- Email: [email protected]
The functional polymorphisms of LIS1 are associated with acute myeloid leukemia risk in a Han Chinese population
Leukemia Research 54 (2017) 7–11
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
The functional polymorphisms of LIS1 are associated with acute myeloid leukemia risk in a Han Chinese population Songyu Cao a , Xiaomei Lu b , Lihua Wang a , Xifeng Qian c , Guangfu Jin a , Hongxia Ma a,∗ a Department of Epidemiology and Biostatistics, Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Collaborative Innovation Center For Cancer Personalized Medicine, School of Public Health, Nanjing Medical University, Nanjing 211166, China b Department of State-Level Model Center of Experimental Teaching, School of Public Health, Nanjing Medical University, Nanjing 211166, China c Department of Hematology, Wuxi Peoples’s Hospital Affiliated to Nanjing Medical University, No. 299 Qingyang Road, Wuxi 214194, China
a r t i c l e
i n f o
Article history: Received 22 September 2016 Received in revised form 13 December 2016 Accepted 28 December 2016 Available online 2 January 2017 Keywords: Acute leukemia LIS1 Polymorphism Susceptibility
a b s t r a c t There is increasing evidence that the human lissencephaly-1 gene, LIS1, plays an important role in carcinogenesis of several malignancies including leukemia. However, little is known about the relationship between single nucleotide polymorphisms (SNPs) in LIS1 and the susceptibility to myeloid leukemia. In the present study, we systematically screened 5 potentially functional polymorphisms in LIS1, and conducted a case-control study including 660 acute myeloid leukemia (AML) patients and 1034 cancer-free controls in a Chinese population, to assess the association between these SNPs and AML risk. We found that the variant alleles of rs4790348, rs4790353, and rs7209748 could significantly increase the AML risk (rs4790348: adjusted OR = 1.31, 95%CI = 1.13–1.53 in additive model; rs4790353: adjusted OR = 4.97, 95%CI = 1.59–15.50 in recessive model; rs7209748: adjusted OR = 2.34, 95%CI = 1.11–4.94 in recessive model). These findings indicated that genetic variants in LIS1 may contribute to AML risk in Chinese population. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction As a kind of fatal hematopoietic stem cell tumor, acute myeloid leukemia (AML) is caused by leukemia cells which invade bone marrow and result in normal hematopoiesis inhibition [1–3]. Several environmental factors have been identified to play important roles in AML development, such as benzene exposure, ionizing radiation, and chemotherapy [4–6]. Furthermore, it is generally accepted that genetic factors are also involved in the pathogenesis of AML. Recently, increasing evidence has shown that the human lissencephaly-1 gene (LIS1, also known as PAFAH1B1) plays important roles in carcinogenesis of several malignancies. It was reported that the mRNA and protein levels of LIS1 were downregulated in about 70% of hepatocellular carcinoma (HCC) tissues, and such alteration was significantly associated with tumor progression [7]. LIS1 was also found to play a role in neuroblastoma cell nucleokinesis and motility [8], glioma migration and proliferation [9], and cholangiocarcinoma cell proliferation and invasion [10]. Besides, Zimdahl et al. reported that conditional deletion of LIS1 in hematopoietic cells could lead to a dramatic “bloodless” pheno-
∗ Corresponding author. E-mail address: [email protected] (H. Ma). http://dx.doi.org/10.1016/j.leukres.2016.12.007 0145-2126/© 2017 Elsevier Ltd. All rights reserved.
type, impaired stem cell function, depletion of the stem cell pool, and accelerated differentiation. They also found that the inhibition of LIS1 increased the differentiation in the short term and blocked the growth in the longer term, identifying LIS1 as a new critical regulator of human leukemia growth and propagation [11]. Although much attention has been focused on the role of LIS1 in carcinogenesis, little is known about the relationship between single nucleotide polymorphisms (SNPs) in LIS1 and the susceptibility to AML. In this study, we hypothesized that potentially functional polymorphisms in LIS1 may alter the expression or function of this gene and consequently influence AML risk. Thus, we systematically screened 5 potentially functional polymorphisms in LIS1, and conducted a case-control study including 660 AML patients and 1034 cancer-free controls to test this hypothesis. 2. Materials and methods 2.1. Ethical approval This work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) and approved by the institutional review board of Nanjing Medical University. A written informed consent was obtained for all the participants. The privacy of all the participants was protected.
8
S. Cao et al. / Leukemia Research 54 (2017) 7–11
2.2. Study subjects A total of 660 AML cases were recruited from Wuxi people’s Hospital Affiliated to Nanjing Medical University from December 2007 to June 2013. All patients diagnosed according to FAB diagnosis and typing standard were included, with no restrictions in terms of age, histology, or stage of disease, but patients with a history of cancer, chemotherapy or hematopoietic stem cell transplantation (HSCT) were excluded. The 1034 controls were randomly selected from more than 30,000 participants in a community screening of noncommunicable disease in Jiangsu Province, conducted during the same period as the cases were recruited. All controls had no selfreported cancer history. All the participants denoted 5 ml of venous blood sample and undergone a face to face interview concerning demographic data (e.g. age and gender), and clinical information of cases were gathered from patients’ medical records, including date of diagnosis, lineage, classification of diagnosis, karyotype, and molecular subtype. 2.3. SNP selection The UCSC database (http://genome.ucsc.edu/, hg19 assembly, Feb. 2009), SNPinfo Web Server (http://snpinfo.niehs.nih.gov/) and RegulomeDB (http://regulomedb.org/) were used to identify common SNPs with potential functions in LIS1 gene region, meeting the following two criteria: (1) with a minor allele frequency (MAF) of more than 5% in Chinese population; and (2) with a predicted function of influencing the binding of specified transcription factors or being expression quantitative trait loci (eQTLs). Linkage disequilibrium (LD) analysis with an r2 threshold of 0.80 was further applied to filter these functional SNPs. As a result, five SNPs were selected for genotyping, including rs1266474, rs4790348, rs4790353, rs7209748, and rs8081803. 