Child apolipoprotein E gene variants and risk of cerebral palsy: estimation from case-parent triads.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 5 ) 1 e6

Official Journal of the European Paediatric Neurology Society

Original article

Child apolipoprotein E gene variants and risk of cerebral palsy: Estimation from caseeparent triads Magne Stoknes a,b,*, Espen Lien a,b, Guro L. Andersen a,c, Yongde Bao d, James A. Blackman e,f, Rolv Terje Lie g, Torstein Vik a,b a

Department of Laboratory Medicine, Children's and Women's Health, Faculty of Medicine, Norwegian University of Science and Technology, Trondheim, Norway b Department of Pediatrics, St. Olavs Hospital, Trondheim University Hospital, Norway c Vestfold Hospital Trust, The Cerebral Palsy Register of Norway, Tønsberg, Norway d DNA Science Core, University of Virginia School of Medicine, Charlottesville, VA, USA e Department of Pediatrics, University of Virginia, Charlottesville, VA, USA f Cerebral Palsy International Research Foundation, New York, USA g Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway

article info

abstract

Article history:

Objective: To use caseeparent triad data to investigate if cerebral palsy (CP) is associated

Received 5 October 2014

with variants of the APOE gene, the rs59007384 SNP of the TOMM40 gene or combined

Received in revised form

haplotypes of the two genes.

21 December 2014

Study design: DNA was analyzed in buccal swabs from 235 children with CP, their parents

Accepted 23 December 2014

and a sibling. The relative risks (RR) with 95% confidence intervals (CI) that the children would have a distribution of APOE genotypes, rs59007384 variants or combined haplotypes

Keywords:

deviating from Mendelian inheritance were estimated.

Cerebral palsy

Results: Children with CP were more likely than expected to carry the APOEε3 allele (RR 7.5;

Apolipoprotein E

CI: 0.99e53.7 for heterozygotes and 10.3; CI: 1.4e79.6 for homozygotes), and to have the

Genetics

haplotype of APOEε3 and rs59007384 G (RR 2.4; CI: 1e5.7 for heterozygotes, RR 3.7; CI: 1.4

Caseeparent triads

e9.5 for homozygotes) whereas the distribution was as expected for rs59007384 alone. In the subgroup analyses the findings were confined to children born preterm. Among siblings the distribution of these genes was as expected according to Mendelian inheritance. Conclusion: We speculate that children with APOEε2/APOEε4 alleles are more likely to die following cerebral injury in utero, resulting in a higher than expected proportion of children with CP carrying the APOEε3 allele. © 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

Abbreviations: AD, Alzheimer disease; APOE, apolipoprotein E (gene); apoE, apolipoprotein E (protein); BFMF, Bimanual Fine Motor Function; CP, cerebral palsy; CPRN, Cerebral Palsy Register of Norway; GMFCS, Gross Motor Function Classification System; HWE, HardyeWeinberg equilibrium; LD, linkage disequilibrium; MBRN, Medical Birth Registry of Norway; REC, Regional Ethics Committee; SNP, single-nucleotide polymorphism; TBI, traumatic brain injury. * Corresponding author. Department of Laboratory Medicine, Children's and Women's Health, Faculty of Medicine, Norwegian University of Science and Technology, PB 8905, 7491 Trondheim, Norway. Tel.: þ47 95001111; fax: þ47 72573801. E-mail address: [email protected] (M. Stoknes). http://dx.doi.org/10.1016/j.ejpn.2014.12.017 1090-3798/© 2014 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Stoknes M, et al., Child apolipoprotein E gene variants and risk of cerebral palsy: Estimation from caseeparent triads, European Journal of Paediatric Neurology (2015), http://dx.doi.org/10.1016/j.ejpn.2014.12.017

2

1.


