CDA gene silencing regulated the proliferation and apoptosis of chronic myeloid leukemia K562 cells
As a disease of hematopoietic stem cell, chronic myeloid leukemia (CML) possesses unique biological and clinical features. However, the biologic mechanism underlying its development remains poorly understood. Thus, the objective of the present study is to discuss the effect of cytidine deaminase (CDA) gene silencing on the apoptosis and proliferation of CML K562 cells.
CDA mRNA expression was detected by reverse transcription-quantitative polymerase chain reaction (RT-qPCR), and enzymatic activity of CDA was measured by a nuclide liquid scintillation method. RT-qPCR and Western blot analysis were used to detect CDA mRNA and protein expression. Cell proliferation, apoptosis and cell cycle were measured by CCK-8 assay and flow cytometry. The expression of proteins relevant to cell proliferation, apoptosis and cell cycle was measured by Western blot analysis. Tumor xenografts were implanted in nude mice to verify the effect of CDA silencing on tumor growth in vivo.
CML and AL patients showed increased mRNA expression and enzymatic activity of CDA. Compared with the blank group, the mRNA and protein expression of CDA in the shRNA-1 and shRNA-2 groups decreased significantly. As a result, the proliferation of K562 cells was inhibited after CDA silencing and the cells were mainly arrested in S and G2 phases, while the apoptosis rate of these cells was increased. In addition, CDA gene silencing in K562 cells led to down-regulated p-ERK1/2, t-AKT, p-AKT and BCL-2 expression and up-regulated expression of P21, Bax, cleaved caspase-3/total caspase-3 and cleaved PARP/total PARP. Finally, CDA gene silencing inhibited tumor growth.
Our study demonstrated that CDA gene silencing could inhibit CML cell proliferation and induce cell apoptosis. Therefore, CDA gene silencing may become an effective target for the treatment of leukemia.
KeywordsCytidine deaminase Chronic myeloid leukemia K562 cells Proliferation Apoptosis Gene silencing
chronic myeloid leukemia
reverse transcription-quantitative polymerase chain reaction
corrected counts per minute
Chronic myeloid leukemia (CML) is a hematopoietic malignant disease characterized by the presence of the Philadelphia chromosome (Ph) due to the reciprocal translocation of chromosomes 9 and 22, resulting in the formation of a BCR–ABL fusion onco-protein [1, 2, 3]. As a myelo-proliferative neoplasm, the incidence of CML is higher than 2/100,000 . In addition, the incidence of CML in females is lower than that in males . The BCR–ABL fusion protein leads to the pathogenicity of CML, while the therapies targeting the BCR–ABL protein have encountered many difficulties . Besides, based on the characteristics related to CML, the prognostic scoring systems of CML tend to classify patients into different risk groups . Currently, the tyrosine kinase inhibitor treatment has been regarded as the fundamental therapy for CML, while imatinib and dasatinib or nilotinib have been used in first-line and second-line treatments, respectively . In addition, gene therapy has become more common recently in the treatment of CML. For example, RNA interference or gene silencing has been used as new strategies for CML treatment .
Cytidine deaminase (CDA) is a leading enzyme that participates in the metabolism of gemcitabine, a pyrimidine analog used for chemotherapy in treating tumors such as biliary tract cancer [10, 11]. According to the crystal structures and possible amino acid sequences of CDA in different species, CDA has been divided into two types, i.e., a homodimeric class and a homotetrameric class . CDA expression is used to classify new subclass of cancers. For anti-tumor drugs, loss of CDA expression seems to act as a new predictive marker of cancer susceptibility . In terms of RNAi-mediated gene silencing, undergoing clinical trials have tried to use it as an alternative to traditional chemotherapies in cancer treatment . In a related study, cytarabine was found to have high efficacy in the treatment of acute myeloid leukemia and its efficacy was delaminated by CDA . Nevertheless, the underlying mechanism of CDA gene silencing in the biological processes of K562 cells remains to be clarified. Therefore, in this study, we aimed to investigate the effects of CDA gene silencing of the apoptosis and proliferation of CML K562 cells.