2.4. Laboratory assays The method of traditional proteinase K digestion followed by phenol-chloroform extraction and ethanol precipitation were used to extract genomic DNA. All SNPs were genotyped on an ABI 7900 system (Applied Biosystems, Foster city, CA) using the TaqMan allelic discrimination assay. The sequence information of primers and probes are shown in Supplymentary Table 1. The genotyping assays were performed without knowing the subjects’ case or control status. More than 10% samples were randomly selected to repeat, and the results were 100% concordant. The genotyping call rates for these polymorphisms were all above 98%. 2.5. Statistical analysis The Student’s t-test and 2 tests were used to evaluate differences between cases and controls on the demographic characteristics, for continuous variables and categorical variables, respectively. Hardy-Weinberg equilibrium (HWE) was tested by a goodness-of-fit 2 test. The associations between LIS1 SNPs and AML risks were estimated by multivariate logistic regression analysis. Likelihood Ratio test was used for stratification analysis. Power was calculated using the Power and Sample Size Calculation (PS) software (v3.1.2), and we set the type I error probability and OR as 0.05 and 1.5, respectively. P < 0.05 was the criterion of statistical significance, and all statistical tests were two sided, performed with SAS 9.1.3 software (SAS Institute, Cary, NC). 3. Results The information of demographic and clinical characteristics of the study subjects was shown in Table 1. The age (44.4 ± 17.8 versus
45.4 ± 9.6) and gender (52.7% versus 56.7% for male) between cases and controls were comparable (P > 0.05). For clinical and cytogenetic characteristics, most cases were M2 AML (33.5%), of myeloid lineage (70.3%), with intermediate Southwest Oncology Group (SWOG) risk assessment (50.9%), and without common fusion gene transcripts (57.1%). The basic information of selected SNPs was presented in Table 2, including locations, alleles, MAF among cases and controls, HWE test among controls, call rate of all samples, and the achieved power. All 5 SNPs located in the intron region of LIS1 had call rate of more than 98% and were consistent with Hardy-Weinberg equilibrium (P > 0.05). Although we selected all SNPs with a MAF of more than 5% in Chinese population according the SNPinfo Web Server (http://snpinfo.niehs.nih.gov/), rs8081803 showed a MAF of 0.049 in the present study. Additionally, two SNPs (rs4790353 and rs8081803) achieved a power of less than 80% (66.3% and 49.8%) when we assumed the type I error probability and OR as 0.05 and 1.5, respectively. We conducted the association analysis between selected SNPs and AML risk in different genetic models, and all analyses were adjusted for age and gender. As shown in Table 3, three SNPs were associated with the increased risk of AML (P < 0.05). The variant A allele of rs4790348 increased AML risk in dominant and additive model (adjusted OR = 1.43, 95%CI = 1.18–1.75 in dominant model; adjusted OR = 1.31, 95%CI = 1.13–1.53 in additive model), and the variant alleles of rs4790353 and rs7209748 increased the risk of AML in recessive model (rs4790353: adjusted OR = 4.97, 95%CI = 1.59–15.50; adjusted rs7209748: OR = 2.34, 95%CI = 1.11–4.94). There was no significant evidence for the relationship between other 2 SNPs and AML risk (Table 3). As 28.2% cases had mixed lineage acute leukemia (myeloid and lymphoid, Table 1), we also conducted the association analysis when those 28.2% patients were excluded and found the similar results (rs4790348: adjusted OR = 1.32, 95%CI = 1.06–1.65 in dominant model and adjusted OR = 1.22, 95%CI = 1.03–1.45 in additive model; rs4790353: adjusted OR = 5.35, 95%CI = 1.63–17.53 in recessive model; rs7209748: adjusted OR = 2.49, 95%CI = 1.13–5.51 in recessive model, Table 3). Then, we examined the association of these 3 SNPs with risk of AML in subgroups stratified by selected variables including age, gender, lineage, SWOG risk assessment and molecular subtype. However, no heterogeneity was found for these 3 SNPs among any subgroup (Table 4). As the fusion gene AML1/ETO and PML/RAR␣ respectively accounted for the majority of molecular subtypes of M2 and M3 AML, we then investigated the association of AML1/ETO and PML/RAR␣ with rs4790348, rs4790353 and rs7209748 in M2 and M3 AML. As shown in Supplymentary Table 2, there was no significant association between AML1/ETO or PML/RAR␣ and these 3 SNPs in M2 and M3 AML.
4. Discussion In the present study, we genotyped 5 potentially functional SNPs of LIS1 in 660 AML patients and 1034 cancer-free controls, and indicated that the variant alleles of rs4790348, rs4790353, and rs7209748 could significantly increase the AML risk in Chinese population. The findings suggested that genetic variants of LIS1 may contribute to AML susceptibility. LIS1 gene encoding the LIS1 protein is located in the region of chromosome 17p13.3. The LIS1 protein colocalizes with cytoplasmic dynein and dynactin, predominantly in prometaphase kinetochores and at the cell cortex of dividing cells [12,13]. LIS1/dynactin plays the function in metaphase spindle assembly and mitotic checkpoint control [12,14,15]. Besides, LIS1 has
S. Cao et al. / Leukemia Research 54 (2017) 7–11
9
Table 1 Demographic and clinical characteristics of study subjects. Characteristics
AML(n = 660)
Controls(n = 1034)
P
N
(%)
N
(%)
Gender Male Female
348 312
52.7 47.3
586 448
56.7 43.3
Age at diagnosis, year mean ± sd ≤45 >45
44.4 ± 17.8 356 304
53.9 46.1
45.4 ± 9.6 510 524
49.3 50.7
Lineage Myeloid Myeloid and Lymphoid
474 186
71.8 28.2
Classification of diagnosis M0 M1 M2 M3 M4 M5 M6 Unknown
5 119 221 119 62 103 4 27
0.8 18.0 33.5 18.0 9.4 15.6 0.6 4.1
109 336 151 64
16.5 50.9 22.9 9.7
96 72 60 377 55
14.5 10.9 9.1 57.1 8.3
Cytogenetic risk categoryb Favorable Intermediate Unfavorable Unknown Molecular subtype Common fusion gene transcripts PML/RAR␣ AML1/ETO Others No common fusion gene ranscripts Unknown a b
0.111
0.178a
P value of independent-sample T test for age. Grouped by Southwest Oncology Group (SWOG) criteria.
Table 2 Basic information of the 5 SNPs of LIS1. SNP
Location
Allele
MAFa
MAFb
HWEc
Call rate
Powerd
rs1266474 rs4790348 rs4790353 rs7209748 rs8081803
intron intron intron intron intron
A>G G>A T>G G>C T>C
0.130 0.321 0.079 0.131 0.053
0.116 0.264 0.077 0.125 0.049
0.351 0.883 0.355 0.231 0.755
0.988 0.989 0.991 0.988 0.995
0.806 0.963 0.663 0.828 0.498
a b c d
Minor allele frequency among case subjects. Minor allele frequency among control subjects. P value of hardy-Weinberg equilibrium test among control subjects. Power was calculated using Power and Sample Size Calculation software, as the type I error probability was set as 0.05, and OR was set as 1.5.
