Research Article | Open Access
Zi-qi Yu, Wen-fei Wang, Chuan-zhi Zhu, Ke-hong Zhang, Xin-chun Chen, Jian-yong Chen, "The Polymorphism rs17525495 of LTA4H Is Associated with Susceptibility of Crohn’s Disease instead of Intestinal Tuberculosis in a Chinese Han Population", BioMed Research International, vol. 2019, Article ID 9537050, 7 pages, 2019. https://doi.org/10.1155/2019/9537050
The Polymorphism rs17525495 of LTA4H Is Associated with Susceptibility of Crohn’s Disease instead of Intestinal Tuberculosis in a Chinese Han Population
Background. Because of the similarity of intestinal tuberculosis and Crohn’s disease in disease phenotype, differential diagnosis has always been a clinical problem. Arachidonic acid metabolites play an important role in the inflammatory response of intestinal tuberculosis and Crohn’s disease. Recent studies have shown that the polymorphism locus in the promoter region of LTA4H gene affects LTB4 expression level and the susceptibility to extrapulmonary tuberculosis. Thus, we identified a total of 148 patients with intestinal tuberculosis, 145 with Crohn’s disease, and 700 normal controls in this study. Methods. All the study participants were local Han people from Jiangxi Province in the past eleven years. DNA was extracted from the paraffin-embedded specimens or the whole blood. The LTA4H promoter SNP (rs17525495) was genotyped with TaqMan assay. Results. The T-alleles frequency was not significantly increased in patients with intestinal tuberculosis compared with healthy control group (p=0.630; OR=1.07; 95%CI=0.81-1.41), while patients with Crohn’s disease have significantly increased T allele frequency compared with healthy population (p=0.032; OR=1.34; 95%CI=1.03-1.75). During treatment, the presence of the T allele significantly increased the proportion of Crohn’s patients requiring glucocorticoids (p<0.05). Conclusions. The T allele of LTA4H gene SNP (rs17525495) is a risk factor for Crohn’s disease instead of intestinal tuberculosis. More importantly, there may be a potential association of the different genotypes of rs17525495 with the treatment efficacy of 5-ASA and glucocorticoids in patients with Crohn’s disease. The association between LTA4H polymorphism and drugs therapeutic effects might contribute to the practice of precision medicine and the prediction of clinical outcomes.
Intestinal tuberculosis (ITB) is a chronic granulomatous inflammation caused by M. tuberculosis invading the gut . Crohn’s disease (CD) is a chronic idiopathic inflammatory disease of the gastrointestinal tract. Although the exact cause is unknown, it’s believed to result from the interplay between genetic susceptibility, environmental factors, and intestinal microflora, showing an abnormal mucosal immune response and compromised epithelial barrier function . Asia was previously considered to be a low-endemic area; however, the incidence has considerably increased over recent decades due to the progress of industrialization and the westernization of lifestyle . On the other hand, globalization and the HIV pandemic have also caused a surge in incidence of TB and consequently abdominal TB in the western world [4, 5]. It has always been a dilemma to distinguish ITB patients from CD patients due to the dynamic changes of epidemiology and the similar characteristics of clinical examination, especially in developing countries [6, 7]. After ITB was misdiagnosed as CD, the immunosuppressive agents treatment may lead to aggravation of tuberculosis infection even systemic dissemination. As for CD misdiagnosed as ITB in reverse, more serious drug side effects and delay treatment will happen after anti-tuberculosis treatment.
Leukotriene A4 hydrolase (LTA4H) is a key enzyme of the 5-lipoxygenase (5-LO) pathway, the activation of which leads to the biosynthesis of proinflammatory leukotriene lipid mediators from arachidonic acid (AA) [8, 9]. The genetic variation of LTA4H affects the balance between proinflammatory and anti-inflammatory responses in the body. Therefore, it is closely related to the occurrence and development of various diseases such as asthma, atherosclerosis, infectious diseases, and tumors [10–13]. Through the 1000 Genomes Project, researchers confirmed that rs17525495 genotype affects LTA4H transcription and interferes with therapeutic outcomes of TB meningitis . In a Chinese population study, LTA4H gene polymorphism was also reported to be associated with extrapulmonary tuberculosis . In Crohn’s disease, James Jupp et al. reported a 4-fold increase in the number of LTA4H-stained cells in colon biopsies from patients with active inflammatory bowel disease compared to healthy controls . However, there is no report on the analysis of the LTA4H gene polymorphism in ITB and CD. Hence, in this study, we attempt to genotype the ITB and CD at the rs17525495 locus of the LTA4H gene in order to find the difference in genetic susceptibility between ITB and CD.