Introduction

We have in an earlier study shown that cerebral palsy (CP) may be the result of a number of risk factors.1 However, few children shared the same risk factors suggesting individual susceptibility to ante- or peri-natal brain injuries. This individual susceptibility may partly be explained by variations of genes encoding factors involved in repair processes or responses to early brain insults.2 The apolipoprotein E gene (APOE), located on chromosome 19, is a gene where variants in single nucleotides may lead to different susceptibility to an early brain injury.2 Three alleles, ε2, ε3 and ε4, of this gene have been identified and described, resulting in corresponding differences in protein structure. The presence of the ε4 allele has in earlier studies been associated with Alzheimer disease (AD) and poor long-term recovery after traumatic brain injury (TBI).3e7 However, whereas APOE association studies have been inconclusive regarding the etiology of CP8,9 we have in an earlier study shown that presence of the APOEε4 allele was associated with more severe clinical manifestations of CP.2 In addition, we have reported that the presence of the T allele of the rs59007384 SNP (single-nucleotide polymorphism) in the TOMM40 gene influencing the production of apoE was associated with more severe manifestation of CP.10 The TOMM40 gene is located centromeric and adjacent to APOE. The results of our two previous studies therefore suggest that both the structure as well as the amount of apoE produced may play a significant role in the pathophysiology leading to various clinical manifestations of CP. Studies using caseeparent triad data are considered more robust than the classical caseecontrol design that so far has been used in studies of the etiology of CP.11 By genotyping both the children and their parents it is possible to assess whether children with a specific disease have a genotype distribution deviating from what would be expected assuming Mendelian transmission. To our knowledge, such caseeparent triads have not been applied in studies of the etiology leading to CP. The aim of the present study was therefore to use caseeparent triad data to investigate if CP is associated with specific APOE alleles and/or specific variants of the rs59007384 SNP in TOMM40. We hypothesized that the proportion of children with the APOEε4 allele, the T allele of the rs59007384 SNP or with the haplotype of APOEε4 and rs59007384 T would be higher in children with CP than expected according to Mendelian inheritance.

parents, as well as one sibling. The complete process from collecting these cells to the DNA analysis has been described in detail in an earlier study.2 The CPRN12 provided clinical data on the children diagnosed with CP. Written informed consent comprising detailed data registration in the CPRN and linkage with MBRN (Medical Birth Registry of Norway) was obtained from the parents. The CP register is an informed consent based register recording detailed information from all the pediatric habilitation centers in Norway. We did not collect any information on the mothers, fathers or siblings.

2.2.

In total, 703 children and families recorded in the CPRN were invited to participate, and 281 (40%) returned swabs. Of these, 26 were of poor quality and could not be used, resulting in a final population of 255 (36.3%) children with reliable APOE and TOMM40 genotyping. In this population of 255 children DNA was available for 235 triads, i.e. the child and its parents, and in addition, DNA was available in 208 siblings. Due to Mendelian inconsistencies 20 triads were removed, leaving 215 triads for further analyses. Among the siblings, 192 triads remained after 16 were removed due to Mendelian inconsistencies. We have, in a previous study, described that participants (those who returned swabs) and non-participants (those who did not return swabs) with CP did not differ in terms of CP subtype, GMFCS and BFMF levels.2

2.3.

Materials and methods

2.1.

Study design

This study is a caseeparent triad study including children diagnosed with CP in Norway born between 1996 and 2003 and recorded with detailed information in the Cerebral Palsy Register of Norway (CPRN). Genotyping was performed on DNA from buccal epithelial cells in children with CP and their

Study variables

Clinical information regarding CP subtypes and severity were abstracted from CPRN. Cerebral palsy was diagnosed and classified according to the recommendations by the Surveillance of Cerebral Palsy in Europe in 1999.13 Severity of CP was defined by the Bimanual Fine Motor Function (BFMF) and the Gross Motor Function Classification System (GMFCS). The GMFCS describes gross motor function within five levels. Level I corresponds to walking without limitations, whereas Level V corresponds to transportation by manual wheelchair.14,15 The BFMF describes fine motor function, also within five levels, where Level V corresponds to only ability to hold or worse.16 In the present study both GMFCS and BFMF were used as dichotomous variables by contrasting levels I and II with levels III to V. Another indicator of severity of CP is epilepsy, in this study defined as current use of antiepileptic drugs.

2.4.

2.

Participants

Ethics

The present study was approved by the Regional Ethics Committee (REC) for medical research in Mid-Norway (reference number 2010/398). Written informed consent was obtained from the parents for participants younger than 16 years, or from the participants themselves if they were older.