Materials and methods
Between January 2012 and January 2015, all leukemia patients diagnosed in the hematology department of Gansu Provincial Hospital were screened. Eventually, 112 CML patients were enrolled in this study, including 70 males and 42 females with a mean age of 34 years (range from 18 to 47 years). This study also enrolled 106 acute leukemia (AL) patients, including 62 males and 44 females with a mean age of 32 years (range from 19 to 48 years), as well as 101 donors for hematopoietic stem cell transplantation, including 53 males and 48 females with a mean age of 32 years (range from 18 to 43 years). All leukemia patients were diagnosed based on MICM (morphology, immunology, cytogenetics, molecular biology and biochemistry) . Bone marrow mononuclear cells were separated using a lymphocyte separation medium (Ficoll separating liquid) (Thermo Fisher Scientific, San Jose, California, USA), while collected cell clusters were used immediately or preserved at − 80 °C for future use. This study was approved by the ethics committee of Gansu Provincial Hospital, and all patients had signed informed consent.
Reveres transcription-quantitative polymerase chain reaction (RT-qPCR)
The total RNA in collected cell samples was extracted using a Trizol-based one-step method according to the instructions of Trizol reagent (Invitrogen Inc., Carlsbad, CA, USA). RNA was dissolved in ultra-pure water pre-treated by diethylpyrocarbonate (DECP). RNA quality was evaluated by measuring its absorbance at 260 and 280 nm on an ND-1000 ultraviolet/visible spectroscopy (UV/VIS) spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Subsequently, the concentration of RNA was adjusted for RT-qPCR. The extracted RNA was reversely transcribed using a two-step method according to the instructions of a reagent kit (Fermentas Inc., Hanover, MD, USA). The reaction conditions of RT were as follows: 70 °C for 10 min, ice bathed for 2 min, 42 °C for 60 min, and 70 °C for 10 min. The obtained cDNA was stored in a − 80 °C freezer. The TaqMan probe method was used in qPCR, and the reaction system was prepared according to the instructions of a reagent kit (Fermentas Inc., Hanover, MD, USA). The upstream and downstream sequences of CDA primers were 5′-CCGTCTCAGAAGGGTACAAG-3′ and 5′-GACAATATACGTACCATCCGG-3′. The reaction conditions of qPCR were as follows: pre-denaturation at 95 °C for 30 s, and 40 cycles of denaturation at 95 °C for 10 s, annealing at 60 °C for 20 s, and extension at 70 °C for 10 s. An RT-qPCR instrument (American Bio-Rad company, model, is Bio-Rad iQ5) was used to carry out RT-qPCR, while glyceraldehyde-3-phosphate dehydrogenase (GAPDH, upstream sequence: 5′-ATTCAACGGCACAGTCAAGG-3′, downstream sequence: 5′-GCAGAAGGGGCGGAGATGA-3′) was used as an internal reference. The relative expression of CDA was calculated using 2−∆∆Ct, and each experiment was repeated 3 times.
Nuclide liquid scintillation
After thawing, the sample cells were rinsed twice with phosphate buffered saline (PBS). Subsequently, the cells (1 × 107) were placed into frozen tubes, followed by the addition of 1 ml CDA buffer solution (pH 7.50, 50 mmol/l Tris–HCl, 2 mmol/l DTT, and 10% glycerol) (Beijing Solarbio Science & Technology Co., Ltd). The cell membranes were ruptured after three freeze–thaw cycles in liquid nitrogen, after which the cells were centrifuged at 1500g and 4 °C for 10 min. Subsequently, 0.1 ml 3H-cytidine (0.03 μ Ci) (Hoffmann-La Roche Ltd., Basel, Switzerland) was added into 0.5 ml supernatant and incubated at 37 °C for 2 h. In the next step, the mixture was dripped slowly into a resin column and slowly chased with 20 ml distilled water. Afterwards, 3H-uridine was collected into a disinfected tube and mixed with 3 ml liquid scintillation cocktail. The value of corrected counts per minute (CCPM) was then measured on a liquid scintillation counter.
