Bio-compatibility and cytotoxicity studies of water-soluble CuInS2-ZnS-AFP fluorescence probe in liver cancer cells

Ming-Ya Yang， Jian Hong， Yan Zhang， Zhen Gao， Tong-Tong Jiang， Jiang-Luqi Song，Xiao-Liang Xu and Li-Xin ZhuHefei， China

?

Bio-compatibility and cytotoxicity studies of water-soluble CuInS2-ZnS-AFP fluorescence probe in liver cancer cells

Ming-Ya Yang， Jian Hong， Yan Zhang， Zhen Gao， Tong-Tong Jiang， Jiang-Luqi Song，Xiao-Liang Xu and Li-Xin Zhu
Hefei， China

BACKGROUND： The oncogenesis of hepatocellular carcinoma （HCC） is not clear. The current methods of the pertinent studies are not precise and sensitive. The present study was to use liver cancer cell line to explore the bio-compatibility and cytotoxicity of ternary quantum dots （QDs） probe and to evaluate the possible application of QDs in HCC.

METHODS： CuInS2-ZnS-AFP fluorescence probe was designed and synthesized to label the liver cancer cell HepG2. The cytotoxicity of CuInS2-ZnS-AFP probe was evaluated by MTT experiments and flow cytometry.

RESULTS： The labeling experiments indicated that CuInS2-ZnS QDs conjugated with AFP antibody could enter HepG2 cells effectively and emit intensive yellow fluorescence by ultraviolet excitation without changing cellular morphology. Toxicity tests suggested that the cytotoxicity of CuInS2-ZnS-AFP probe was significantly lower than that of CdTe-ZnS-AFP probe （t test，

F=0.8， T=-69.326， P＜0.001）. For CuInS2-ZnS-AFP probe， timeeffect relationship was presented in intermediate concentration（＞20%） groups （P＜0.05） and dose-effect relationship was presented in almost all of the groups （P＜0.05）.

CONCLUSION： CuInS2-ZnS-AFP QDs probe had better biocompatibility and lower cytotoxicity compared with CdTe-ZnSAFP probe， and could be used for imaging the living cells in vitro. （Hepatobiliary Pancreat Dis Int 2016；15：406-411）

CuInS2-ZnS quantum dot；

HepG2 cells；

bio-compatibility；

cytotoxicity

Introduction

H epatocellular carcinoma （HCC） is universally a common malignancy and the prognosis is poor.［1， 2］In 2010， the new HCC cases and death in China were 358 840 and 312 432， respectively.［3］Studies on HCC oncogenesis and progression are important to improve the outcomes of this population.

Quantum dot （QD） is an artificial nanoparticle （2-10 nm）.［4］The excellent optical properties of QDs attract a lot of attentions from applications of medicine to bioimaging field.［5］Fluorescence probe QDs have already been used in biological imaging in vivo and in vitro.［6-8］The surface of QDs can be functionalized by conjugating with polypeptide， protein and antibody and therefore，extraordinary combining capacity of QDs was observed in tumor targeting and imaging in vivo.［9-11］Originally the cadmium-series QDs were the most attractive in the cell fluorescent labeling due to their easily growth craft，but high cytotoxicity restricted its further application.［12］On the other hand， in order to seek a better QDs with higher optical stability， CuInS2and AgInS2QDs are fabricated and become the commonest free cadmium QDs with good bio-imaging capacity.［13］Although CuInS2and AgInS2are cadmium free QDs with a potential in biomedical applications， the literature on the synthesis，imaging of the cells and cytotoxicity in living cells are paucity. The aim of the present study was to expand the bio-function of the ternary QDs， via testing the cytotoxicity and bio-compatibility of the QDs probes labeling in the HepG2 cells.

Alpha-fetoprotein （AFP） is considered a HCC marker.［14］In this study， AFP antibody was selected to conjugate the QDs CuInS2-ZnS to form a new fluorescence probe CuInS2-ZnS-AFP. So the probe could be used to combine and target the HepG2 cells， an HCC cell line that can positively express the AFP. The purpose of the study was to synthesize the probe， to detect the cytotoxicity and to investigate the application of the probe in liver cancer.

