Increasing studies showed that miR-200 family (miR-200s) clusters are aberrantly expressed in multiple human cancers, and miR-200s clusters function as tumor suppressor genes by affecting cell proliferation, self-renewal, differentiation, division and apoptosis. Herein, we aimed to investigate the expression and clinical implication of miR-200s clusters in acute myeloid leukemia (AML).
Methods
RT-qPCR was performed to detect expression of miR-200s clusters in 19 healthy donors, 98 newly diagnosed AML patients, and 35 AML patients achieved complete remission (CR).
Results
Expression of miR-200a/200b/429 cluster but not miR-200c/141 cluster was decreased in newly diagnosed AML patients as compared to healthy donors and AML patients achieved CR. Although no significant differences were observed between miR-200s clusters and most of the features, low expression of miR-200s clusters seems to be associated with higher white blood cells especially for miR-200a/200b. Of the five members of miR-200s clusters, low expression of miR-200b/429/200c was found to be associated with lower CR rate. Logistic regression analysis further revealed that low expression of miR-429 acted as an independent risk factor for CR in AML. Based on Kaplan–Meier analysis, low expression of miR-200b/429/200c was associated with shorter OS, whereas miR-200a/141 had a trend. Moreover, multivariate analysis of Cox regression models confirmed the independently prognostic value of miR-200b expression for OS in AML.
Conclusions
Expression of miR-200a/200b/429 cluster was frequently down-regulated in AML, and low expression of miR-429 as an independent risk factor for CR, whereas low expression of miR-200b as an independent prognostic biomarker for OS.
Jing-dong Zhou and Liu-chao Zhang contributed equally to this work
Abkürzungen
AML
acute myeloid leukemia
3′-UTR
3′-untranslated region
APL
acute promyelocytic leukemia
BM
bone marrow
CR
complete remission
BMMNCs
BM mononuclear cells
FAB
French–American–British
WHO
World Health Organization
RT-qPCR
real-time quantitative PCR
OS
overall survival
WBC
white blood cell
Background
Acute myeloid leukemia (AML) is a highly heterogeneous malignant hematological disorder with complex molecular pathophysiology. Although the treatment strategies against AML have been updated in the past decades, the majority of patients eventually succumb to relapse after induction chemotherapy [1]. Clinical outcome of AML remains unsatisfactory especially in those with specific karyotypes/biomarkers such as inv(3)(q21q26.2), t(6;9)(p23; q34), 11q abnormalities other than t(9;11), -5/del(5q), -7, TP53 mutations, FLT3-ITD mutations, C-KIT mutations, WT1 overexpression, and BAALC overexpression [2‐4]. The development of effective therapeutic options against AML relies on mechanistic understanding of AML biology, especially in molecular regulators of AML pathogenesis and molecular predictor of AML prognosis [5].
MicroRNAs, a class of small (19–22 nucleotides) single-stranded RNAs, negatively regulate various genes by targeting 3′-untranslated region (3′-UTR) of mRNAs, thereby facilitating translational silencing or degradation of targeted genes [6]. Mounting evidences have implicated that microRNAs play crucial roles in regulating many fundamental and biological processes including cancer development [7]. Moreover, microRNAs have been reported as novel biomarkers for diagnosis and prognosis, and regarded as potential therapeutic targets in AML [8]. For instance, recent studies implicated that several microRNAs such as miR-216b, miR-362-5p, miR-217, and miR-193b were prognosis-related predictors in AML and may involve in AML biology [9‐12].
Anzeige
The miR-200 family (miR-200s) clusters includes five members (miR-200a, miR-200b, miR-200c, miR-141, and miR-429) and can be divided into two clusters (miR-200a/b/429 cluster and miR-200c/141 cluster) based on chromosomal location (chromosome 1p36 and chromosome 12p13) [13]. Numerous studies showed that miR-200s clusters are aberrantly expressed in multiple human cancers, and miR-200s clusters function as tumor suppressor genes by affecting cell proliferation, self-renewal, differentiation, division and apoptosis [14]. Although the tumor-suppressive roles of miR-200s clusters have also been reported in solid tumors with prognostic value [14, 15], the expression and clinical implication of miR-200s clusters in AML remains poorly revealed.