2.1. Study Participants
This case-control study was conducted between January 2008 and December 2018 at the Jiangxi Provincial People’s Hospital (Nanchang, China) and The First Affiliated Hospital of Nanchang University (Nanchang, China) and The Second Affiliated Hospital of Nanchang University (Nanchang, China). A total of 148 patients with ITB, 145 with CD, and 700 normal controls were enrolled in this study. All of the patients and normal controls were randomly selected from the three hospitals, and all the subjects were local Han people from Jiangxi Province. Inclusion and exclusion criteria for patients with intestinal tuberculosis were the following: age 18 years or older; diagnosed by clinical manifestations, endoscopy, X-ray imaging, pathological changes, positive acid-fast staining, and anti-tuberculosis treatment; no other infectious diseases, chronic diseases, and autoimmune diseases. Inclusion and exclusion criteria for patients with Crohn’s disease are the following: age 18 years or older; CD diagnosed based on clinical symptoms, X-ray findings, serological examination results, endoscopic findings, and histopathology following the Chinese Society of Gastroenterology consensus on the diagnosis and management of inflammatory bowel disease, and exclusion of infectious and other noninfectious colitis. The inclusion and exclusion criteria for healthy controls were the following: age 18 years or older; tuberculosis-specific IGRA release test negative; no tuberculosis clinical manifestations (no abnormalities in chest radiology); no other infectious diseases, chronic diseases, and autoimmune diseases. We find out the Paraffin-embedded biopsy specimens of ITB and CD patients and then cut the FFPE sample into 3 sections 20 μm thick and transfer into EP tube stored at −20°C immediately. 2 ml of lithium heparin anticoagulated whole blood was collected from the healthy controls, at 4°C for storage. The present study protocol was reviewed and approved by the institutional review board of the above three hospitals and complied with the principles of the Declaration of Helsinki.
2.2. Genotyping for LTA4H Gene rs17525495 Locus Polymorphisms
We used FFPE DNA Kit (OMERA, US) for extraction of genomic DNA from paraffin-embedded specimens of ITB and CD patients as described above. We also used DNeasy Blood & Tissue (QIAGEN, US) for extraction of genomic DNA from whole blood in healthy controls (HC). The SNP rs17525495 was genotyped with TaqMan assays in 20 μl reaction volume, using ABI 7500 real-time PCR systems (ABI, US). Cycling conditions were set up as follows: pre-PCR Read (Holding Stage) 60°C for 1 min and then Holding Stage 95°C for 10 min, followed by 40 cycles at 95°C for 15 s, 60°C for 1 min, and Post-PCR Read (Holding Stage) 60°C for 1 min. Results were analyzed by using ABI 7500 PCR instrument software.
2.3. Statistical Analysis
In this case-control study, we first compared the baseline characteristics of study population. The categorical variables were analyzed by chi-square test or Fisher’s exact test, and the continuous variables were compared using independent sample t-test. The goodness-of-fit χ2 test was performed to check if the genotype frequencies of case and control groups conformed to Hardy-Weinberg equilibrium. Power calculations were performed using Power and Sample Size Calculation Statistical Software . Unconditional logistic regression analysis was used to compare the genotype and allele frequencies among the three groups. Covariates as gender, age, smoking, and drinking were included to adjust the odds ratio (OR) value. The disease-associated Additive genetic model was evaluated by the Cochran-Armitage trend test. For the disease with significant results in the additive model text, logistic regression analysis was used to analyze the three genetic models (dominant, recessive, and codominant) that the rs17525495 locus may follow in the disease; OR and 95% confidence interval (CI) were calculated and then adjusted by using the above covariates. All statistical analyses were performed by using SPSS (statistical software version 24). The value of p<0.05 (two-sided) was considered statistically significant.