CML K526 cells were purchased from Shanghai cell bank of the Chinese Academy of Sciences and maintained in a Roswell Park Memorial Institute (RPMI)-1640 (Gibco Company Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Hyclone Company, Logan, UT, USA). During this study, the cells were cultured in a 5% CO2 incubator (Thermo Fisher Scientific, San Jose, California, USA. model, Thermo Scientific 8000) at 37 °C and 95% humidity. During cell passage, the original medium was removed through centrifugation at 1000 rpm and the cells were rinsed twice with PBS before adding a new RPMI-1640 medium (containing 10% FBS).
Construction of CDA gene silencing and CDA overexpression models
Interfering RNA (siRNA) sequences for CDA gene
Cell transfection and grouping
The cells were assigned into the following groups: a blank group, a shRNA-con group, a shRNA-1 group, a shRNA-2 group, and a group of over-expressed CDA. K526 cells in the logarithmic growth phase were selected for transfection. During transfection, shRNA plasmids (at a final concentration of 50 nM) were diluted in 250 µl serum-free Opti-MEM medium (Gibco Company, Grand Island, NY, USA), gently mixed and incubated at room temperature for 5 min. At the same time, 5 µl Lipofectamine 2000 were diluted in 250 µl serum-free Opti-MEM medium, gently mixed and incubated at room temperature for 5 min. The above two solutions were then mixed, incubated at room temperature for 20 min and added onto the cells. After 24–48 h transfection, the cells were collected for further experiments.
Cell counting kit-8 (CCK-8) assay
After 24 h of transfection, cells were centrifuged to remove the original medium, rinsed twice with PBS and made into a single cell suspension. After counting, the cells were seeded into a 96-well plate in 100 μl/well medium at a density of 3–6 × 103 cells per well. Each cell treatment was tested in triplicate. During the assay, each well was added with 10 µl of CCK-8 solution and cultured for 4–6 h (American Sigma Company). The optical density (OD) of each well was detected at 450 nm using an enzyme-linked immuno-sorbent assay (ELISA) after 24, 48 and 72 h of incubation. Cell viability curves were drawn using time as the abscissa and survival rate (%) as the ordinate.
The cells were detached with trypsin, suspended and counted. Afterwards, the cells were seeded into a 6-well plate at a density of 1000 cells/well, and cultured in a semi-fixed medium under 5% CO2 and 37 °C. After 2 weeks, the cells were stained with crystal violet, and the number and size of cell colonies were observed. The experiment was repeated 3 times.
Detection of cell cycle: after 48 h of transfection, the cells were collected, rinsed 3 times with ice-cold PBS and centrifuged to remove the supernatant. The concentration of the cells was adjusted to approximately 1 × 105/ml. Subsequently, the cells were fixed in 1 ml ice-cold 75% ethanol at 4 °C overnight. Before staining, the cells were rinsed twice with PBS, added into 100 μl RNaseA and incubated at 37 °C for 30 min in the dark. Subsequently, the cells were stained with 400 μl PI (Sigma-Aldrich Chemical Company, St Louis MO, USA) at 4 °C for 30 min in the dark. Cell cycle was detected by flow cytometry (American BD Biosciences Company. Model, FACSCanto II) at 488 nm.
Detection of cell apoptosis: after 48 h of transfection, the cells were collected into flow tubes and centrifuged at 178 g for 5 min. Subsequently, the cells were rinsed 3 times with ice-cold PBS and centrifuged again. In accordance with the instruction of an Annexin-V-FITC apoptosis determination kit (Sigma-Aldrich Chemical Company, St Louis MO, USA), 150 μl binding buffer and 5 μl Annexin-V-FITC were added into each tube. After oscillation, the cells were incubated in the dark at room temperature for 15 min. Subsequently, another 100 μl of binding buffer and 5 μl PI stain (Sigma-Aldrich Chemical Company, St Louis MO, USA) were added into each tube. After oscillation, cell apoptosis was detected by flow cytometry.