In this study, we investigated expression of miR-200s clusters in AML patients except for acute promyelocytic leukemia (APL), and found that low expression of miR-200s clusters acted as potential prognostic biomarkers in AML.
Methods
Patients and treatment
A total of 98 de novo AML patients except for APL and 19 healthy donors were enrolled in this study. Bone marrow (BM) was collected from all the patients at diagnosis time as well as 35 patients at complete remission (CR) time. AML was diagnosed based on the French–American–British (FAB) and 2016 revised World Health Organization (WHO) criteria [16, 17]. All the patients received chemotherapy as reported [18]. Induction chemotherapy therapy was 1–2 courses of daunorubicin combined with cytarabine. Subsequent consolidation treatment after CR for younger patients included high-dose cytarabine, mitoxantrone with cytarabine, and homoharringtonine combined with cytarabine, whereas for older patients received in an individualized manner decided by the physicians, such as CHG protocol (cytarabine, homoharringtonine, and G-CSF). This study was approved by the Ethics Committee of the Affiliated People’s Hospital of Jiangsu University, and written informed consents were informed and signed by all participants in accordance with the Declaration of Helsinki Principles.
Cytogenetic analysis and mutation detection
BM cells were harvested after 1–3 days of unstimulated culture in RPMI 1640 medium (BOSTER, Wuhan, China) containing 20% fetal calf serum (ExCell Bio, Shanghai, China). Cytogenetics for AML patients were analyzed at the newly diagnosis time by conventional R-banding method and karyotype risk was classified according to reported previously [19, 20]. Hotspot mutations in NPM1, C-KIT, DNMT3A, N/K-RAS, IDH1/2, U2AF1, SRSF2 and SETBP1 were detected by high-resolution melting analysis [21‐25], whereas mutations in FLT3-ITD and CEBPA were examined by DNA sequencing [26].
Anzeige
RNA isolation and reverse transcription
BM mononuclear cells (BMMNCs) were extracted as reported using Lymphocyte Separation Medium (Absin, Shanghai, China) [27]. According to the manufacturer’s protocols, RNA was extracted from BMMNCs using the mirVana miRNA isolation kit (Ambion, Austin, TX, USA), and was synthesized to cDNA by reverse transcription using MiScript Reverse Transcription Kit (Qiagen, Duesseldorf, Germany).
Real-time quantitative PCR
The level of miR-200s clusters was detected by real-time quantitative PCR (RT-qPCR) using miScript SYBR green PCR kit (Qiagen, Duesseldorf, Germany). The primers were miR-200s specific (Additional file 1: Table S1) and the manufacturer-provided miScript universal primer (Qiagen, Duesseldorf, Germany). The programs for RT-qPCR reactions were performed as reported [28]. U6 small nuclear RNA was selected as the endogenous normalizer detected by RT-qPCR using 2× SYBR Green PCR Mix (Multisciences, Hangzhou, China). Relative miR-200s level was calculated by 2−ΔΔCT method. The healthy donors that possessed the minimal ΔCT between miR-200s (each member) and U6 expression was selected as control, and was defined as 100% expression.
Statistical analysis
Mann–Whitney’s U test was carried to compare the difference of continuous variables between two groups, whereas Pearson Chi square analysis/Fisher exact test were applied to compare the difference of categorical variables between two groups. The impact of miR-200s clusters expression on overall survival (OS) was analyzed by Kaplan–Meier analysis, and Cox regression models (univariate and multivariate analyses) were further used to determine the independently prognostic value of miR-200s cluster expression. The effect of miR-200s clusters expression on CR was determined by Logistic regression analysis (univariate and multivariate analyses). All tests were two sided, and P < 0.05 was defined as statistically significant. SPSS software 20.0 and GraphPad Prism 5.0 was used to conduct the statistical analyses in this study.
Results
Expression of miR-200s in AML
We analyzed miR-200s clusters expression in BM from 19 healthy donors, 98 AML patients, and 35 AML patients achieved CR by RT-qPCR. As presented in Fig. 1, expression of miR-200a/200b/429 clusters but not miR-200c/141 clusters was significantly decreased in AML patients as compared to healthy donors and AML patients achieved CR.