been involved in the development of some malignancies, including HCC, neuroblastoma, glioma, and cholangiocarcinoma [7–10]. Zimdahl et al. demonstrated that LIS1 deficiency impaired primitive (fetal) hematopoietic stem cells (HSCs) expansion and caused rapid depletion of definitive (adult) HSCs [11,16]. They also found that transformed hematopoietic cells in AML rely on LIS1-mediated regulation of cell division mode during a critical phase of leukemogenesis, and knockdown of LIS1 in primary human-derived CD34+ cells led to significantly impaired leukemic growth [11,16]. All three SNPs (rs4790348, rs4790353, and rs7209748) located in the intron region of LIS1, and none of them had been reported so far. To investigate the potentially functional significance of these SNPs, we used an online prediction tool, RegulomeDB (http:// regulomedb.org/), which implemented the data of the Encyclopedia of DNA Elements (ENCODE) [17]. All three SNPs were predicted to influence the chromatin structure and histone modifications. Additionally, rs4790348 might influence the binding of several
transcription factors, such as POLR3A and Zscan4, and is a trans eQTL linked to the expression of RTF1 [18]. Meanwhile, rs4790353 and rs7209748 were also located in the binding sites of some transcriptional factors including Brachyury and Irx3. All these predictions indicated that these three SNPs may had potential function to affect transcriptional regulatory of LIS1 in the development of AML. Several limitations need to be addressed in our study. Firstly, the present study was based on candidate gene approach, which could only detect limited genes and SNPs on AML. Secondly, the sample size might be relatively small, as two SNPs (rs4790353 and rs8081803) didn’t achieve a statistical power of greater than 80% when we assumed the OR was 1.5. Thirdly, we lacked environmental information like ionizing radiation and benzene, so that gene-environment interaction cannot be evaluated. Therefore, studies with larger sample size and more environmental informa-
10
S. Cao et al. / Leukemia Research 54 (2017) 7–11
Table 3 Association between selected SNPs and AML risk in different models. AMLa (n = 660)
AMLb (n = 474)
Control (n = 1034)
Adjusted OR (95%CI)a
Pa
Adjusted OR(95%CI)b
Pb
rs1266474 A > G AA AG GG Dominant model Recessive model Additive model
491 131 18 – – –
350 94 14
811 206 17 – – –
– 1.04(0.81, 1.33) 1.82(0.93, 3.57) 1.10(0.87, 1.39) 1.80(0.92, 3.53) 1.14(0.92, 1.39)
– 0.747 0.083 0.432 0.086 0.227
– 1.04(0.79, 1.37) 2.01(0.98, 4.14) 1.11(0.86, 1.45) 1.99(0.97, 4.10) 1.16(0.92, 1.45)
– 0.778 0.058 0.430 0.060 0.207
rs4790348 G > A GG GA AA Dominant model Recessive model Additive model
289 293 59 – – –
215 206 37
561 400 73 – – –
– 1.41(1.15, 1.74) 1.55(1.07, 2.26) 1.43(1.18, 1.75) 1.33(0.93, 1.90) 1.31(1.13, 1.53)
– 0.001 0.020 <0.001 0.122 0.001
– 1.33(1.05, 1.67) 1.31(0.85, 2.01) 1.32(1.06, 1.65) 1.15(0.76, 1.74) 1.22(1.03, 1.45)
– 0.017 0.217 0.013 0.504 0.024
rs4790353 T > G TT TG GG Dominant model Recessive model Additive model
555 78 12 – – –
396 56 9
879 151 4 – – –
– 0.82(0.61, 1.09) 4.83(1.55, 15.09) 0.92(0.69, 1.22) 4.97(1.59, 15.50) 1.03(0.80, 1.33)
– 0.173 0.007 0.548 0.006 0.829
– 0.82(0.59, 1.14) 5.21(1.59, 17.08) 0.93(0.68, 1.27) 5.35(1.63, 17.53) 1.05(0.79, 1.39)
– 0.239 0.007 0.650 0.006 0.747
rs7209748 G > C GG GC CC Dominant model Recessive model Additive model
488 134 17 – – –
349 95 13
787 235 12 – – –
– 0.92(0.72, 1.17) 2.30(1.09, 4.86) 0.99(0.78, 1.25) 2.34(1.11, 4.94) 1.06(0.86, 1.30)
– 0.506 0.030 0.921 0.026 0.588
– 0.92(0.70, 1.20) 2.44(1.10, 5.42) 0.99(0.76, 1.28) 2.49(1.13, 5.51) 1.07(0.85, 1.35)
– 0.524 0.028 0.942 0.025 0.571
rs8081803 T > C TT TC CC Dominant model Recessive model Additive model
586 63 3 – – –
420 43 3
935 97 2 – – –
– 1.05(0.75, 1.46) 2.35(0.39, 14.14) 1.07(0.77, 1.49) 2.34(0.39, 14.08) 1.10(0.80, 1.50)
– 0.791 0.351 0.674 0.354 0.569
– 1.00(0.69, 1.47) 3.13(0.52, 18.88) 1.05(0.73, 1.52) 3.13(0.52, 18.87) 1.09(0.77, 1.54)
– 0.987 0.213 0.801 0.213 0.627
a b
Those patients with myeloid and lymphoid lineage were included. Those patients with myeloid and lymphoid lineage were excluded.
Table 4 Stratification analysis of genotypes of the 3 SNPs by selected variables in AML patients and controls. Characteristics
rs4790348a
rs4790353b
rs7209748b
OR(95%CI)c
Pd
OR(95%CI)c
Pd
OR(95%CI)c
Pd
Age ≤45 >45
1.31(1.00, 1.72) 1.40(1.05, 1.88)
0.743
12.57(1.56, 100.98) 2.42(0.54, 10.90)
0.209
5.11(1.40, 18.72) 1.38(0.51, 3.75)
0.117
Gender Male Female
1.33(0.97, 1.81) 1.31(0.95, 1.80)
0.947
5.22(1.40, 19.42) 4.28(0.44, 41.50)
0.882
1.75(0.65, 4.72) 3.47(1.06, 11.39)
0.386
Lineage Myeloid Myeloid and Lymphoid
1.32(1.06, 1.65) 1.75(1.27, 2.41)
0.362
5.49(1.68, 18.00) 3.97(0.87, 18.04)
0.741
2.15(0.94, 4.93) 2.96(1.09, 8.05)
0.629
Cytogenetic risk categorye Favorable Intermediate Unfavorable
1.31(0.86, 1.98) 1.39(1.09, 1.79) 1.41(0.99, 2.00)
0.962
5.66(0.98, 32.75) 3.98(1.06, 14.93) 5.61(1.24, 25.45)
0.927
3.90(1.13, 13.47) 2.40(1.00, 5.75) 1.83(0.51, 6.58)
0.695
Molecular subtype PML/RAR␣ AML1/ETO Others No common fusion gene transcripts
1.24(0.80, 1.91) 1.37(0.84, 2.24) 1.88(1.09, 3.25) 1.44(1.13, 1.83)
0.704
6.66(1.23, 36.00) 5.71(0.61, 53.51) – 5.04(1.47, 17.37)
0.966
3.29(0.84, 12.79) 1.28(0.16, 10.37) 1.76(0.22, 13.95) 2.62(1.14, 5.99)
0.879
a b c d e
in dominant model. in recessive model. adjusted for age and gender. P value for heterogeneity test. Grouped by Southwest Oncology Group (SWOG) criteria.