Methods

Materials

HepG2 cell line （from cell bank of Chinese Academy of Science）， serum-free high glucose DMEM medium， fetal calf serum， PBS solution， penicillin-streptomycin， trypsin， dimethyl sulfoxide （DMSO）， 0.5% 3-（4， 5-dimethyl-2-thiazolyl）-2， 5-diphenyl-2-H-tetrazolium bromide （MTT），4'， 6-diamidino-2-phenylindole （DAPI）， rabbit anti-AFP antibody， 3， 3'-diaminobenzidine （DAB） and secondary antibody were all purchased from Beijing Zhongshan Biotechnology Co.， Ltd. Annexin V-FITC and leucocytic permeabilization reagent were purchased from American Beckman Coulter commercial enterprise Co.， Ltd. CuInS2-ZnS QDs synthesis related reagents were provided by University of Science and Technology of China.

Synthesis of CuInS2-ZnS and CdTe-ZnS core-shell QDs

Hydrothermal synthesis method was used to synthesize CuInS2QDs. At room temperature， 0.15 mmol CuCl2and InCl3were added into 10.5 mL deionized water， then 1.8 mmol GSH was added into the solution. Solution PH was adjusted to 11.3 with 2 mol/L NaOH solution，0.3 mmol thiourea was added and dissolved thoroughly by stirring the solution for 10 minutes. The CuInS2-ZnS core-shell QDs were prepared by adding 0.8 mmol Zn（OAc）2·2H2O， 1 mmol sodium citrate and 1.2 mmol GSH into the CZIS QD solution. Then 0.8 mmol Na2S solution was slowly added. After stirring for 10 minutes，the reaction solution was heated for 2 hours at 100 ℃. The mixed liquor was transferred into a 15 mL autoclave and heated at 150 ℃ for 21 hours. After the solution was cooled to room temperature in cold water， ethanol was added， the precipitate was centrifuged with high-speed centrifuge and the pellet was dissolved in the deionized water. CdTe-ZnS core-shell QDs were provided by University of Science and Technology of China and the preparation was according to the instructions.［15］

Preparation of CuInS2-ZnS-AFP and CdTe-ZnS-AFP QDs probe

100 μL of these two QDs solution （12.8 mmol/L） was added to the mixture solution which contained 10 μL 200 mg/mL EDC， 10 μL 20 mg/mL sulfo-NHS and 190 μL PBS. After 1 hour incubation at 37 ℃， PBS was added and the solution was centrifuged for 3 times， the supernatant was discarded and precipitate was resuspended by 600 μL PBS. 100 μL of 0.1 g/mL rabbit anti-AFP antibody was added and the reaction solution was shaking for 1 hour at room temperature， and after 3 times centrifuging QDs probe precipitate was resuspended in PBS. Both of the densities of these two QDs probes were 1.2 mg/mL.

Preparation of cell-coverslips

The exponential phase HepG2 cells were trypsinized and harvested. Sterile coverslips were placed in the sixwell plate； cell suspension was dropped into each well with the density of 50-100×103/well and cultured in an incubator with 5% CO2at 37 ℃ for 24 hours.

Cell labeling and imaging

At room temperature， cell coverslips were washed for 3 times with PBS and 50 μL fixation agent was dropped and maintained on the glasses for 10 minutes. The cell glasses were washed as above and 50 μL permeability agent was added and washed after 5 minutes.

50 μL 0.24 mg/mL either CuInS2-ZnS-AFP probe solution （experimental group） or CuInS2-ZnS QDs solution （control group） was dropped onto the fixed cell glasses， respectively， and these glasses were placed in a humidified box and incubated at 37 ℃ for 40 minutes. These coverslips were observed under a fluorescence microscope （Olympus IX73） and a confocal laser scanning microscope （Leica SP5）.

MTT assay

The suspension of exponential phase cells were harvested at the density of 5-10×103/100 μL， 100 μL cell suspension was planted in 96-well plates， the fringe wells were filled with 100 μL PBS. After 24 hours culture in incubator， all solution in these 96-well plates was discarded excepting for the fringe wells and two types of QDs probe solution with a range of concentrations （2.5%， 5%，10%， 15%， 20%， 30%， 40% and 50%） was added， each concentration had five accessory wells， that were experimental groups. Control groups and zero setting groups were HepG2 cells+diluent and pure diluent， respectively. The diluent was high glucose DMEM medium with 10% fetal calf serum. After incubating with 5% CO2at 37 ℃for 2 hours， 6 hours， 24 hours， 48 hours and 72 hours，cellular morphology was observed under microscope，then 0.5% MTT solution and DMSO were added. After that all the plates were shaken immediately and simulta-neously for 10 minutes in dark to dissolve the produced formazan. Finally OD value was read with a micro-plate reader （ELX 800， Bio-Tek） at 490 nm. Before analyzing with SPSS 22.0， MTT data was processed with a formula as follow： IR=（E-Z）/（C-Z）. IR， E， C， Z are the inhibition rate， the OD values of the experimental group， control group and zero settings， respectively.