Fig. 1
Expression of miR-200s clusters in controls, newly diagnosed AML patients and AML patients achieved CR. a For miR-200a; b For miR-200b; c For miR-429; d For miR-200c; e For miR-141. The distributions of the miR-200s clusters expression in controls, newly diagnosed AML patients and AML patients achieved CR were presented with scatter plots. The median level of miR-200s clusters expression in each group was shown with horizontal line
×
Relationship between miR-200s and clinical features in AML
To investigate clinical implication of miR-200s clusters expression, the whole-cohort patients were classified into two groups (high and low miR-200s clusters expression) based on the median level of each member of miR-200s clusters, respectively. We analyzed the association between each member of miR-200s clusters expression and clinic-pathologic features including gender, age, white blood cell (WBC) counts, hemoglobin content, platelet counts, blasts (%), FAB subtypes, karyotypes, and common gene mutations. As shown in Table 1, no significant differences were observed between miR-200s clusters expression and most of the features. However, low expression of miR-200s clusters seems to be associated with higher WBC counts especially for miR-200a/200b (P = 0.001 and 0.041, respectively). In addition, low expression of miR-200a was related to male, whereas low expression of miR-141 was correlated with higher hemoglobin content (P = 0.013 and 0.024, respectively).
Table 1
Correlation of miR-200s cluster expression with clinical/laboratory features in AML patients
Patient’s features
miR-200a expression
miR-200b expression
miR-429 expression
miR-200c expression
miR-141 expression
Low (n = 49)
High (n = 49)
P
Low (n = 49)
High (n = 49)
P
Low (n = 49)
High (n = 49)
P
Low (n = 49)
High (n = 49)
P
Low (n = 49)
High (n = 49)
P
Sex (male/female)
36/13
23/26
0.013
32/17
27/22
0.409
31/18
28/21
0.680
32/17
27/22
0.409
31/18
28/21
0.680
Age (years)
58 (21–81)
61 (18–87)
0.842
60 (18–81)
59 (18–87)
0.831
60 (21–81)
59 (18–87)
0.741
60 (18–81)
59 (18–87)
0.743
60 (18–81)
59 (18–87)
1.000
WBC (× 109/L)
38.7 (1.3–528.0)
9.6 (1.1–130.2)
0.001
35.5 (1.3–528.0)
9.6 (1.1–130.2)
0.041
34.5 (1.3–528.0)
13.3 (1.1–130.2)
0.099
34.9 (1.3–528.0)
13.2 (1.1–130.2)
0.094
35.5 (1.3–528.0)
10.2 (1.1–116.6)
0.090
Hemoglobin (g/L)
78 (53–138)
76.5 (32–144)
0.085
80 (53–138)
76.5 (32–144)
0.330
77 (32–138)
78 (34–134)
0.940
84 (53–138)
76 (32–144)
0.124
88 (53–138)
74.5 (32–144)
0.024
Platelets (× 109/L)
37 (3–447)
47 (4–264)
0.558
37 (3–447)
46 (4–264)
0.460
36 (3–447)
47 (4–264)
0.512
30 (5–447)
49 (3–264)
0.233
40 (5–125)
46 (3–447)
0.769
BM blasts (%)
60% (20–99%)
58% (20–95%)
0.892
63% (20–99%)
58% (21–95%)
0.651
60% (20–98%)
59% (20–99%)
0.757
63% (20–99%)
54% (20–95%)
0.226
62% (20–98%)
57% (20–99%)
0.598
FAB subtypes
0.660
0.945
0.681
0.827
0.900
M0
1
0
1
0
0
1
1
0
1
0
M1
4
2
3
3
3
3
2
4
2
4
M2
26
24
23
27
22
28
23
27
24
26
M4
12
15
14
13
15
12
15
12
15
12
M5
6
6
7
5
8
4
7
5
6
6
M6
0
2
1
1
1
1
1
1
1
1
Karyotypes
0.982
0.813
0.606
0.948
0.578
Normal
25
28
23
30
23
30
24
29
24
29
t(8;21)
5
5
5
5
6
4
5
5
6
4
+ 8
2
1
2
1
2
1
2
1
3
0
− 5/5q-
1
2
2
1
2
1
2
1
2
1
− 7/7q-
1
0
1
0
0
1
1
0
1
0
t(9;22)
1
0
1
0
0
1
1
0
1
0
Complex
8