3.1. Clinical Characteristics
The clinical characteristics of study population are shown in Table 1. No significant differences were found between the ITB and CD groups in terms of gender, smoking and drinking history, WBC, Hb, ESR, PPD, or disease related surgery (p>0.05). However, the average diagnosis age of patients with CD is earlier than ITB, and both of T-SPOT positive rate and abnormal lung imaging findings showed significant differences between the two groups (p<0.05), indicating their essential role in identifying ITB and CD.
Smoker: subjects who had smoked more than one cigarette per day for at least one year; drinker: subjects who drank alcohol per week for more than six months; ESR: erythrocyte sedimentation rate; PPD, tuberculin skin testing. Crohn's disease phenotype defined by the Montreal Classification System, L1, Terminal ileum; L2, colonic; L3, ileocolonic; L4, upper GI. P-value means comparison between ITB and CD groups. Bold italics indicate the statistical significance (P < 0.05).
3.2. SNP Distributions
The genotype distributions of rs17525495 in the case and control groups conformed to Hardy-Weinberg equilibrium (p>0.05). Retrieving rs17525495 gene in the dbSNP database, we found that the MAF (Minor Allele Frequency) was 0.29 in the East Asian population (China & Japan), and the statistical power of the polymorphism study exceeded 0.75. The genotype and allele frequency distribution of rs17525495 and its relationship with ITB and CD in Chinese Han population were shown in Table 2. After adjusting for age, gender, and other covariates, there was no significant difference in genotype and allele frequency regarding LTA4H rs17525495 when we compared with HC and ITB, or ITB and CD. However, a comparison between patients and controls showed a significant association between the TT genotype and Crohn’s disease (p=0.034; OR=1.92; 95%CI=1.05-3.51); the presence of the T allele might function as one of the genetic risk factors for Crohn’s disease (p=0.032; OR=1.34; 95%CI=1.03-1.75).
The major genotype or allele was used as baseline. Adjusted for gender, age, smoking, and drinking history. Bold italics indicate the statistical significance (P < 0.05).
3.3. Genetic Model Analysis
In the additive model of ITB (χ2=1.07; p=0.30) and CD (χ2=5.98; p=0.01), there was no significant correlation between the genotype of LTA4H rs17525495 locus and ITB susceptibility, whereas there was significant association with CD. In addition, a significant difference was found between the wild-type and mutant homozygotes under the codominant model, and the difference was still statistically significant after adjusting for the covariates (p=0.035; OR=1.94; 95%CI=1.05-3.57). The results were shown in Table 3. It suggests that this site may follow a codominant model in the CD.
The major genotype or allele was used as baseline. Adjusted for gender, age, smoking, and drinking history. Bold italics indicate the statistical significance (P < 0.05).
3.4. The Association of Hematological Examination and Genotype
There were no statistical differences among the three clinical indicators (WBC, Hb, and ESR) in patients with different genotypes of rs17525495 in both ITB and CD patients. The result was shown in Supplementary Figure 1 (Figure S1).
3.5. The Association of Medicinal Cure Effect and Genotype in CD Patients
Among CD patients, we compared TT+CT with CC genotype to see if genotypes affect the efficacy of drug therapy (Figure 1). The presence of the T allele significantly increased the proportion of Crohn’s patients with requiring GS (p<0.05).
The development and maintenance of inflammation are controlled by a complex network of cells and cytokines. Among these are the eicosanoids: a class of structurally related paracrine hormones, derived from the metabolism of arachidonic acid that includes the prostaglandins, the leukotrienes, and lipoxins, play an important role in the inflammatory response . In the 5-LO pathway, LTA4H catalyzes the conversion of LT4H to LTB4 . LTB4 is a potent lipid chemoattractant involved in inflammation, which causes adhesion and chemotactic movement of inflammatory cells and induces macrophage release of oxygen radicals , which is also reported to stimulate the production of IL-6 and IFN-γ and TNF [21–23]. In addition, lipoxygenase-derived metabolites of arachidonic acid may also be involved in activation, proliferation, and differentiation of human lymphocytes [24, 25]. For several reasons, the expression level of LTA4H plays an important role in the body’s inflammatory response.