2.5.

Statistical analysis

We estimated relative risks (RR) with 95% confidence intervals (CI) of child APOE genotypes, rs59007384 variants and


3


combined haplotypes compared to population average risk using Haplin 5.3; a statistical software designed to study multi-marker associations among case triads.11 The software assesses if the distribution of haplotypes among the children with CP deviates from what would be predicted through Mendelian inheritance from the parents. The estimation of haplotypes and relative risks is done with a maximum likelihood approach, which includes triads with missing data. For the main analyses, the effect of each haplotype was measured relative to the remaining haplotypes. This reference model has advantages when estimating single- and doubledose risk, and is the default option in Haplin. However, in analyses within subgroups of children with CP (for example preterm and term born children and within groups with more or less clinical manifestations of CP), the most frequent allele APOEε3 was used as reference due to small numbers. The present study is a convenient study using data from a study of the relationship between variants of the APOE gene and severity of CP.2 In that study a priori power calculations were conducted, and indicated that a study population of 200 children would allow to demonstrate a twofold difference in prevalence of APOEε4 between the groups, with a two-sided significance level of 5% and 80% power.2

3.

Results

Out of the 215 children diagnosed with CP, 94 (44%) had spastic unilateral CP, 91 (42%) had spastic bilateral, 12 (6%) dyskinetic, 8 (4%) ataxic CP and 10 (5%) were not classified. Ninety-two (43%) were born preterm, 102 (47%) were born at term, and 21 (10%) had missing gestational age. Furthermore, 141 (66%) were assessed as GMFCS levels I and II and 71 (33%) as levels III to V, whereas 3 (1%) were not classified. For BFMF, 134 (62%) were assessed as levels I and II and 74 (34%) were as levels III to V, whereas 7 (3%) were not classified. A higher than expected proportion of children with CP had the APOEε3 allele, with an RR of 7.5 (CI: 0.99e53.7) for heterozygotes and 10.3 (CI: 1.4e79.6) for homozygotes (Table 1). The distribution of rs59007384 did not deviate from the expected (Table 2). When we combined the APOE gene with rs59007384, the haplotype APOEε3 and rs59007384 G occurred more often than expected among children with CP (RR ¼ 2.4; CI: 1e5.7 for heterozygotes and RR ¼ 3.7; CI: 1.4e9.5 for homozygotes) (Table 3). In addition, the haplotype APOEε2 and rs59007384 G was less frequent than expected among children with CP for heterozygotes (RR ¼ 0.46; CI: 0.2e1.03). Among the siblings we found no associations with APOE or rs59007384 alone (Online