S. Cao et al. / Leukemia Research 54 (2017) 7–11
tion were needed to further elucidate the impact of LIS1 SNPs on AML susceptibility. Conflicts of interest none. Acknowledgements This work was supported by the National Natural Science Foundation for Young Scholars of China [grant number 81402739]; Excellent Youth Foundation of Jiangsu Province [grant number BK20160046]; Priority Academic Program Development of Jiangsu Higher Education Institutions (Public Health and Preventive Medicine); and Top-notch Academic Programs Project of Jiangsu Higher Education Institutions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.leukres.2016. 12.007. References [1] M.R. O’Donnell, C.N. Abboud, J. Altman, F.R. Appelbaum, S.E. Coutre, L.E. Damon, J.M. Foran, S. Goorha, L.J. Maness, G. Marcucci, P. Maslak, M.M. Millenson, J.O. Moore, F. Ravandi, P.J. Shami, B.D. Smith, R.M. Stone, S.A. Strickland, M.S. Tallman, E.S. Wang, National comprehensive cancer network. Acute myeloid leukemia, J. Natl. Compr. Cancer Netw. 9 (March (3)) (2011) 280–317. [2] J.E. Rubnitz, B. Gibson, F.O. Smith, Acute myeloid leukemia, Pediatr. Clin. North Am. 55 (February (1)) (2008) 21–51, http://dx.doi.org/10.1016/j.pcl. 2007.11.003. [3] J.L. Shipley, J.N. Butera, Acute myelogenous leukemia, Exp. Hematol. 37 (June (6)) (2009) 649–658, http://dx.doi.org/10.1016/j.exphem.2009.04.002. [4] F. Ferrara, C.A. Schiffer, Acute myeloid leukaemia in adults, Lancet 381 (February (9865)) (2013) 484–495, http://dx.doi.org/10.1016/S01406736(12)61727-9. [5] M.R. O’Donnell, M.S. Tallman, C.N. Abboud, J.K. Altman, F.R. Appelbaum, D.A. Arber, E. Attar, U. Borate, S.E. Coutre, L.E. Damon, J. Lancet, L.J. Maness, G. Marcucci, M.G. Martin, M.M. Millenson, J.O. Moore, F. Ravandi, P.J. Shami, B.D. Smith, R.M. Stone, S.A. Strickland, E.S. Wang, K.M. Gregory, M. Naganuma, National comprehensive cancer network acute myeloid leukemia, version 2.2013, J. Natl. Compr. Cancer Netw. 11 (September (9)) (2013) 1047–1055.
11
[6] S. Fircanis, P. Merriam, N. Khan, J.J. Castillo, The relation between cigarette smoking and risk of acute myeloid leukemia: an updated meta-analysis of epidemiological studies, Am. J. Hematol. 89 (August (8)) (2014) E125–E132, http://dx.doi.org/10.1002/ajh.23744. [7] Z. Xing, X. Tang, Y. Gao, L. Da, H. Song, S. Wang, P. Tiollais, T. Li, M. Zhao, The human LIS1 is downregulated in hepatocellular carcinoma and plays a tumor suppressor function, Biochem. Biophys. Res. Commun. 409 (June (2)) (2011) 193–199, http://dx.doi.org/10.1016/j.bbrc.2011.04.117. [8] E. Messi, M.C. Florian, C. Caccia, M. Zanisi, R. Maggi, Retinoic acid reduces human neuroblastoma cell migration and invasiveness: effects on DCX LIS1, neurofilaments-68 and vimentin expression, BMC Cancer 8 (January (30)) (2008), http://dx.doi.org/10.1186/1471-2407-8-30. [9] S.O. Suzuki, R.J. McKenney, S.Y. Mawatari, M. Mizuguchi, A. Mikami, T. Iwaki, J.E. Goldman, P. Canoll, R.B. Vallee, Expression patterns of LIS1, dynein and their interaction partners dynactin, NudE, NudEL and NudC in human gliomas suggest roles in invasion and proliferation, Acta Neuropathol. 113 (May (5)) (2007) 591–599. [10] R. Yang, Y. Chen, C. Tang, H. Li, B. Wang, Q. Yan, J.1 Hu, S. Zou, MicroRNA-144 suppresses cholangiocarcinoma cell proliferation and invasion through targeting platelet activating factor acetylhydrolase isoform 1b, BMC Cancer 5 (December (14)) (2014) 917, http://dx.doi.org/10.1186/1471-2407-14-917. [11] B. Zimdahl, T. Ito, A. Blevins, J. Bajaj, T. Konuma, J. Weeks, C.S. Koechlein, H.Y. Kwon, O. Arami, D. Rizzieri, H.E. Broome, C. Chuah, V.G. Oehler, R. Sasik, G. Hardiman, T. Reya, Lis1 regulates asymmetric division in hematopoietic stem cells and in leukemia, Nat. Genet. 46 (March (3)) (2014) 245–252, http://dx. doi.org/10.1038/ng.2889. [12] K.H. Siller, C.Q. Doe, Lis1/dynactin regulates metaphase spindle orientation in Drosophila neuroblasts, Dev. Biol. 319 (July (1)) (2008) 1–9, http://dx.doi.org/ 10.1016/j.ydbio.2008.03.018. [13] J. Yingling, Y.H. Youn, D. Darling, K. Toyo-Oka, T. Pramparo, S. Hirotsune, A. Wynshaw-Boris, Neuroepithelial stem cell proliferation requires LIS1 for precise spindle orientation and symmetric division, Cell 132 (February (3)) (2008) 474–486, http://dx.doi.org/10.1016/j.cell.2008.01.026. [14] K.H. Siller, M. Serr, R. Steward, T.S. Hays, C.Q. Doe, Live imaging of Drosophila brain neuroblasts reveals a role for Lis1/dynactin in spindle assembly and mitotic checkpoint control, Mol. Biol. Cell 16 (November (11)) (2005) 5127–5140. [15] N.E. Faulkner, D.L. Dujardin, C.Y. Tai, K.T. Vaughan, C.B. O’Connell, Y. Wang, R.B. Vallee, A role for the lissencephaly gene LIS1 in mitosis and cytoplasmic dynein function, Nat. Cell Biol. 2 (November (11)) (2000) 784–791. [16] B. Will, U. Steidl, Stem cell fate regulation by dynein motor protein Lis1, Nat. Genet. 46 (March (3)) (2014) 217–218, http://dx.doi.org/10.1038/ng.2902. [17] A.P. Boyle, E.L. Hong, M. Hariharan, Y. Cheng, M.A. Schaub, M. Kasowski, K.J. Karczewski, J. Park, B.C. Hitz, S. Weng, J.M. Cherry, M. Snyder, Annotation of functional variation in personal genomes using RegulomeDB, Genome Res. Sep. 22 (9) (2012) 1790–1797, http://dx.doi.org/10.1101/gr.137323.112. [18] G. Badis, M.F. Berger, A.A. Philippakis, S. Talukder, A.R. Gehrke, S.A. Jaeger, E.T. Chan, G. Metzler, A. Vedenko, X. Chen, H. Kuznetsov, C.F. Wang, D. Coburn, D.E. Newburger, Q. Morris, T.R. Hughes, M.L. Bulyk, Diversity and complexity in DNA recognition by transcription factors, Science 324 (June (5935)) (2009) 1720–1723, http://dx.doi.org/10.1126/science.1162327.