7
9
6
10
5
8
7
7
8
Others
6
5
6
5
6
5
6
5
5
6
No data
0
1
0
1
0
1
0
1
0
1
Gene mutations
CEBPA (±)
7/33
6/38
0.765
5/38
8/33
0.376
8/36
5/35
0.555
7/34
6/37
0.768
6/37
7/34
0.768
NPM1 (±)
5/35
4/40
0.730
3/40
6/35
0.307
4/40
5/35
0.730
5/36
4/39
0.735
4/39
5/36
0.735
FLT3-ITD (±)
6/34
3/41
0.298
6/37
3/38
0.484
6/38
3/37
0.488
5/36
4/39
0.735
6/37
3/38
0.484
C-KIT (±)
1/39
1/43
1.000
1/42
1/40
1.000
1/43
1/39
1.000
1/40
1/42
1.000
1/42
1/40
1.000
N/K-RAS (±)
4/36
6/38
0.741
3/40
7/34
0.190
5/39
5/35
1.000
6/35
4/39
0.515
6/37
4/37
0.739
IDH1/2 (±)
1/39
3/41
0.618
3/40
1/40
0.616
4/40
0/40
0.118
4/37
0/43
0.052
2/41
2/39
1.000
DNMT3A (±)
4/36
2/42
0.418
4/39
2/39
0.676
4/40
2/38
0.678
3/38
3/40
1.000
3/40
3/38
1.000
U2AF1 (±)
2/38
2/42
1.000
1/42
3/38
0.354
1/43
3/37
0.343
2/39
2/41
1.000
3/40
1/40
0.616
SRSF2 (±)
1/39
3/41
0.618
1/42
3/38
0.354
1/43
3/37
0.343
1/40
3/40
0.616
1/42
3/38
0.354
SETBP1 (±)
2/38
0/44
0.224
2/41
0/41
0.494
2/42
0/40
0.495
2/39
0/43
0.235
2/41
0/41
0.494
CR (±)
16/33
23/26
0.215
14/35
25/24
0.038
14/35
25/24
0.038
14/35
25/24
0.038
15/34
24/25
0.098
WBC white blood cells, BM bone marrow, FAB French–American–British classification, CR complete remission
Prognostic value of miR-200s in AML
To observe the impact of miR-200s clusters expression on clinical outcome in AML, we first determined the association of each member of miR-200s clusters expression with CR. Of the five members of miR-200s clusters, low expression of miR-200b/429/200c was found to be associated with lower CR rate (Table 1, all P = 0.038). Additionally, Logistic regression analysis was further performed to confirm and verify the effect of miR-200s clusters’ expression on CR, and revealed low expression of miR-429 as an independent risk factor for CR in AML (Table 2, P = 0.023).
Table 2
Univariate and multivariate analyses of variables for overall survival in AML patients
Variables
Complete remission
Overall survival
Univariate analysis
Multivariate analysis
Univariate analysis
Multivariate analysis
OR (95% CI)
P
OR (95% CI)
P
HR (95% CI)
P
HR (95% CI)
P
miR-200a
0.548 (0.242–1.244)
0.150
1.029 (0.285–3.722)
0.965
0.662 (0.415–1.022)
0.082
1.425 (0.651–3.120)
0.376
miR-200b
0.384 (0.167–0.885)
0.025
0.823 (0.199–3.401)
0.788
0.511 (0.319–0.819)
0.005
0.524 (0.305–0.902)
0.020
miR-429
0.384 (0.167–0.885)
0.025
0.331 (0.128–0.858)
0.023
0.558 (0.350–0.891)
0.015
0.820 (0.325–2.073)
0.675
miR-200c
0.384 (0.167–0.885)
0.025
0.977 (0.149–6.400)
0.981
0.606 (0.380–0.965)
0.035
0.649 (0.190–2.217)
0.491
miR-141
0.460 (0.201–1.050)
0.065
0.594 (0.192–1.833)
0.364
0.695 (0.437–1.104)
0.123
1.152 (0.582–2.279)
0.684
Age
4.229 (1.742–10.266)
0.001
4.555 (1.715–12.095)
0.002
2.046 (1.282–3.266)
0.003
1.732 (1.033–2.902)
0.037
WBC
2.367 (1.015–5.520)
0.046
1.846 (0.715–4.767)
0.206
2.002 (1.253–3.199)
0.004
1.560 (0.925–2.629)
0.095
Karyotype
3.108 (1.338–7.220)
0.008
2.862 (1.164–7.042)
0.022
1.875 (1.295–2.715)
0.001
1.874 (1.210–2.902)
0.005
CEBPA mutations
0.526 (0.160–1.731)
0.290
0.870 (0.413–1.829)
0.713
NPM1 mutations
0.833 (0.207–3.358)
0.798
1.200 (0.516–2.793)
0.672
FLT3-ITD mutations
0.833 (0.207–3.358)
0.798
0.935 (0.403–2.170)
0.876
C-KIT mutations
0.673 (0.041–11.150)
0.783
0.479 (0.066–3.458)