Table 1 e Children with CP, associations with APOE. Genotype 2 2 3 3 4 4

Dose

N

RR

CI

P-value

Single Double Single Double Single Double

24 0 79 127 57 8

0.73 N/A 7.49 10.3 1.04 1.09

0.40e1.34 N/A 0.99e53.7 1.35e79.6 0.57e1.88 0.42e2.95

0.31 N/A 0.05 0.02 0.89 0.85

Table 2 e Children with CP, associations with rs59007384. Genotype G G T T

Dose

N

RR

CI

P-value

Single Double Single Double

60 113 60 10

1 (Reference) 1 (Reference) 0.85 0.89

0.58e1.26 0.41e1.9

0.41 0.77

Supplementary material: Tables 4 and 5), nor for the haplotype APOEε3 and rs59007384 G (Online Supplementary material: Table 6). In the analyses of subgroups of children with CP, using the APOEε3 allele as reference we found that within the group born preterm, the proportion who carried the APOEε2 allele was lower than expected according to Mendelian inheritance with an RR of 0.37 (CI: 0.15e0.91, p ¼ 0.03) for heterozygotes and 0.13 (CI: 0.02e0.82, p ¼ 0.03) for homozygotes. Also within the group of children with GMFCS levels I and II, and within the group of children with BFMF levels I and II, the proportions carrying the APOEε2 allele were lower than expected. For children with CP and GMFCS level I and II the RR for carrying the APOEε2 allele was 0.40 (CI: 0.21e0.80, p ¼ 0.01) for heterozygotes and 0.16 (CI: 0.04e0.64, p ¼ 0.01) for homozygotes, and for children with BFMF levels I and II carrying the APOEε2 allele, we found an RR of 0.41 (CI: 0.21e0.83, p ¼ 0.01) for heterozygotes and 0.17 (CI: 0.17e0.69, p ¼ 0.01) for homozygotes. Among children born at term, among boys and girls, among children who used, or did not use antiepileptic drugs, among children with GMFCS levels III to V and among children with BFMF levels III to V, the proportions of children carrying the APOEε2 and APOEε4 alleles were consistent with Mendelian inheritance, using the APOEε3 allele as reference (data not shown). Due to small numbers, we were not able to carry out the subgroup analyses for rs59007384. HardyeWeinberg equilibrium (HWE) in the total sample was calculated for APOE (p ¼ 0.016) and rs59007384 (p ¼ 0.34). In addition, HWE were calculated separately for the siblings (APOE, p ¼ 0.014 and rs59007384, p ¼ 0.274).

Table 3 e Children with CP, associations with APOE and rs59007384. Haplotype 2-G 2-G 2-T 2-T 3-G 3-G 3-T 3-T 4-G 4-G 4-T 4-T

Dose

N

RR

CI

P-value

Single Double Single Double Single Double Single Double Single Double Single Double

19 0 14 0 70 101 63 0 45 1 45 5

0.46 N/A 1.41 N/A 2.37 3.67 0.85 N/A 0.39 5.11 1.4 1.43

0.20e1.03 N/A 0.63e3.11 N/A 1.0e5.65 1.42e9.49 0.36e1.95 N/A 0.11e1.3 0.42e60.6 0.80e2.44 0.50e4.00

0.06 N/A 0.40 N/A 0.05 0.006 0.70 N/A 0.13 0.20 0.23 0.50


4

4.


Discussion

In contrast to our hypothesis, we found that a higher than expected proportion of children with CP had the APOEε3 allele, and the haplotype APOEε3 and rs59007384 G. Among siblings the distributions of these alleles were consistent with Mendelian transmission. The subgroup analyses suggested that the unexpected distribution mainly was found in children born preterm and among children with less severe gross and fine motor impairment. The wide confidence intervals for the association between APOEε3 and CP suggest that these findings should be interpreted with caution. However, when we studied the haplotype APOEε3 and rs59007384 G, the confidence intervals were narrower, strengthening our findings. The results of the subgroup analyses must be interpreted cautiously due to limited statistical power. Twenty children were excluded from the study due to Mendelian inconsistencies. This may be attributed to misattributed paternity, swapping of test tubes or errors in the swabs analysis. We cannot exclude the possibility that there were other children in the study with misclassified alleles. However, the quality of the DNA analyses was considered excellent, and data failing a consistency check at a confidence level of 95% were subjected to repeat analyses.2 Neither the study population nor the siblings were in HWE for the APOE genotype distribution, which may not be surprising in the group of children with CP. Although the siblings are healthy, they still share a certain amount of genes with their siblings with CP, and it is within realm of possibility that they also have deleterious effects of alleles, even if the results of the Haplin analyses were consistent with Mendelian inheritance. In addition, we find it most unlikely that the response rate of the study was affected by the genetic profile of parents and their children. Many comparisons may increase the risk for chance findings leading to type I errors. However, and according to Rosner (2000), care is needed when corrections for multiple comparisons are considered.17 Correction for multiple comparisons may be particularly relevant when comparing several groups and each pair of comparisons is equally interesting. However, in the present study we were primarily interested in comparing the genotypes/haplotypes of children with CP with their parents (the main comparison), and with their siblings to provide evidence that the unexpected distribution of alleles was specific to the CP group. Thus, the main findings with statistical significance were obtained in the principal comparison, and we consider it most appropriate to present the results without correction for multiple testing. We are not aware of previous studies that have used caseeparent triads to study APOE or rs59007384 as candidate genes in the etiology of CP. The unexpectedly high proportion of children with CP carrying the APOEε3 allele, and the haplotype APOEε3 and rs59007384 G, and the lower than expected proportion carrying the APOEε2 allele in children born preterm and children with less severe gross and fine motor function, were surprising in light of our own previous studies on severity of CP and etiologic studies by other authors. Although previous studies on etiology have been inconclusive,