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
The functional polymorphisms of LIS1 are associated with acute myeloid leukemia risk in a Han Chinese population Songyu Cao a , Xiaomei Lu b , Lihua Wang a , Xifeng Qian c , Guangfu Jin a , Hongxia Ma a,∗ a Department of Epidemiology and Biostatistics, Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Collaborative Innovation Center For Cancer Personalized Medicine, School of Public Health, Nanjing Medical University, Nanjing 211166, China b Department of State-Level Model Center of Experimental Teaching, School of Public Health, Nanjing Medical University, Nanjing 211166, China c Department of Hematology, Wuxi Peoples’s Hospital Affiliated to Nanjing Medical University, No. 299 Qingyang Road, Wuxi 214194, China
a r t i c l e
i n f o
Article history: Received 22 September 2016 Received in revised form 13 December 2016 Accepted 28 December 2016 Available online 2 January 2017 Keywords: Acute leukemia LIS1 Polymorphism Susceptibility
a b s t r a c t There is increasing evidence that the human lissencephaly-1 gene, LIS1, plays an important role in carcinogenesis of several malignancies including leukemia. However, little is known about the relationship between single nucleotide polymorphisms (SNPs) in LIS1 and the susceptibility to myeloid leukemia. In the present study, we systematically screened 5 potentially functional polymorphisms in LIS1, and conducted a case-control study including 660 acute myeloid leukemia (AML) patients and 1034 cancer-free controls in a Chinese population, to assess the association between these SNPs and AML risk. We found that the variant alleles of rs4790348, rs4790353, and rs7209748 could significantly increase the AML risk (rs4790348: adjusted OR = 1.31, 95%CI = 1.13–1.53 in additive model; rs4790353: adjusted OR = 4.97, 95%CI = 1.59–15.50 in recessive model; rs7209748: adjusted OR = 2.34, 95%CI = 1.11–4.94 in recessive model). These findings indicated that genetic variants in LIS1 may contribute to AML risk in Chinese population. © 2017 Elsevier Ltd. All rights reserved.
1. Introduction As a kind of fatal hematopoietic stem cell tumor, acute myeloid leukemia (AML) is caused by leukemia cells which invade bone marrow and result in normal hematopoiesis inhibition [1–3]. Several environmental factors have been identified to play important roles in AML development, such as benzene exposure, ionizing radiation, and chemotherapy [4–6]. Furthermore, it is generally accepted that genetic factors are also involved in the pathogenesis of AML. Recently, increasing evidence has shown that the human lissencephaly-1 gene (LIS1, also known as PAFAH1B1) plays important roles in carcinogenesis of several malignancies. It was reported that the mRNA and protein levels of LIS1 were downregulated in about 70% of hepatocellular carcinoma (HCC) tissues, and such alteration was significantly associated with tumor progression [7]. LIS1 was also found to play a role in neuroblastoma cell nucleokinesis and motility [8], glioma migration and proliferation [9], and cholangiocarcinoma cell proliferation and invasion [10]. Besides, Zimdahl et al. reported that conditional deletion of LIS1 in hematopoietic cells could lead to a dramatic “bloodless” pheno-
∗ Corresponding author. E-mail address: [email protected] (H. Ma). http://dx.doi.org/10.1016/j.leukres.2016.12.007 0145-2126/© 2017 Elsevier Ltd. All rights reserved.
type, impaired stem cell function, depletion of the stem cell pool, and accelerated differentiation. They also found that the inhibition of LIS1 increased the differentiation in the short term and blocked the growth in the longer term, identifying LIS1 as a new critical regulator of human leukemia growth and propagation [11]. Although much attention has been focused on the role of LIS1 in carcinogenesis, little is known about the relationship between single nucleotide polymorphisms (SNPs) in LIS1 and the susceptibility to AML. In this study, we hypothesized that potentially functional polymorphisms in LIS1 may alter the expression or function of this gene and consequently influence AML risk. Thus, we systematically screened 5 potentially functional polymorphisms in LIS1, and conducted a case-control study including 660 AML patients and 1034 cancer-free controls to test this hypothesis. 2. Materials and methods 2.1. Ethical approval This work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) and approved by the institutional review board of Nanjing Medical University. A written informed consent was obtained for all the participants. The privacy of all the participants was protected.
8
S. Cao et al. / Leukemia Research 54 (2017) 7–11
2.2. Study subjects A total of 660 AML cases were recruited from Wuxi people’s Hospital Affiliated to Nanjing Medical University from December 2007 to June 2013. All patients diagnosed according to FAB diagnosis and typing standard were included, with no restrictions in terms of age, histology, or stage of disease, but patients with a history of cancer, chemotherapy or hematopoietic stem cell transplantation (HSCT) were excluded. The 1034 controls were randomly selected from more than 30,000 participants in a community screening of noncommunicable disease in Jiangsu Province, conducted during the same period as the cases were recruited. All controls had no selfreported cancer history. All the participants denoted 5 ml of venous blood sample and undergone a face to face interview concerning demographic data (e.g. age and gender), and clinical information of cases were gathered from patients’ medical records, including date of diagnosis, lineage, classification of diagnosis, karyotype, and molecular subtype. 2.3. SNP selection The UCSC database (http://genome.ucsc.edu/, hg19 assembly, Feb. 2009), SNPinfo Web Server (http://snpinfo.niehs.nih.gov/) and RegulomeDB (http://regulomedb.org/) were used to identify common SNPs with potential functions in LIS1 gene region, meeting the following two criteria: (1) with a minor allele frequency (MAF) of more than 5% in Chinese population; and (2) with a predicted function of influencing the binding of specified transcription factors or being expression quantitative trait loci (eQTLs). Linkage disequilibrium (LD) analysis with an r2 threshold of 0.80 was further applied to filter these functional SNPs. As a result, five SNPs were selected for genotyping, including rs1266474, rs4790348, rs4790353, rs7209748, and rs8081803. 