0.465
N/K-RAS mutations
3.048 (0.605–15.343)
0.177
1.311 (0.621–2.770)
0.478
IDH1/2 mutations
Undetermined
0.999
4.671 (1.637–13.326)
0.004
6.662 (1.757–25.268)
0.005
DNMT3A mutations
1.391 (0.240–8.057)
0.712
1.590 (0.634–3.987)
0.323
U2AF1 mutations
Undetermined
0.999
2.791 (0.987–7.890)
0.053
5.130 (1.714–15.355)
0.003
SRSF2 mutations
Undetermined
0.999
1.934 (0.693–5.400)
0.208
SETBP1 mutations
0.673 (0.041–11.150)
0.783
0.637 (0.088–4.613)
0.656
OR odd ratio, HR hazard ratio, CI confidence interval. Variables including miR-200s cluster expression (Low vs. High), age (≤ 60 vs. > 60 years), WBC (≥ 30 × 109 vs. < 30 × 109/L), karyotype (favorable vs. intermediate vs. poor), and gene mutations (mutant vs. wild-type). Multivariate analysis includes variables with P < 0.200 in univariate analysis
We next evaluated the correlation of each member of miR-200s clusters expression with survival. Based on Kaplan–Meier analysis, low expression of miR-200b/429/200c was associated with shorter OS, whereas miR-200a/141 had a trend (Fig. 2). In addition, we also analyzed the impact of composite members of miR-200s clusters expression on OS by Kaplan–Meier analysis as shown in Fig. 3.
Fig. 2
Prognostic value of each member of miR-200s clusters expression in AML. a For miR-200a. b For miR-200b. c For miR-429. d For miR-200c. e For miR-141. Overall survival analyzed between two groups based on median level of each member of miR-200s clusters, and performed by Kaplan–Meier methods
Fig. 3
Prognostic value of composite members of miR-200s clusters expression in AML. A Overall survival (OS) analyzed among AML patients with different numbers of abnormal miR-200s clusters expression; a: without miR-200s clusters abnormalities; b: with only one member of miR-200s clusters abnormalities; c: with two members of miR-200s clusters abnormalities; d: with three members of miR-200s clusters abnormalities; e: with four members of miR-200s clusters abnormalities; f: with all members of miR-200s clusters abnormalities. B OS analyzed between two groups (all members of miR-200s clusters abnormalities vs. equal or less than four members of miR-200s clusters abnormalities). C OS analyzed between two groups (equal or more than four members of miR-200s clusters abnormalities vs. equal or less than three members of miR-200s clusters abnormalities). D OS analyzed between two groups (equal or more than three members of miR-200s clusters abnormalities vs. equal or less than two members of miR-200s clusters abnormalities). E OS analyzed between two groups (equal or more than two members of miR-200s clusters abnormalities vs. equal or less than one member of miR-200s clusters abnormalities). F OS analyzed between two groups (equal or more than one member of miR-200s clusters abnormalities vs. without any members of miR-200s clusters abnormalities)
×
×
Since miR-200s clusters expression was associated with well-established prognostic factor such as WBC counts, we further conducted a Cox regression model adjusting for prognosis-related factors (age, WBC counts, karyotypic classifications, and gene mutations) for OS. Results showed that low expression of miR-200b acted as an independent prognostic biomarker for OS (P = 0.020, Table 2).