those who have reported an association have found increased risk to be associated with the presence of APOEε2 and/or APOEε4 alleles.9,18 In our own previous studies we found that more severe clinical manifestations of CP were associated with the presence of APOEε4 or rs59007384 T alleles.2,10 Interestingly, Blackman et al. (2009) found a potential protective effect of the APOEε4 allele after fetal brain injury.19 They suggested that APOEε4 may have an advantageous effect early in life, but have a detrimental effect later on, consistent with the effect of antagonistic pleiotropy proposed by Williams in 1957.20 Our results may be consistent with the hypothesis proposed by Blackman. The discrepancy in findings between the present study and earlier studies may be explained by different study designs and study populations, but also suggest that our understanding of how genes affect brain development, in particular following brain injury, is limited, and that we still have a long way to go in understanding the modulating effect of genes on brain insults. The high relative risks of children with CP to carry an APOEε3 allele and/or the APOEε3-rs59007384 haplotype may be consistent with a causal relationship. The finding of higher relative risks in homozygotes than in heterozygotes may be consistent with a dose dependent increase in risk further supporting a causal relationship. However, it may seem biologically less plausible that the most common alleles should be considered risk factors for CP. We therefore speculate that fetuses carrying the APOEε2 and/or APOEε4 alleles exposed to an early injurious event to the brain may be at higher risk to die, whereas those carrying the APOEε3 allele are more likely to survive but with some degree of impairment. This could explain that the proportion of children with CP carrying the APOEε3 allele was higher than expected according to Mendelian inheritance. This would be consistent with the observations that APOEε2 and APOEε4 are related to poor repair processes in the brain after insults in adults.3e7 This interpretation is also supported by a finding in a recent study by Li et al. (2014) who found an association between recurrent pregnancy loss and the APOEε2 and APOEε4 alleles in the mother, whereas the APOEε3 polymorphism was interpreted by Li to be a protective factor.21 Molecular mechanisms explaining a possible increased risk for death after fetal brain injury when APOEε4 is present, may be impaired repair processes following an adverse event in utero, in particular through reduced ability to modulate the inflammatory response. Zhang et al., in 2010 pointed to immunoregulatory dysfunction in APOEε4 carriers, and Gale et al. (2014) also found that APOEε4 allele carriers differed in their immune responses compared to non-carriers.22,23 In addition, presence of at least one APOEε4 allele may negatively affect synapse formation, dendritic branching and axonal sprouting compared with the other APOE alleles.24e26 In the analyses within subgroups of children with CP, we found that among children born preterm, children with GMFCS levels I and II, and among children with BFMF levels I and II fewer children than expected carried the APOEε2 allele. Since we in these subgroup analyses used the APOEε3 allele as reference, these findings suggest that the main finding in the total study population of a higher than expected proportion of children with the APOEε3 allele is mainly confined to children born preterm. Since children born preterm are more likely to



have less severe motor impairments than children born at term, this could explain why similar results were obtained within the groups with less severe fine and gross motor impairments.27 The findings may therefore reflect that children born preterm carrying the APOEε2 allele or the APOEε4 allele are particularly vulnerable to an adverse brain injury, and therefore more likely to die, whereas those carrying the APOEε3 are more likely to survive but with CP. A further potential explanation of our findings could be that the APOEε3 allele is closely linked to another gene, responsible for the increased risk. It is important to acknowledge that the APOE alleles may have a number of different complex effects not known, contributing to our findings. In addition, there is most likely strong linkage disequilibrium (LD) between SNPs in rs59007384 and APOE alleles.28e30 Thus, the APOEε4 allele will in most instances be passed on as a unit together with the rs59007384 T allele due to the close proximity of the two genes and the strong LD between SNPs in this region, whereas the APOEε3 allele is linked to the rs59007384 G allele. However, in our study we found no association with CP and rs59007384 G, although we found an association with CP and the APOEε3 allele. This discrepancy requires further studies. In conclusion, the present study has shown a dosedependent association between CP and the APOEε3 allele, and between CP and the haplotype APOEε3 and rs59007384 G. Healthy siblings had no associations of APOE or rs59007384. The subgroup analyses revealed that the findings may be confined to children born preterm. We speculate that children with the APOEε2 and/or APOEε4 alleles are more likely to die following a brain injury in utero, resulting in a higher than expected proportion of children with CP carrying the APOEε3 allele. The apparent discrepancy between this and other studies may suggest that our understanding of how genes affect the outcome of early brain injuries is still very limited, and a multicentric approach should ideally be conducted in future research in order to document our findings. Further multicenter studies should collect large enough populations to study risk factors and genetic factors for specific subtypes of CP, as the causes of unilateral and bilateral spastic CP, as well as dyskinetic and ataxic CP are likely to be very different.1