2.4. Laboratory assays The method of traditional proteinase K digestion followed by phenol-chloroform extraction and ethanol precipitation were used to extract genomic DNA. All SNPs were genotyped on an ABI 7900 system (Applied Biosystems, Foster city, CA) using the TaqMan allelic discrimination assay. The sequence information of primers and probes are shown in Supplymentary Table 1. The genotyping assays were performed without knowing the subjects’ case or control status. More than 10% samples were randomly selected to repeat, and the results were 100% concordant. The genotyping call rates for these polymorphisms were all above 98%. 2.5. Statistical analysis The Student’s t-test and 2 tests were used to evaluate differences between cases and controls on the demographic characteristics, for continuous variables and categorical variables, respectively. Hardy-Weinberg equilibrium (HWE) was tested by a goodness-of-fit 2 test. The associations between LIS1 SNPs and AML risks were estimated by multivariate logistic regression analysis. Likelihood Ratio test was used for stratification analysis. Power was calculated using the Power and Sample Size Calculation (PS) software (v3.1.2), and we set the type I error probability and OR as 0.05 and 1.5, respectively. P < 0.05 was the criterion of statistical significance, and all statistical tests were two sided, performed with SAS 9.1.3 software (SAS Institute, Cary, NC). 3. Results The information of demographic and clinical characteristics of the study subjects was shown in Table 1. The age (44.4 ± 17.8 versus
45.4 ± 9.6) and gender (52.7% versus 56.7% for male) between cases and controls were comparable (P > 0.05). For clinical and cytogenetic characteristics, most cases were M2 AML (33.5%), of myeloid lineage (70.3%), with intermediate Southwest Oncology Group (SWOG) risk assessment (50.9%), and without common fusion gene transcripts (57.1%). The basic information of selected SNPs was presented in Table 2, including locations, alleles, MAF among cases and controls, HWE test among controls, call rate of all samples, and the achieved power. All 5 SNPs located in the intron region of LIS1 had call rate of more than 98% and were consistent with Hardy-Weinberg equilibrium (P > 0.05). Although we selected all SNPs with a MAF of more than 5% in Chinese population according the SNPinfo Web Server (http://snpinfo.niehs.nih.gov/), rs8081803 showed a MAF of 0.049 in the present study. Additionally, two SNPs (rs4790353 and rs8081803) achieved a power of less than 80% (66.3% and 49.8%) when we assumed the type I error probability and OR as 0.05 and 1.5, respectively. We conducted the association analysis between selected SNPs and AML risk in different genetic models, and all analyses were adjusted for age and gender. As shown in Table 3, three SNPs were associated with the increased risk of AML (P < 0.05). The variant A allele of rs4790348 increased AML risk in dominant and additive model (adjusted OR = 1.43, 95%CI = 1.18–1.75 in dominant model; adjusted OR = 1.31, 95%CI = 1.13–1.53 in additive model), and the variant alleles of rs4790353 and rs7209748 increased the risk of AML in recessive model (rs4790353: adjusted OR = 4.97, 95%CI = 1.59–15.50; adjusted rs7209748: OR = 2.34, 95%CI = 1.11–4.94). There was no significant evidence for the relationship between other 2 SNPs and AML risk (Table 3). As 28.2% cases had mixed lineage acute leukemia (myeloid and lymphoid, Table 1), we also conducted the association analysis when those 28.2% patients were excluded and found the similar results (rs4790348: adjusted OR = 1.32, 95%CI = 1.06–1.65 in dominant model and adjusted OR = 1.22, 95%CI = 1.03–1.45 in additive model; rs4790353: adjusted OR = 5.35, 95%CI = 1.63–17.53 in recessive model; rs7209748: adjusted OR = 2.49, 95%CI = 1.13–5.51 in recessive model, Table 3). Then, we examined the association of these 3 SNPs with risk of AML in subgroups stratified by selected variables including age, gender, lineage, SWOG risk assessment and molecular subtype. However, no heterogeneity was found for these 3 SNPs among any subgroup (Table 4). As the fusion gene AML1/ETO and PML/RAR␣ respectively accounted for the majority of molecular subtypes of M2 and M3 AML, we then investigated the association of AML1/ETO and PML/RAR␣ with rs4790348, rs4790353 and rs7209748 in M2 and M3 AML. As shown in Supplymentary Table 2, there was no significant association between AML1/ETO or PML/RAR␣ and these 3 SNPs in M2 and M3 AML.
4. Discussion In the present study, we genotyped 5 potentially functional SNPs of LIS1 in 660 AML patients and 1034 cancer-free controls, and indicated that the variant alleles of rs4790348, rs4790353, and rs7209748 could significantly increase the AML risk in Chinese population. The findings suggested that genetic variants of LIS1 may contribute to AML susceptibility. LIS1 gene encoding the LIS1 protein is located in the region of chromosome 17p13.3. The LIS1 protein colocalizes with cytoplasmic dynein and dynactin, predominantly in prometaphase kinetochores and at the cell cortex of dividing cells [12,13]. LIS1/dynactin plays the function in metaphase spindle assembly and mitotic checkpoint control [12,14,15]. Besides, LIS1 has
S. Cao et al. / Leukemia Research 54 (2017) 7–11
9
Table 1 Demographic and clinical characteristics of study subjects. Characteristics
AML(n = 660)
Controls(n = 1034)
P
N
(%)
N
(%)
Gender Male Female
348 312
52.7 47.3
586 448
56.7 43.3
Age at diagnosis, year mean ± sd ≤45 >45
44.4 ± 17.8 356 304
53.9 46.1
45.4 ± 9.6 510 524
49.3 50.7
Lineage Myeloid Myeloid and Lymphoid
474 186
71.8 28.2
Classification of diagnosis M0 M1 M2 M3 M4 M5 M6 Unknown
5 119 221 119 62 103 4 27
0.8 18.0 33.5 18.0 9.4 15.6 0.6 4.1
109 336 151 64
16.5 50.9 22.9 9.7
96 72 60 377 55
14.5 10.9 9.1 57.1 8.3
Cytogenetic risk categoryb Favorable Intermediate Unfavorable Unknown Molecular subtype Common fusion gene transcripts PML/RAR␣ AML1/ETO Others No common fusion gene ranscripts Unknown a b
0.111
0.178a
P value of independent-sample T test for age. Grouped by Southwest Oncology Group (SWOG) criteria.
Table 2 Basic information of the 5 SNPs of LIS1. SNP
Location
Allele
MAFa
MAFb
HWEc
Call rate
Powerd
rs1266474 rs4790348 rs4790353 rs7209748 rs8081803
intron intron intron intron intron
A>G G>A T>G G>C T>C
0.130 0.321 0.079 0.131 0.053
0.116 0.264 0.077 0.125 0.049
0.351 0.883 0.355 0.231 0.755
0.988 0.989 0.991 0.988 0.995
0.806 0.963 0.663 0.828 0.498
a b c d
Minor allele frequency among case subjects. Minor allele frequency among control subjects. P value of hardy-Weinberg equilibrium test among control subjects. Power was calculated using Power and Sample Size Calculation software, as the type I error probability was set as 0.05, and OR was set as 1.5.