Flow cytometry

Exponential phase HepG2 cells were planted into 6-well plate， 1 mL 0.24 mg/mL （20%） QDs probe solution was added to each well after HepG2 cells adhered to the bottom （diluent was high glucose DMEM with 10% fetal calf serum）. After culture in an incubator for 24 hours and 48 hours， cell culture was stopped by trypsin and cells were harvested. AnnexinV-PI kit was used to detect cell apoptosis with a flow cytometer （Beckman Coulter Navios）.

Optical experiments

Absorption spectrum （UV-Vis）： UV-visible absorbance was measured with a Shimadzu UV-VIS-NIR spectrophotometer （UV-3600）. Fluorescence spectrum （PL）： photoluminescence spectrum was measured with a fluorescence spectrophotometer （F-4600）.

Statistical analysis

ANOVA （one-way） was used to analyze the time-effect and dose-effect relationship of these two QDs probe in HepG2 cells， that LSD（L）， S-N-K（s） and Dunnett（E）were included. When time was fixed， dose-effect relationship was analyzed as well as time-effect relationship was analyzed when dose was fixed. Independent-samples t test including F test and T test were used to analyze the difference of cytotoxicity between two QDs probe with SPSS 22.0. A P＜0.05 was considered statistically significant.

Fig. 1. Fluorescence microscope images of HepG2 cells after incubating with CuInS2-ZnS-AFP probe （0.24 mg/mL 40 minutes） in bright-field （A）， and in dark-field （B）. Control group （CuInS2-ZnS QDs， 0.24 mg/mL， 40 minutes） in bright-field （C） and in darkfield （D）. The yellow fluorescence was generated by QDs probe under UV light （dark-field） （original magnification ×400）. In the experimental group， confocal fluorescence images of HepG2 cells after co-culturing with CuInS2-ZnS-AFP probe （0.24 mg/mL 40 minutes） and being stained by DAPI dye were taken in blueyellow dual channel （E） and blue channel （F）. Confocal fluorescence images in the control group （CuInS2-ZnS QDs， 0.24 mg/mL，40 minutes） were taken in blue-yellow dual channel （G） and blue channel （H）. The yellow fluorescence was generated by QDs probe under UV light （yellow channel） and the blue fluorescence was produced by DAPI in blue channel （original magnification ×630）.

Results

Cell labeling and imaging

Immunohistochemistry showed that the AFP was positive in HepG2 cytoplasm. It was found that cellular morphology of the experimental group was clear and intact without any fluorescence emitting in bright field （Fig. 1A）under an inverted fluorescence microscope. Intensive yellow fluorescence was observed in cytoplasm but negative in cell nucleus with a normal cellular morphology when excited by UV light （Fig. 1B）. No fluorescence was found in cytoplasm both in bright field （Fig. 1C） and dark field （Fig. 1D） in the control group （original magnification × 400）. Further observation was conducted with a confocal microscope. In the experimental group，yellow fluorescence was observed in cytoplasm and blue fluorescence in cell nucleus in blue-yellow dual channel （Fig. 1E）， and in blue channel output there was only blue cell nucleus （Fig. 1F）. For the control group， only blue nucleus appeared in both channel outputs （Fig. 1G， H），which meant no QDs were remained in cells （originalmagnification ×630）.

MTT analysis

For the CuInS2-ZnS-AFP probe， when the probe concentration was low （＜20%）， there was no exact relationship between cell inhibition rate and incubation time that indicated no time-effect relationship （P=0.286，0.401 and 0.807， respectively， in comparison among the 20%-2 h， 6 h， 24 h groups）. However， when the probe concentration was higher （＞20%）， cell inhibition rate was positively correlated with incubation time. In addition， the cell inhibition rate of this QDs probe was positively correlated with probe concentration， that dose-effect relationship was presented in most comparisons among groups with statistic significance （P＜0.05）excepting for that in 2.5%， 5%-48 h， 72 h （P=0.279 and 0.107， respectively） （Fig. 2）.