Western blot analysis
Cells were collected and incubated with a 1 × SDS lysis buffer (Beyotime Biotechnology Co., Shanghai, China). The extracted protein was heated at 100 °C for 5 min and the samples (20 μl) were loaded onto 10% polyacrylamide gel electrophoresis (PAGE). Subsequently, the protein was transferred at voltage of 48 V onto a polyvinylidene fluoride (PVDF) membrane for 3.5 h, followed by incubation with 5% bovine serum albumin (BSA) at room temperature for 2 h to block the membrane. After being rinsed by 1× tris-buffered saline with tween 20 (TBST), the membrane was incubated at 4 °C overnight with rabbit anti-human primary antibodies [CDA, t-AKT, t-extracellular signal-regulated kinase (ERK)1/2, p-ERK1/2, phosphatidylinositol-3 kinase (PI3K), B cell lymphoma/leukemia-2 (BCL-2), BCL-XL, bcl-2-associated X (Bax), p21, cleaved caspase-3, total caspase-3, cleaved poly (ADP-ribose) polymerase (PARP), and total PARP, all diluted by 1: 500, (Cell Signaling Technologies (CST), Beverly, MA, USA)]. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody was used as an internal reference. After washing with 1 × TBST, the cells were incubated at room temperature for 1 h with peroxidase labeled goat anti-rabbit secondary antibody (1:2000) (Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA). After washing the cells with TBST again, chemiluminescence reagents (Thermo Fisher Scientific, California, USA) were added onto the cells and incubated for 1 min. The enhanced chemiluminescence (ECL) of all protein bands was visualized and recorded for subsequent Western blot analysis.
Tumor xenograft implantation in nude mice
Specific pathogen free (SPF) nude mice (body weight: 18–25 g; age: 5–6 weeks old) purchased from Hunan SLAC Jingda Laboratory Animal Co., Ltd. (Changsha, Hunan, China) were raised in a key laboratory for SPF animals. The mice were housed in a SPF grade chamber under aseptic laminar flow and a constant temperature (22–25 °C)/humidity (55 ± 5%). All food, water and padding for the mice were sterilized. In this study, 9 nude mice were allocated into a blank, a shRNA-con and a shRNA-1 group, with 3 mice in each group. Before tumor inoculation, the back of each nude mouse was cleaned with 75% alcohol, and 0.2 ml of cell suspension (106 cells/ml) were injected into each mouse using a skin test needle. The maximum diameter (a) and the maximum transverse diameter (b) of the tumor were measured every 3 days after inoculation, and the volume of the tumor (V = ab2/2) was calculated. The diet, body shape and defecation of nude mice were observed. After 28 days of inoculation, all nude mice were sacrificed by cervical dislocation. Subsequently, the tumor was isolated and weighed to calculate the tumor volume.
SPSS 21.0 software (IBM Corp., Armonk, NY, USA) was used for data analysis. The measurement data were expressed as mean ± standard deviation. Comparison between two groups was conducted by t test. One-way analysis of variance (ANOVA) was used for comparing multiple groups. p < 0.05 was considered statistically significant.
CDA mRNA and protein expression was increased in CML and AL patients
Enzymatic activity of CDA was increased in CML and AL patients
Comparison of the enzymatic activity of CDA
1375.76 ± 321.69*
1452.52 ± 432.43*
923.42 ± 269.02
1542.52 ± 202.65
990.97 ± 186.72#
1666.58 ± 344.66
1000.62 ± 170.17#
CDA gene silencing decreased CDA mRNA and protein expression
CDA gene silencing inhibited cell proliferation
Decreased clone cells in the shRNA-1 and shRNA-2 groups
More cells were arrested in the S and G2 phases after CDA gene silencing
CDA gene silencing induced cell apoptosis
CDA gene silencing down-regulated BCL-2 and up-regulated p21, Bax, cleaved caspase-3/total caspase-3 and cleaved PARP/total PARP
Underlying mechanism of CDA in K562 cell proliferation, apoptosis and cell cycle
CDA gene silencing inhibited tumor growth
It seems that CML mainly affects young people from Asia . A previous study revealed that high expression of CDA was responsible for cytidine resistance in cancer cells . Therefore, the aim of our study was to determine the effects of CDA gene silencing on the apoptosis and proliferation of K562 cells. Besides, our study confirmed that down-regulated CDA gene expression could inhibit cell proliferation and induce apoptosis.