been involved in the development of some malignancies, including HCC, neuroblastoma, glioma, and cholangiocarcinoma [7–10]. Zimdahl et al. demonstrated that LIS1 deficiency impaired primitive (fetal) hematopoietic stem cells (HSCs) expansion and caused rapid depletion of definitive (adult) HSCs [11,16]. They also found that transformed hematopoietic cells in AML rely on LIS1-mediated regulation of cell division mode during a critical phase of leukemogenesis, and knockdown of LIS1 in primary human-derived CD34+ cells led to significantly impaired leukemic growth [11,16]. All three SNPs (rs4790348, rs4790353, and rs7209748) located in the intron region of LIS1, and none of them had been reported so far. To investigate the potentially functional significance of these SNPs, we used an online prediction tool, RegulomeDB (http:// regulomedb.org/), which implemented the data of the Encyclopedia of DNA Elements (ENCODE) [17]. All three SNPs were predicted to influence the chromatin structure and histone modifications. Additionally, rs4790348 might influence the binding of several
transcription factors, such as POLR3A and Zscan4, and is a trans eQTL linked to the expression of RTF1 [18]. Meanwhile, rs4790353 and rs7209748 were also located in the binding sites of some transcriptional factors including Brachyury and Irx3. All these predictions indicated that these three SNPs may had potential function to affect transcriptional regulatory of LIS1 in the development of AML. Several limitations need to be addressed in our study. Firstly, the present study was based on candidate gene approach, which could only detect limited genes and SNPs on AML. Secondly, the sample size might be relatively small, as two SNPs (rs4790353 and rs8081803) didn’t achieve a statistical power of greater than 80% when we assumed the OR was 1.5. Thirdly, we lacked environmental information like ionizing radiation and benzene, so that gene-environment interaction cannot be evaluated. Therefore, studies with larger sample size and more environmental informa-
10
S. Cao et al. / Leukemia Research 54 (2017) 7–11
Table 3 Association between selected SNPs and AML risk in different models. AMLa (n = 660)
AMLb (n = 474)
Control (n = 1034)
Adjusted OR (95%CI)a
Pa
Adjusted OR(95%CI)b
Pb
rs1266474 A > G AA AG GG Dominant model Recessive model Additive model
491 131 18 – – –
350 94 14
811 206 17 – – –
– 1.04(0.81, 1.33) 1.82(0.93, 3.57) 1.10(0.87, 1.39) 1.80(0.92, 3.53) 1.14(0.92, 1.39)
– 0.747 0.083 0.432 0.086 0.227
– 1.04(0.79, 1.37) 2.01(0.98, 4.14) 1.11(0.86, 1.45) 1.99(0.97, 4.10) 1.16(0.92, 1.45)
– 0.778 0.058 0.430 0.060 0.207
rs4790348 G > A GG GA AA Dominant model Recessive model Additive model
289 293 59 – – –
215 206 37
561 400 73 – – –
– 1.41(1.15, 1.74) 1.55(1.07, 2.26) 1.43(1.18, 1.75) 1.33(0.93, 1.90) 1.31(1.13, 1.53)
– 0.001 0.020 <0.001 0.122 0.001
– 1.33(1.05, 1.67) 1.31(0.85, 2.01) 1.32(1.06, 1.65) 1.15(0.76, 1.74) 1.22(1.03, 1.45)
– 0.017 0.217 0.013 0.504 0.024
rs4790353 T > G TT TG GG Dominant model Recessive model Additive model
555 78 12 – – –
396 56 9
879 151 4 – – –
– 0.82(0.61, 1.09) 4.83(1.55, 15.09) 0.92(0.69, 1.22) 4.97(1.59, 15.50) 1.03(0.80, 1.33)
– 0.173 0.007 0.548 0.006 0.829
– 0.82(0.59, 1.14) 5.21(1.59, 17.08) 0.93(0.68, 1.27) 5.35(1.63, 17.53) 1.05(0.79, 1.39)
– 0.239 0.007 0.650 0.006 0.747
rs7209748 G > C GG GC CC Dominant model Recessive model Additive model
488 134 17 – – –
349 95 13
787 235 12 – – –
– 0.92(0.72, 1.17) 2.30(1.09, 4.86) 0.99(0.78, 1.25) 2.34(1.11, 4.94) 1.06(0.86, 1.30)
– 0.506 0.030 0.921 0.026 0.588
– 0.92(0.70, 1.20) 2.44(1.10, 5.42) 0.99(0.76, 1.28) 2.49(1.13, 5.51) 1.07(0.85, 1.35)
– 0.524 0.028 0.942 0.025 0.571
rs8081803 T > C TT TC CC Dominant model Recessive model Additive model
586 63 3 – – –
420 43 3
935 97 2 – – –
– 1.05(0.75, 1.46) 2.35(0.39, 14.14) 1.07(0.77, 1.49) 2.34(0.39, 14.08) 1.10(0.80, 1.50)
– 0.791 0.351 0.674 0.354 0.569
– 1.00(0.69, 1.47) 3.13(0.52, 18.88) 1.05(0.73, 1.52) 3.13(0.52, 18.87) 1.09(0.77, 1.54)
– 0.987 0.213 0.801 0.213 0.627
a b
Those patients with myeloid and lymphoid lineage were included. Those patients with myeloid and lymphoid lineage were excluded.
Table 4 Stratification analysis of genotypes of the 3 SNPs by selected variables in AML patients and controls. Characteristics
rs4790348a
rs4790353b
rs7209748b
OR(95%CI)c
Pd
OR(95%CI)c
Pd
OR(95%CI)c
Pd
Age ≤45 >45
1.31(1.00, 1.72) 1.40(1.05, 1.88)
0.743
12.57(1.56, 100.98) 2.42(0.54, 10.90)
0.209
5.11(1.40, 18.72) 1.38(0.51, 3.75)
0.117
Gender Male Female
1.33(0.97, 1.81) 1.31(0.95, 1.80)
0.947
5.22(1.40, 19.42) 4.28(0.44, 41.50)
0.882
1.75(0.65, 4.72) 3.47(1.06, 11.39)
0.386
Lineage Myeloid Myeloid and Lymphoid
1.32(1.06, 1.65) 1.75(1.27, 2.41)
0.362
5.49(1.68, 18.00) 3.97(0.87, 18.04)
0.741
2.15(0.94, 4.93) 2.96(1.09, 8.05)
0.629
Cytogenetic risk categorye Favorable Intermediate Unfavorable
1.31(0.86, 1.98) 1.39(1.09, 1.79) 1.41(0.99, 2.00)
0.962
5.66(0.98, 32.75) 3.98(1.06, 14.93) 5.61(1.24, 25.45)
0.927
3.90(1.13, 13.47) 2.40(1.00, 5.75) 1.83(0.51, 6.58)
0.695
Molecular subtype PML/RAR␣ AML1/ETO Others No common fusion gene transcripts
1.24(0.80, 1.91) 1.37(0.84, 2.24) 1.88(1.09, 3.25) 1.44(1.13, 1.83)
0.704
6.66(1.23, 36.00) 5.71(0.61, 53.51) – 5.04(1.47, 17.37)
0.966
3.29(0.84, 12.79) 1.28(0.16, 10.37) 1.76(0.22, 13.95) 2.62(1.14, 5.99)
0.879
a b c d e
in dominant model. in recessive model. adjusted for age and gender. P value for heterogeneity test. Grouped by Southwest Oncology Group (SWOG) criteria.