Rs17525495 is one of the tag SNPs of LTA4H, which is located in the promoter region of the gene. David M. Tobin et al. confirmed protein expression of rs17525495 TT group shows approximately 2.3-fold higher than CC group, with intermediate expression in the CT cell lines, and the locus affects transcriptional activity through conducting dual-luciferase assays . Therefore, rs17525495 is a functional SNP. This locus was originally studied in tuberculosis; among TB meningitis infected patients without dexamethasone treatment, the patients with TT genotype had the highest mortality. In dexamethasone treated group, the result is in reverse [14, 26]. Interestingly, SNP heterozygosity in tuberculosis infection showed a protective effect, whereas both homozygous appeared to be increasing disease severity. The reason might be inadequate or excessive acute inflammation influences the susceptibility to M. tuberculosis [14, 26, 27].
We found that the frequency of T allele showed no significant difference between ITB group and control. This is different from the findings of tuberculous meningitis in Vietnamese population . This may be related to the following three reasons. Firstly, there were differential effects of genetic polymorphisms on susceptibility to disease in various populations. In a large sample of Russian population research, the results suggest that the other six common polymorphisms in the LTA4H gene do not play any major role in susceptibility to clinical pulmonary tuberculosis . Secondly, related reports on the impact of this locus on susceptibility to tuberculosis are mainly in pulmonary tuberculosis and tuberculous meningitis. As we know, M. tuberculosis genotype influences clinical disease phenotype. More virulent Beijing/W lineage strains are associated with tuberculous meningitis . However, some of the infections in ITB are caused by the ingestion of dairy products contaminated with Mycobacterium bovis, which are less virulent mycobacteria . Furthermore, because the sample of ITB is relatively small, the possibility of significant differences after expanding the sample size cannot be ruled out.
The contribution of AA metabolites pathway in CD has been recognized . These metabolites are important in promoting the migration of neutrophils to the intestinal mucosa, which cause substantial epithelial damage by producing reactive oxygen species and proteases . Previous studies have shown that mucosal synthesis of LTB4 is greatly enhanced in CD. In particular, the increased, sustained, and recurrent inflammation characteristic of CD may be the result of an inherent defect in the metabolism of LTB4, leading to its enhanced accumulation and activity . The classical anti-inflammatory drugs for CD treatment such as 5-aminosalicylic acid and glucocorticoids are mainly through inhibition of PGE2 and LTB4 release in colonic mucosa to reduce inflammation . A Canadian study evaluated the proteomes of biopsies collected from 99 pediatric patients at diagnosis-stage and discovered and validated the potential protein biomarkers through quantifying over 3500 proteins. They found proteins involved in fatty acid metabolism to be elevated in the CD over UC pediatric patients, especially the protein LTA4H . Another research suggested that LTA4H is a special biomarker of CD diagnosis and severity evaluation. Whatever in intestinal mucosal-luminal interface (MLI) or stool samples, higher protein expression of LTA4H is related to CD . We hypothesize that patients with high expression of LTA4H produce more LTB4, causing an excessive acute inflammatory response, and thus glucocorticoids can effectively inhibit the progression of inflammation and improve the symptoms of patients.
In conclusion, we first report that the LTA4H SNP rs17525495 T allele is a risk factor for CD. More importantly, there is a clear correlation between genotype and drug efficacy. Further research on the interaction of LTA4H SNPs will provide a theoretical background for elucidating the role of arachidonic acid metabolites in tuberculosis and inflammatory bowel disease and may help the susceptible population to take host genotype-specific therapies earlier to achieve the best curative effect.
The data used to support the findings of this study are included within the article.
Conflicts of Interest
There are no conflicts of interest.
Zi-qi Yu and Wen-fei Wang contributed equally to the writing of this article.
The authors would like to acknowledge Science and Technology Department of Jiangxi Province, Key R&D Plan, Grant Number: 20171BBG70087, and Health Commission of Jiangxi Province, Grant Number: 20195016.
Figure S1: the association of hematological examination and genotype. (Supplementary Materials)
- H. D. Donoghue and J. Holton, “Intestinal tuberculosis,” Current Opinion in Infectious Diseases, vol. 22, no. 5, pp. 490–496, 2009.
- J. Torres, S. Mehandru, J.-F. Colombel, and L. Peyrin-Biroulet, “Crohn's disease,” The Lancet, vol. 389, no. 10080, pp. 1741–1755, 2017.
- S. C. Ng, H. Y. Shi, N. Hamidi et al., “Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies,” The Lancet, vol. 390, no. 10114, pp. 2769–2778, 2017.