Anzeige
Discussion
In the current study, we for the first time investigated expression of miR-200s clusters in AML, and revealed that most of the members of miR-200s clusters were down-regulated in de novo AML patients. Recently, Li et al. revealed that introduction of a pre-miR-200c reduced the expression of ZEB2 protein and inhibited the proliferation of human leukemia cell lines (HL-60, MOLM-13, and THP-1), and mouse miR-200c significantly impaired the proliferation of mouse leukemia cells [29]. Taken together, these results emphasized the crucial role of miR-200s clusters in leukemogenesis. Although the biological role of miR-200s clusters in AML was less studied, tumor suppressor roles of miR-200s clusters have been identified in a variety of human solid cancers, such as bladder cancer, gastric cancer, colorectal cancer, breast cancer, ovarian cancer, endometrial cancer, pancreatic cancer, gliomas, hepatocellular carcinoma, and lung cancer [14, 30]. The miR-200s clusters were reported as key inhibitors of epithelial-to-mesenchymal transition by directly targeting transcriptional repressors of E-cadherin, ZEB1, and ZEB2 [13]. Moreover, miR-200s clusters also played crucial roles in the repression of cancer stem cells self-renewal and differentiation, modulation of cell division and apoptosis, and reversal of chemoresistance [14, 30]. Notably, in some other hematological malignancies, expression or biological role of miR-200s clusters has been preliminary studied. For instance, Choi et al. reported that miR-200c was decreased in patients with myelodysplastic syndrome (MDS) [31]. González-Gugel et al. revealed that down-regulation of mmu-miR-30a and mmu-miR-141 as well as hsa-miR-193b clearly contributed to enhance the expression of Smoothened (SMO) gene in mouse and human lymphomas and, subsequently, to activate the GLI/Hh signalling [32].
In addition to basic research before, it has been noted that low expression of miR-200s clusters could correlate with adverse clinical outcome and serve as a prognostic biomarker for various cancer patients [15]. Although the potential prognostic value of miR-200s clusters in several human cancers remains controversial, a recent meta-analysis demonstrated that lower tissue expression of miR-200s clusters’ members were associated with poor OS and progression-free survival, whereas lower expression of circulating miR-200s clusters’ members were correlated with favorable prognosis [15]. From our study, we showed the negative effect of low expression of miR-200s clusters on AML chemotherapy response and survival. Moreover, multivariate analysis showed that low expression of miR-429 as an independent risk factor for CR, whereas low expression of miR-200b as an independent prognostic biomarker for OS in AML. Due to some limitations in this study (such as patients numbers, treatment regimens, and single center), prospective studies are needed to verify our results before miR-200s clusters expression could be used routinely as a promising biomarker for risk stratification in AML.
Conclusion
Expression of miR-200a/200b/429 cluster was frequently down-regulated in AML, and low expression of miR-429 as an independent risk factor for CR, whereas low expression of miR-200b as an independent prognostic biomarker for OS.
Authors’ contributions
JQ and JL conceived and designed the experiments; JZ and LZ performed the experiments; JZ and TZ analyzed the data; YG, WZ and DW collected the clinical data; JM, XW and HG offered technique support; JZ wrote the paper. All authors read and approved the final manuscript.
Anzeige
Acknowledgements
None.
Competing interests
The authors declare that they have no competing interests.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Consent for publication
Written informed consents were obtained from all enrolled individuals prior to their participation.
Anzeige
Ethical approval and consent to participate
The present study approved by the Ethics Committee and Institutional Review Board of the Affiliated People’s Hospital of Jiangsu University.
Funding
This work was supported by National Natural Science foundation of China (81270630), Medical Innovation Team of Jiangsu Province (CXTDB2017002), 333 Project of Jiangsu Province (BRA2016131), Six Talent Peaks Project in Jiangsu Province (2015-WSN-115), Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX17_1821), Youth Medical Talents Project of “Ke Jiao Qiang Wei” project of Jiangsu province (QNRC2016450, QNRC2016449), China Postdoctoral Science Foundation funded project (2016M601748), Clinical Medical Science and Development Foundation of Jiangsu University (JLY20160011), Social Development Foundation of Kunshan (KS1624), Social Development Foundation of Zhenjiang (SH2016045, SH2016046, SH2017040), Key Medical Talent Program of Zhenjiang City.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 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.