In our study, the mRNA expression and enzymatic activity of CDA in CML and AL patients remarkably increased compared with those in the donors of hematopoietic stem cell transplantation. CDA is a widely used chemotherapeutic agent in the treatment of certain leukemia and has high expression in cancer cells, which is regarded as a prominent enzyme in the catabolism of cytosine nucleoside analogues [19, 20]. In addition, CDA was reported to be the prominent enzyme in the metabolism of gemcitabine, a pyrimidine analog widely used in the chemotherapy of tumors such as biliary tract cancer. Therefore, the cellular activity of CDA is crucial for understanding gemcitabine-induced toxicity [10, 11]. Whether in vitro or in vivo, CDA could play a major role to regulate the susceptibility to antitumor dCyd analogs and hence affect the antitumor activity of 2′-C-cyano-2′-deoxy-1-beta-d-arabinofuranosylcytosine (CNDAC) . One former study found that, in many cancers including myeloid leukemia, the CDA expression in male cancer patients was remarkably higher than that in females . Besides, when compared with patients of complete remission, CDA expression was higher in previously untreated and relapsed AL patients . Additionally, in AML patients, lower CDA levels predicted a longer disease-free survival . Carpi et al. demonstrated that the average value of CDA activity was 0.051 ± 0.024 mU/mg, which also followed a normal distribution in healthy subjects . For this reason, we speculated that CDA could become a promising biomarker and therapeutic target for the treatment of leukemia.
In order to study the effect of CDA gene on the biological property of K562 cells, we carried out CDA gene silencing in K562 cells. Upon CDA gene silencing, the cell proliferation was inhibited and the cells were arrested in S and G2/M phases, accompanied by an increased apoptotic rate. CDA treatment can cause the death of Jurkat human T leukemia cells by inducing apoptosis in a time- and dose-dependent fashion . A previous study found that miR-484 could inhibit CDA and suppress the proliferation of breast cancer cells . Furthermore, when compared with the blank group, the shRNA-1 and shRNA-2 groups were associated with down-regulated expression of PI3K, p-AKT/t-AKT, p-ERK1/2/t-ERK1/2 and Bcl-2, and up-regulated expression of p21, Bax, cleaved caspase-3/total caspase-3 and cleaved PARP/total PARP. It implicated that CDA gene silencing could involve in the ability of K562 cell proliferation and apoptosis by regulating the PI3K/Akt signaling pathway and MAPK/ERK signaling pathway PARPs were found to promote the transformation of ADP-ribose and play critical roles in several cellular processes, such as transcription, replication, recombination, and DNA repair . A recent study found that CDA deficiency could impair sister chromatid disjunction by decreasing PARP-1 activity . ERK is important for transducing extracellular signals to cell nuclei, and hence acts as a main downstream transducer of Ras. In addition, the invasiveness of tumor cells can be altered by nuclear localization of phosphorylated or activated ERK . CDA treatment has been reported to increase the phosphorylation of ERK1/2 in Jurkat T leukemia cells . It was reported that the chronic phase of CML and the blast crisis of CML were accompanied by AKT overexpression . It has also been reported that the reduction in AKT expression enhanced gemcitabine-induced apoptosis and its anti-tumor activity in pancreatic carcinoma . PI3Ks are lipid kinases that can regulate various cellular processes, such as proliferation, adhesion and survival . PI3Ks are frequently upregulated in cancer , suggesting that the down-regulated expression of PI3K upon CDA gene silencing may be helpful for the treatment of CML. Our study agrees well with the above, that CDA gene silencing mediated the PI3K/Akt signaling pathway and MAPK/ERK signaling pathway, resulting in under controlled proliferation of K562 cells. The mediation of PI3K/Akt signaling pathway and MAPK/ERK signaling pathway on apoptotic factors such as Bax and Bcl-2 have been illustrated in lots of studies. The Bcl-2 family proteins are important regulators and their main acting site is on the outer mitochondrial membrane . As an anti-apoptotic protein, the high-expression of Bcl-2 could impede apoptosis in cancer cells . A previous study found that the expression of pro-apoptotic factors, Bax and Caspase3, was up-regulated in chemotherapy, while the expression of anti-apoptotic factor Bcl-2 was markedly down-regulated . As a cyclin-dependent kinase inhibitor, p21 can stop cellular proliferation and block cell cycle progression . As a member of the Bcl-2 family, Bax is a pro-apoptotic factor translocated from the cytosol to the mitochondria during programmed cell death . As a result, we observed activation of anti-apoptotic factor Bcl-2, together with suppression of pro-apoptotic factor Bax. In related to silenced CDA, K562 cell proliferation is restricted, and apoptosis is accelerated.