- K. D. Horvath and R. L. Whelan, “Intestinal tuberculosis: return of an old disease,” American Journal of Gastroenterology, vol. 93, no. 5, pp. 692–696, 1998.
- V. Sibartie, W. O. Kirwan, S. O'Mahony, W. Stack, and F. Shanahan, “Intestinal tuberculosis mimicking Crohn's disease: Lessons relearned in a new era,” European Journal of Gastroenterology & Hepatology, vol. 19, no. 4, pp. 347–349, 2007.
- M. A. Almadi, S. Ghosh, and A. M. Aljebreen, “Differentiating intestinal tuberculosis from crohn's disease: A diagnostic challenge,” American Journal of Gastroenterology, vol. 104, no. 4, pp. 1003–1012, 2009.
- J. Y. Ma, J. L. Tong, and Z. H. Ran, “Intestinal tuberculosis and Crohn's disease: Challenging differential diagnosis,” Journal of Digestive Diseases, vol. 17, no. 3, pp. 155–161, 2016.
- T. Szul, P. Castaldi, M. H. Cho, J. Edwin Blalock, and A. Gaggar, “Genetic regulation of expression of leukotriene A4 hydrolase,” ERJ Open Research, vol. 2, no. 1, 2016.
- J. Z. Haeggström, “Structure, function, and regulation of leukotriene A4 hydrolase,” American Journal of Respiratory and Critical Care Medicine, vol. 161, no. 2, pp. S25–S31, 2000.
- J. J. Lima, S. Zhang, A. Grant et al., “Influence of leukotriene pathway polymorphisms on response to montelukast in asthma,” American Journal of Respiratory and Critical Care Medicine, vol. 173, no. 4, pp. 379–385, 2006.
- J. Zhao, J. Goldberg, and V. Vaccarino, “Leukotriene A4 hydrolase haplotype, diet and atherosclerosis: A twin study,” Atherosclerosis, vol. 226, no. 1, pp. 238–244, 2013.
- V. M. Fava and E. Schurr, “Evaluating the impact of LTA4H genotype and immune status on survival from tuberculous meningitis,” The Journal of Infectious Diseases, vol. 215, no. 7, pp. 1011–1013, 2017.
- E. S. Amirian, M. M. Ittmann, and M. E. Scheurer, “Associations between arachidonic acid metabolism gene polymorphisms and prostate cancer risk,” The Prostate, vol. 71, no. 13, pp. 1382–1389, 2011.
- D. M. Tobin, F. J. Roca, S. F. Oh et al., “Host genotype-specific therapies can optimize the inflammatory response to mycobacterial infections,” Cell, vol. 148, no. 3, pp. 434–446, 2012.
- J. Yang, J. Chen, J. Yue, L. Liu, M. Han, and H. Wang, “Relationship between human LTA4H polymorphisms and extra-pulmonary tuberculosis in an ethnic Han Chinese population in Eastern China,” Tuberculosis, vol. 94, no. 6, pp. 657–663, 2014.
- J. Jupp, K. Hillier, D. H. Elliott et al., “Colonic expression of leukotriene-pathway enzymes in inflammatory bowel diseases,” Inflammatory Bowel Diseases, vol. 13, no. 5, pp. 537–546, 2007.
- W. D. Dupont and W. D. Plummer Jr., “Power and sample size calculations for studies involving linear regression,” Controlled Clinical Trials, vol. 19, no. 6, pp. 589–601, 1998.
- X.-S. Chen, J. R. Sheller, E. N. Johnson, and C. D. Funk, “Role of leukotrienes revealed by targeted disruption of the 5-lipoxygenase gene,” Nature, vol. 372, no. 6502, pp. 179–182, 1994.
- M. Divangahi, D. Desjardins, C. Nunes-Alves, H. G. Remold, and S. M. Behar, “Eicosanoid pathways regulate adaptive immunity to Mycobacterium tuberculosis,” Nature Immunology, vol. 11, no. 8, pp. 751–758, 2010.
- T. Lämmermann, P. V. Afonso, B. R. Angermann et al., “Neutrophil swarms require LTB4 and integrins at sites of cell death in vivo,” Nature, vol. 498, no. 7454, pp. 371–375, 2013.
- C. E. Barry III, H. I. Boshoff, V. Dartois et al., “The spectrum of latent tuberculosis: rethinking the biology and intervention strategies,” Nature Reviews Microbiology, vol. 7, no. 12, pp. 845–855, 2009.