For the CdTe-ZnS-AFP probe， probe concentration and incubation time were positively correlated with cell inhibition rate （P＜0.05）， obvious time-effect and doseeffect relationship was presented （Fig. 3）.

Independent-samples t test presented that the cytotoxicity of CuInS2-ZnS-AFP probe was significantly lower than that of CdTe-ZnS-AFP probe （F=0.8， T=-69.326，P＜0.001）.

Flow cytometry

Survival rates of CuInS2-ZnS-AFP in the 20%-24 h and 20%-48 h groups detected by flow cytometry were 91.2% and 80.2%， the apoptosis rates were 8.6% and 19.4%， which were similar with the results of MTT assay which were 8.3% and 19.2% （Fig. 4）.

Optical experiments

The absorption spectrum （UV-Vis） and fluorescence spectrum （PL） of CuInS2-ZnS were measured with UV-3600 UV-VIS-NIR spectrophotometer and F-4600 fluorescence spectrophotometer （Figs. 5， 6）. According to the data of optical property experiment， the intensive absorption spectrum was in ultraviolet band， and fluorescence intensity reached the peak when emission wavelength was 600-650 nm. Stokes shift is the difference （in wavelength or frequency units） between positions of the band maxima of the absorption and emission spectra of the same electronic transition. When a QD absorbs a photon， it gains energy and enters an excited state. One path for the QD to release is to emit a photon and another method would be the loss of heat energy. When the emitted photon has less energy than the absorbed one，this energy difference is the Stokes Shift. As it was shown in Figs. 5 and 6， the suitable Stokes Shift made QDs have no overlap between UV-Vis and PL which made QDs work well in cell labeling.

Fig. 2. Cell inhibition rates （Y-axis） of different CuInS2-ZnS-AFP concentrations （X-axis） with different co-culturing time.

Fig. 3. Cell inhibition rates （Y-axis） of different CdTe-ZnS-AFP concentrations （X-axis） with different co-culturing time.

Fig. 4. The apoptosis rates of these two flow cytometry pictures were 8.6% （left 20%-24 h） and 19.4% （right 20%-48 h）， respectively. B1： cell debris； B2： non-viable apoptotic cells； B3： alive cells；B4： viable apoptotic cell.

Discussion

In the imaging experiment， intensive fluorescence was observed in cytoplasm. The principle of synthesizing theQDs-antibody probe is to form amide bonds with carboxyl and amino in both of QDs and antibody.［16］Then the antibody domain combines to the corresponding antigen in cells. When cells were exposed under UV light，the combined sites emit fluorescence， which presented the targeting capacity.

In this study， the lower concentrations （2.5%-10%）of the CuInS2-ZnS-AFP probe promoted the cell proliferation when incubation time was 24 hours. It has been reported that cytotoxicity of gold nanoparticle was associated with particle size and coagulation. When particle size was bigger than the threshold of endocytosis，nanoparticles would adhere to membrane surface rather than enter into Hela cells，［17］which contributed to cell proliferation instead of promoting apoptosis. The specific mechanism is still unknown. According to this， it is safe to assume that proliferative effect of CuInS2-ZnSAFP at low concentrations was superior to the apoptotic effect after co-culturing with HepG2 cells for a period of time （24 hours in this experiment）. While along with extension of culture time， apoptotic effect turned to be dominant. This result indicated that it is important to consider the proliferative effect of QDs probe rather than their cytotoxicity in clinical applications.

Obvious time-effect and dose-effect relationship of cell inhibition rate in CdTe-ZnS-AFP probe and in CuInS2-ZnS-AFP probe with intermediate concentration were observed， the specific mechanism was still unclear which might be related to the synthetic materials and degradation product of QDs. Semiconductor nanoparticles are usually composed of atoms from groups II-VI，III-V， or IV-VI of the periodic table，［18］which were binary QDs. Ternary QD was a new technological achievement which elements from I-III-IV groups. As QDs are effective energy donors，［19， 20］they could transfer energy to adjacent oxygen molecules， inducing the generation of reactive oxygen species （ROS）， and resulting in cell damage or death.［21］For instance， cadmium selenide （CdSe）QDs have been demonstrated to produce singlet oxygen in vitro.［22］While other than free radical formation， the mechanism of QD-induced cytotoxicity might be also mediated by other factors.［23］It was documented that cytotoxicity of QDs is related to the concentration of Cd2+released as an intermediate product of the deterioration in CdSe lattice，［12］and surface coatings such as ZnS， dihydrolipoic acid，［24］and polyacrylate［25］that reduce QD surface oxidation.