In conclusion, our study demonstrated that CDA was highly expressed in CML cells, and CDA gene silencing could inhibit the proliferation of CML cells and promote their apoptosis. These processes might be closely associated with the development of malignant tumors. Thus, CDA could become a new target for the treatment of CML. However, we have not studied other cell lines due to limited time and resources, and the specific mechanism underlying the function of CDA in CML remains elusive.
XFW and QKZ designed the study and prepared the figures. QLC and QKZ collated the data, designed and developed the database, carried out data analyses and produced the initial draft of the manuscript. XFW and YFF contributed to drafting the manuscript. All authors read and approved the final manuscript.
The authors want to show their appreciation to reviewers for their helpful comments.
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
Ethics approval and consent to participate
This study was approved by the ethics committee of Gansu Provincial Hospital, and all patients had signed informed consent.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 13.Mameri H, Bieche I, Meseure D, Marangoni E, Buhagiar-Labarchede G, Nicolas A, Vacher S, Onclercq-Delic R, Rajapakse V, Varma S, Reinhold WC, Pommier Y, Amor-Gueret M. Cytidine deaminase deficiency reveals new therapeutic opportunities against cancer. Clin Cancer Res. 2017;23:2116–26.CrossRefPubMedGoogle Scholar
- 15.Abraham A, Varatharajan S, Abbas S, Zhang W, Shaji RV, Ahmed R, Abraham A, George B, Srivastava A, Chandy M, Mathews V, Balasubramanian P. Cytidine deaminase genetic variants influence RNA expression and cytarabine cytotoxicity in acute myeloid leukemia. Pharmacogenomics. 2012;13:269–82.CrossRefPubMedGoogle Scholar
- 16.McGrath JP, Williamson KE, Balasubramanian S, Odate S, Arora S, Hatton C, Edwards TM, O’Brien T, Magnuson S, Stokoe D, Daniels DL, Bryant BM, Trojer P. Pharmacological inhibition of the histone lysine demethylase KDM1A suppresses the growth of multiple acute myeloid leukemia subtypes. Cancer Res. 2016;76:1975–88.CrossRefPubMedGoogle Scholar
- 22.Mahfouz RZ, Jankowska A, Ebrahem Q, Gu X, Visconte V, Tabarroki A, Terse P, Covey J, Chan K, Ling Y, Engelke KJ, Sekeres MA, Tiu R, Maciejewski J, Radivoyevitch T, Saunthararajah Y. Increased CDA expression/activity in males contributes to decreased cytidine analog half-life and likely contributes to worse outcomes with 5-azacytidine or decitabine therapy. Clin Cancer Res. 2013;19:938–48.CrossRefPubMedPubMedCentralGoogle Scholar
- 29.Tai CJ, Lee CH, Chen HC, Wang HK, Jiang MC, Su TC, Shen KH, Lin SH, Yeh CM, Chen CJ, Yeh KT, Chang CC. High nuclear expression of phosphorylated extracellular signal-regulated kinase in tumor cells in colorectal glands is associated with poor outcome in colorectal cancer. Ann Diagn Pathol. 2013;17:165–71.CrossRefPubMedGoogle Scholar
- 37.Winter JN, Li S, Aurora V, Variakojis D, Nelson B, Krajewska M, Zhang L, Habermann TM, Fisher RI, Macon WR, Chhanabhai M, Felgar RE, Hsi ED, Medeiros LJ, Weick JK, Weller EA, Melnick A, Reed JC, Horning SJ, Gascoyne RD. Expression of p21 protein predicts clinical outcome in DLBCL patients older than 60 years treated with R-CHOP but not CHOP: a prospective ECOG and Southwest Oncology Group correlative study on E4494. Clin Cancer Res. 2010;16:2435–42.CrossRefPubMedPubMedCentralGoogle Scholar
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.