Funding source Liaison Committee (ID: 4230/2010) between the Central Norway Regional Health Authority (RHA) and the Norwegian University of Science and Technology (NTNU).

Financial disclosure/conflict of interest The authors declare that they have no conflict of interest. The authors have no financial relationship relevant to this article to disclose.

5

Acknowledgments We would like to thank Ms. Jodi Darring at the University of Virginia, and Ms. Sandra Julsen Hollung at the Cerebral Palsy Register of Norway for their assistance and professor Frank Skorpen of the Norwegian University of Science and Technology (NTNU) for valuable advice in interpreting the genetic data. We would also like to thank professor Ha˚kon Gjessing at University of Bergen for valuable help with the statistical analyses. Last but not least, thank you to all the children and their families who contributed to the study.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejpn.2014.12.017.

references

1. Stoknes M, Andersen GL, Elkamil AI, Irgens LM, Skranes J, Salvesen KA, et al. The effects of multiple pre- and perinatal risk factors on the occurrence of cerebral palsy. A Norwegian register based study. Eur J Paediatr Neurol 2012;16(1):56e63. 2. Lien E, Andersen GL, Bao Y, Gordish-Dressman H, Skranes JS, Vik T, et al. Apolipoprotein E polymorphisms and severity of cerebral palsy: a cross-sectional study in 255 children in Norway. Dev Med Child Neurol 2013;55(4):372e7. 3. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 1993;261(5123):921e3. 4. Schiefermeier M, Kollegger H, Madl C, Schwarz C, Holzer M, Kofler J, et al. Apolipoprotein E polymorphism: survival and neurological outcome after cardiopulmonary resuscitation. Stroke 2000;31(9):2068e73. 5. Tanriverdi F, Taheri S, Ulutabanca H, Caglayan AO, Ozkul Y, Dundar M, et al. Apolipoprotein E3/E3 genotype decreases the risk of pituitary dysfunction after traumatic brain injury due to various causes: preliminary data. J Neurotrauma 2008;25(9):1071e7. 6. Zeng S, Jiang JX, Xu MH, Xu LS, Shen GJ, Zhang AQ, et al. Prognostic value of apolipoprotein E epsilon4 allele in patients with traumatic brain injury: a meta-analysis and meta-regression. Genet Test Mol Biomark 2014;18(3):202e10. 7. Zhou W, Xu D, Peng X, Zhang Q, Jia J, Crutcher KA. Metaanalysis of APOE4 allele and outcome after traumatic brain injury. J Neurotrauma 2008;25(4):279e90. 8. Cotten CM, Goldstein RF, McDonald SA, Goldberg RN, Salhab WA, Carlo WA, et al. Apolipoprotein E genotype and outcome in infants with hypoxic-ischemic encephalopathy. Pediatr Res 2013;75(3):424e30. 9. Kuroda MM, Weck ME, Sarwark JF, Hamidullah A, Wainwright MS. Association of apolipoprotein E genotype and cerebral palsy in children. Pediatrics 2007;119(2):306e13. 10. Lien E, Andersen GL, Bao Y, Gordish-Dressman H, Skranes J, Blackman JA, et al. Gene sequences regulating the production of apoE and cerebral palsy of variable severity. Eur J Paediatr Neurol 2014;18(5):591e6. 11. Gjessing HK, Lie RT. Caseeparent triads: estimating singleand double-dose effects of fetal and maternal disease gene haplotypes. Ann Hum Genet 2006;70(Pt 3):382e96.