This study showed that the cytotoxicity of CdTe-ZnSAFP probe was severer than that of CuInS2-ZnS-AFP probe. The survival rate was 80% or 92.7% after incubating with 0.24 mg/mL CdTe-ZnS-AFP probe or CuInS2-ZnS-AFP probe solutions （20%） for 24 hours， respectively. Ternary QD used elements from the I-III-IV groups to replace the previous heavy metals， such as cadmium，which could greatly decrease the cytotoxicity caused by synthetic materials and degradation product. In some old reports stabilizer of CdTe-ZnS QD was MPA （3-mercaptopropionic acid） which is toxic.［15］In this study GSH was used to decrease the cytotoxicity. However， studies are still needed to further decrease the cytotoxicity.

This study was a preliminary research to verify the targeting capacity and the cytotoxicity of the CuInS2-ZnS-AFP probe rather than the specificity. Further study should be focused on bio-toxicity， fluorescence lifetime and specificity in animal experiment. In this study，CuInS2-ZnS QDs were produced with hydrothermal synthesis method， and the AFP antibody was conjugated with the QDs to form the CuInS2-ZnS-AFP probe. The probe， which is of a character of lower cytotoxicity， can label and track the liver cancer cells in vitro. Based on its stable optical properties and bio-compatibility， the QDs probe may provide a new method to study the oncogenesis and progression of HCC.

Fig. 5. Absorption spectrum （UV-Vis）.

Fig. 6. Fluorescence spectrum （PL）.

Contributors： YMY proposed the study. YMY and HJ performed the research and experiments draft， JTT and SJ performed the material synthesis， ZY and GZ analyzed the data. XXL and ZLXmade modifications to the manuscript. YMY and HJ contributed equally to this work. All authors made contributions to this study，and ZLX is the guarantor.

Funding： This research was supported by grants from the National Natural Science Foundation of China （51272246 and 81172082）. Ethical approval： Not needed.

Competing interest： No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

1 Wang L， Yao M， Pan LH， Qian Q， Yao DF. Glypican-3 is a biomarker and a therapeutic target of hepatocellular carcinoma. Hepatobiliary Pancreat Dis Int 2015；14：361-366.

2 Ha SY， Song DH， Hwang SH， Cho SY， Park CK. Expression of prothymosin alpha predicts early recurrence and poor prognosis of hepatocellular carcinoma. Hepatobiliary Pancreat Dis Int 2015；14：171-177.

3 Wei KR， Yu X， Zheng RS， Peng XB， Zhang SW， Ji MF， et al. Incidence and mortality of liver cancer in China， 2010. Chin J Cancer 2014；33：388-394.

4 Redl FX， Cho KS， Murray CB， O'Brien S. Three-dimensional binary superlattices of magnetic nanocrystals and semiconductor quantum dots. Nature 2003；423：968-971.

5 Chan WC， Maxwell DJ， Gao X， Bailey RE， Han M， Nie S. Luminescent quantum dots for multiplexed biological detection and imaging. Curr Opin Biotechnol 2002；13：40-46.

6 Roco MC. Nanotechnology： convergence with modern biology and medicine. Curr Opin Biotechnol 2003；14：337-346.

7 Katz E， Willner I. Integrated nanoparticle-biomolecule hybrid systems： synthesis， properties， and applications. Angew Chem Int Ed Engl 2004；43：6042-6108.

8 Cai W， Chen X. Nanoplatforms for targeted molecular imaging in living subjects. Small 2007；3：1840-1854.

9 Michalet X， Pinaud FF， Bentolila LA， Tsay JM， Doose S， Li JJ， et al. Quantum dots for live cells， in vivo imaging， and diagnostics. Science 2005；307：538-544.