S. Cao et al. / Leukemia Research 54 (2017) 7–11
tion were needed to further elucidate the impact of LIS1 SNPs on AML susceptibility. Conflicts of interest none. Acknowledgements This work was supported by the National Natural Science Foundation for Young Scholars of China [grant number 81402739]; Excellent Youth Foundation of Jiangsu Province [grant number BK20160046]; Priority Academic Program Development of Jiangsu Higher Education Institutions (Public Health and Preventive Medicine); and Top-notch Academic Programs Project of Jiangsu Higher Education Institutions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.leukres.2016. 12.007. References [1] M.R. O’Donnell, C.N. Abboud, J. Altman, F.R. Appelbaum, S.E. Coutre, L.E. Damon, J.M. Foran, S. Goorha, L.J. Maness, G. Marcucci, P. Maslak, M.M. Millenson, J.O. Moore, F. Ravandi, P.J. Shami, B.D. Smith, R.M. Stone, S.A. Strickland, M.S. Tallman, E.S. Wang, National comprehensive cancer network. Acute myeloid leukemia, J. Natl. Compr. Cancer Netw. 9 (March (3)) (2011) 280–317. [2] J.E. Rubnitz, B. Gibson, F.O. Smith, Acute myeloid leukemia, Pediatr. Clin. North Am. 55 (February (1)) (2008) 21–51, http://dx.doi.org/10.1016/j.pcl. 2007.11.003. [3] J.L. Shipley, J.N. Butera, Acute myelogenous leukemia, Exp. Hematol. 37 (June (6)) (2009) 649–658, http://dx.doi.org/10.1016/j.exphem.2009.04.002. [4] F. Ferrara, C.A. Schiffer, Acute myeloid leukaemia in adults, Lancet 381 (February (9865)) (2013) 484–495, http://dx.doi.org/10.1016/S01406736(12)61727-9. [5] M.R. O’Donnell, M.S. Tallman, C.N. Abboud, J.K. Altman, F.R. Appelbaum, D.A. Arber, E. Attar, U. Borate, S.E. Coutre, L.E. Damon, J. Lancet, L.J. Maness, G. Marcucci, M.G. Martin, M.M. Millenson, J.O. Moore, F. Ravandi, P.J. Shami, B.D. Smith, R.M. Stone, S.A. Strickland, E.S. Wang, K.M. Gregory, M. Naganuma, National comprehensive cancer network acute myeloid leukemia, version 2.2013, J. Natl. Compr. Cancer Netw. 11 (September (9)) (2013) 1047–1055.
11
[6] S. Fircanis, P. Merriam, N. Khan, J.J. Castillo, The relation between cigarette smoking and risk of acute myeloid leukemia: an updated meta-analysis of epidemiological studies, Am. J. Hematol. 89 (August (8)) (2014) E125–E132, http://dx.doi.org/10.1002/ajh.23744. [7] Z. Xing, X. Tang, Y. Gao, L. Da, H. Song, S. Wang, P. Tiollais, T. Li, M. Zhao, The human LIS1 is downregulated in hepatocellular carcinoma and plays a tumor suppressor function, Biochem. Biophys. Res. Commun. 409 (June (2)) (2011) 193–199, http://dx.doi.org/10.1016/j.bbrc.2011.04.117. [8] E. Messi, M.C. Florian, C. Caccia, M. Zanisi, R. Maggi, Retinoic acid reduces human neuroblastoma cell migration and invasiveness: effects on DCX LIS1, neurofilaments-68 and vimentin expression, BMC Cancer 8 (January (30)) (2008), http://dx.doi.org/10.1186/1471-2407-8-30. [9] S.O. Suzuki, R.J. McKenney, S.Y. Mawatari, M. Mizuguchi, A. Mikami, T. Iwaki, J.E. Goldman, P. Canoll, R.B. Vallee, Expression patterns of LIS1, dynein and their interaction partners dynactin, NudE, NudEL and NudC in human gliomas suggest roles in invasion and proliferation, Acta Neuropathol. 113 (May (5)) (2007) 591–599. [10] R. Yang, Y. Chen, C. Tang, H. Li, B. Wang, Q. Yan, J.1 Hu, S. Zou, MicroRNA-144 suppresses cholangiocarcinoma cell proliferation and invasion through targeting platelet activating factor acetylhydrolase isoform 1b, BMC Cancer 5 (December (14)) (2014) 917, http://dx.doi.org/10.1186/1471-2407-14-917. [11] B. Zimdahl, T. Ito, A. Blevins, J. Bajaj, T. Konuma, J. Weeks, C.S. Koechlein, H.Y. Kwon, O. Arami, D. Rizzieri, H.E. Broome, C. Chuah, V.G. Oehler, R. Sasik, G. Hardiman, T. Reya, Lis1 regulates asymmetric division in hematopoietic stem cells and in leukemia, Nat. Genet. 46 (March (3)) (2014) 245–252, http://dx. doi.org/10.1038/ng.2889. [12] K.H. Siller, C.Q. Doe, Lis1/dynactin regulates metaphase spindle orientation in Drosophila neuroblasts, Dev. Biol. 319 (July (1)) (2008) 1–9, http://dx.doi.org/ 10.1016/j.ydbio.2008.03.018. [13] J. Yingling, Y.H. Youn, D. Darling, K. Toyo-Oka, T. Pramparo, S. Hirotsune, A. Wynshaw-Boris, Neuroepithelial stem cell proliferation requires LIS1 for precise spindle orientation and symmetric division, Cell 132 (February (3)) (2008) 474–486, http://dx.doi.org/10.1016/j.cell.2008.01.026. [14] K.H. Siller, M. Serr, R. Steward, T.S. Hays, C.Q. Doe, Live imaging of Drosophila brain neuroblasts reveals a role for Lis1/dynactin in spindle assembly and mitotic checkpoint control, Mol. Biol. Cell 16 (November (11)) (2005) 5127–5140. [15] N.E. Faulkner, D.L. Dujardin, C.Y. Tai, K.T. Vaughan, C.B. O’Connell, Y. Wang, R.B. Vallee, A role for the lissencephaly gene LIS1 in mitosis and cytoplasmic dynein function, Nat. Cell Biol. 2 (November (11)) (2000) 784–791. [16] B. Will, U. Steidl, Stem cell fate regulation by dynein motor protein Lis1, Nat. Genet. 46 (March (3)) (2014) 217–218, http://dx.doi.org/10.1038/ng.2902. [17] A.P. Boyle, E.L. Hong, M. Hariharan, Y. Cheng, M.A. Schaub, M. Kasowski, K.J. Karczewski, J. Park, B.C. Hitz, S. Weng, J.M. Cherry, M. Snyder, Annotation of functional variation in personal genomes using RegulomeDB, Genome Res. Sep. 22 (9) (2012) 1790–1797, http://dx.doi.org/10.1101/gr.137323.112. [18] G. Badis, M.F. Berger, A.A. Philippakis, S. Talukder, A.R. Gehrke, S.A. Jaeger, E.T. Chan, G. Metzler, A. Vedenko, X. Chen, H. Kuznetsov, C.F. Wang, D. Coburn, D.E. Newburger, Q. Morris, T.R. Hughes, M.L. Bulyk, Diversity and complexity in DNA recognition by transcription factors, Science 324 (June (5935)) (2009) 1720–1723, http://dx.doi.org/10.1126/science.1162327.