- I. S. Leal, M. Flórido, P. Andersen, and R. Appelberg, “Interleukin-6 regulates the phenotype of the immune response to a tuberculosis subunit vaccine,” The Journal of Immunology, vol. 103, no. 3, pp. 375–381, 2001.
- W. R. Henderson Jr., “The role of leukotrienes in inflammation,” Annals of Internal Medicine, vol. 121, no. 9, pp. 684–697, 1994.
- N. Gualde, D. Atluru, and J. S. Goodwin, “Effect of lipoxygenase metabolites of arachidonic acid on proliferation of human T cells and T cell subsets,” The Journal of Immunology, vol. 134, no. 2, pp. 1125–1129, 1985.
- K. A. Yamaoka, H. E. Claesson, and A. Rosen, “Leukotriene B4 enhances activation, proliferation, and differentiation of human B lymphocytes,” The Journal of Immunology, vol. 143, no. 6, pp. 1996–2000, 1989.
- D. M. Tobin, J. C. Vary Jr., J. P. Ray et al., “The lta4h locus modulates susceptibility to mycobacterial infection in zebrafish and humans,” Cell, vol. 140, no. 5, pp. 717–730, 2010.
- H. Clay, H. E. Volkman, and L. Ramakrishnan, “Tumor necrosis factor signaling mediates resistance to Mycobacteria by inhibiting bacterial growth and macrophage death,” Immunity, vol. 29, no. 2, pp. 283–294, 2008.
- S. J. Dunstan, T. T. Tram, G. E. Thwaites et al., “LTA4H genotype is associated with susceptibility to bacterial meningitis but is not a critical determinant of outcome,” PLoS ONE, vol. 10, no. 3, p. e0118789, 2015.
- J. Curtis, L. Kopanitsa, E. Stebbings et al., “Association analysis of the LTA4H gene polymorphisms and pulmonary tuberculosis in 9115 subjects,” Tuberculosis, vol. 91, no. 1, pp. 22–25, 2011.
- M. Caws, G. Thwaites, S. Dunstan et al., “The influence of host and bacterial genotype on the development of disseminated disease with Mycobacterium tuberculosis,” PLoS Pathogens, vol. 4, no. 3, Article ID e1000034, 2008.
- E. H. Choi and W. J. Coyle, “Gastrointestinal tuberculosis,” Microbiology Spectrum, vol. 4, no. 6, 2016.
- D. R. Shores, D. G. Binion, B. A. Freeman, and P. R. S. Baker, “New insights into the role of fatty acids in the pathogenesis and resolution of inflammatory bowel disease,” Inflammatory Bowel Diseases, vol. 17, no. 10, pp. 2192–2204, 2011.
- W. F. Stenson, “The universe of arachidonic acid metabolites in inflammatory bowel disease: Can we tell the good from the bad?” Current Opinion in Gastroenterology, vol. 30, no. 4, pp. 347–351, 2014.
- I. Costea, D. R. Mack, D. Israel et al., “Genes involved in the metabolism of poly-unsaturated fatty-acids (PUFA) and risk for Crohn's disease in children & young adults,” PLoS One, vol. 5, no. 12, Article ID e15672, 2010.
- K. Lauritsen, L. S. Laursen, K. Bukhave, and J. Rask-Madsen, “Effects of topical 5-aminosalicylic acid and prednisolone on prostaglandin E2 and leukotriene B4 levels determined by equilibrium in vivo dialysis of rectum in relapsing ulcerative colitis,” Gastroenterology, vol. 91, no. 4, pp. 837–844, 1986.
- A. E. Starr, S. A. Deeke, Z. Ning et al., “Proteomic analysis of ascending colon biopsies from a paediatric inflammatory bowel disease inception cohort identifies protein biomarkers that differentiate Crohn's disease from UC,” Gut, vol. 66, no. 9, pp. 1573–1583, 2017.
- S. A. Deeke, A. E. Starr, Z. Ning et al., “Mucosal-luminal interface proteomics reveals biomarkers of pediatric inflammatory bowel disease-associated colitis,” American Journal of Gastroenterology, vol. 113, no. 5, pp. 713–724, 2018.
Copyright © 2019 Zi-qi Yu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.