10 Cai W， Hsu AR， Li ZB， Chen X. Are quantum dots ready for in vivo imaging in human subjects？ Nanoscale Res Lett 2007；2：265-281.

11 Cai W， Shin DW， Chen K， Gheysens O， Cao Q， Wang SX， et al. Peptide-labeled near-infrared quantum dots for imaging tumor vasculature in living subjects. Nano Lett 2006；6：669-676.

12 Chen N， He Y， Su Y， Li X， Huang Q， Wang H， et al. The cytotoxicity of cadmium-based quantum dots. Biomaterials 2012；33： 1238-1244.

13 Mashinchian O， Johari-Ahar M， Ghaemi B， Rashidi M， Barar J，Omidi Y. Impacts of quantum dots in molecular detection and bioimaging of cancer. Bioimpacts 2014；4：149-166.

14 Mizejewski GJ. Alpha-fetoprotein structure and function： relevance to isoforms， epitopes， and conformational variants. Exp Biol Med （Maywood） 2001；226：377-408.

15 He H， Sun X， Wang X， Xu H. Synthesis of highly luminescent and biocompatible CdTe/CdS/ZnS quantum dots using microwave irradiation： a comparative study of different ligands. Luminescence 2014；29：837-845.

16 Song J， Jiang T， Guo T， Liu L， Wang H， Xia T， et al. Facile synthesis of water-soluble Zn-doped AgIn5S8/ZnS core/shell fluorescent nanocrystals and their biological application. Inorg Chem 2015；54：1627-1633.

17 Cui W， Li J， Zhang Y， Rong H， Lu W， Jiang L. Effects of aggregation and the surface properties of gold nanoparticles on cytotoxicity and cell growth. Nanomedicine 2012；8：46-53.

18 Cassette E， Helle M， Bezdetnaya L， Marchal F， Dubertret B，Pons T. Design of new quantum dot materials for deep tissue infrared imaging. Adv Drug Deliv Rev 2013；65：719-731.

19 Clapp AR， Medintz IL， Mauro JM， Fisher BR， Bawendi MG，Mattoussi H. Fluorescence resonance energy transfer between quantum dot donors and dye-labeled protein acceptors. J Am Chem Soc 2004；126：301-310.

20 Medintz IL， Trammell SA， Mattoussi H， Mauro JM. Reversible modulation of quantum dot photoluminescence using a protein-bound photochromic fluorescence resonance energy transfer acceptor. J Am Chem Soc 2004；126：30-31.

21 Zhan Q， Tang M. Research advances on apoptosis caused by quantum dots. Biol Trace Elem Res 2014；161：3-12.

22 Samia AC， Chen X， Burda C. Semiconductor quantum dots for photodynamic therapy. J Am Chem Soc 2003；125：15736-15737.

23 Lovri? J， Cho SJ， Winnik FM， Maysinger D. Unmodified cadmium telluride quantum dots induce reactive oxygen species formation leading to multiple organelle damage and cell death. Chem Biol 2005；12：1227-1234.

24 Jaiswal JK， Mattoussi H， Mauro JM， Simon SM. Long-term multiple color imaging of live cells using quantum dot bioconjugates. Nat Biotechnol 2003；21：47-51.

25 Wu X， Liu H， Liu J， Haley KN， Treadway JA， Larson JP， et al. Immunofluorescent labeling of cancer marker Her2 and other cellular targets with semiconductor quantum dots. Nat Biotechnol 2003；21：41-46.

Accepted after revision May 18， 2016

Author Affiliations： Center Laboratory， First Affiliated Hospital of Anhui Medical University， Hefei 230022， China （Yang MY， Hong J， Zhang Y， Gao Z and Zhu LX）； Key Laboratory of Strongly-Coupled Quantum Matter Physics， Chinese Academy of Sciences， School of Physical Sciences， University of Science and Technology of China， Hefei 230026， China （Jiang TT，Song J and Xu XL）

Li-Xin Zhu， PhD， Center Laboratory， First Affiliated Hospital of Anhui Medical University， Hefei 230022， China （Tel：+86-551-62922299； Fax： +86-551-62922335； Email： lx-zhu@163.com）
? 2016， Hepatobiliary Pancreat Dis Int. All rights reserved.

10.1016/S1499-3872（16）60112-1
Published online July 4， 2016.

March 11， 2016