CrossRefGoogle Scholar
- 25.Huang H, Gandhi JK, Zhong X, Wei Y, Gong J, Duh EJ, Vinores SA. TNFalpha is required for late BRB breakdown in diabetic retinopathy, and its inhibition prevents leukostasis and protects vessels and neurons from apoptosis. Invest Ophthalmol Vis Sci. 2011;52(3):1336–44. https://doi.org/10.1167/iovs.10-5768.CrossRefPubMedPubMedCentralGoogle Scholar
- 26.Chang CK, LoCicero J III. Overexpressed nuclear factor kappaB correlates with enhanced expression of interleukin-1beta and inducible nitric oxide synthase in aged murine lungs to endotoxic stress. Ann Thorac Surg. 2004;77(4):1222–7. https://doi.org/10.1016/j.athoracsur.2003.09.128.CrossRefPubMedGoogle Scholar
- 30.Zhang Z, Wang C. Effects of Tongxinluo capsule on p38 mitogen-activated protein kinase signal pathway in retina of diabetic mice. He Bei Yi Yao. 2015;37(4):504–7.Google Scholar
- 31.Huang CP, Yang HP, Zhang ZX, Yang YJ. Role of p38 mitogen-activated protein kinase in the activation of nuclear factor-κB in lipopolysaccharide-stimulated alveolar macrophages. Acta Med Univ Sci Technol Huazhong. 2007;36(6):804–7.Google Scholar
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International 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://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.