6


12. Andersen GL, Irgens LM, Haagaas I, Skranes JS, Meberg AE, Vik T. Cerebral palsy in Norway: prevalence, subtypes and severity. Eur J Paediatr Neurol 2008;12(1):4e13. 13. Surveillance of cerebral palsy in Europe: a collaboration of cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Dev Med Child Neurol 2000;42(12):816e24. 14. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997;39(4):214e23. 15. Palisano RJ, Rosenbaum P, Bartlett D, Livingston MH. Content validity of the expanded and revised Gross Motor Function Classification System. Dev Med Child Neurol 2008;50(10):744e50. 16. Beckung E, Hagberg G. Neuroimpairments, activity limitations, and participation restrictions in children with cerebral palsy. Dev Med Child Neurol 2002;44(5):309e16. 17. Rosner B. Fundamentals of biostatistics. 5th ed., vol. xix. Pacific Grove, CA: Duxbury; 2000. p. 792. 18. O'Callaghan ME, MacLennan AH, Haan EA, Dekker G. The genomic basis of cerebral palsy: a HuGE systematic literature review. Hum Genet 2009;126(1):149e72. 19. Blackman JA, Gurka MJ, Bao Y, Dragulev BP, Chen WM, Romness MJ. Apolipoprotein E and functional motor severity in cerebral palsy. J Pediatr Rehabil Med 2009;2(1):67e74. 20. Williams GC. Pleiotropy, natural selection, and the evolution of sensescence. Evolution 1957;11:398e411. 21. Li J, Chen Y, Wu H, Li L. Apolipoprotein E (Apo E) gene polymorphisms and recurrent pregnancy loss: a metaanalysis. J Assist Reprod Genet 2014;31(2):139e48. 22. Gale SC, Gao L, Mikacenic C, Coyle SM, Rafaels N, Murray Dudenkov T, et al. APOepsilon4 is associated with enhanced in vivo innate immune responses in human subjects. J Allergy Clin Immunol 2014;134(1):127e34.

23. Zhang HL, Wu J, Zhu J. The immune-modulatory role of apolipoprotein E with emphasis on multiple sclerosis and experimental autoimmune encephalomyelitis. Clin Dev Immunol 2010;2010:186813. 24. Dumanis SB, Tesoriero JA, Babus LW, Nguyen MT, Trotter JH, Ladu MJ, et al. ApoE4 decreases spine density and dendritic complexity in cortical neurons in vivo. J Neurosci Off J Soc Neurosci 2009;29(48):15317e22. 25. Guo L, LaDu MJ, Van Eldik LJ. A dual role for apolipoprotein e in neuroinflammation: anti- and pro-inflammatory activity. J Mol Neurosci 2004;23(3):205e12. 26. Kim J, Basak JM, Holtzman DM. The role of apolipoprotein E in Alzheimer's disease. Neuron 2009;63(3):287e303. 27. Himpens E, Oostra A, Franki I, Calders P, Vanhaesebrouck P, Van den Broeck C. Influence of gestational age on nosologic CP characteristics in a high-risk population. Eur J Pediatr 2010;169(3):305e10. 28. Bekris LM, Millard SP, Galloway NM, Vuletic S, Albers JJ, Li G, et al. Multiple SNPs within and surrounding the apolipoprotein E gene influence cerebrospinal fluid apolipoprotein E protein levels. J Alzheimer's Dis 2008;13(3):255e66. 29. Mulder M, Ravid R, Swaab DF, de Kloet ER, Haasdijk ED, Julk J, et al. Reduced levels of cholesterol, phospholipids, and fatty acids in cerebrospinal fluid of Alzheimer disease patients are not related to apolipoprotein E4. Alzheimer Dis Assoc Disord 1998;12(3):198e203. 30. Yu CE, Seltman H, Peskind ER, Galloway N, Zhou PX, Rosenthal E, et al. Comprehensive analysis of APOE and selected proximate markers for late-onset Alzheimer's disease: patterns of linkage disequilibrium and disease/ marker association. Genomics 2007;89(6):655e65.


Apolipoprotein E gene variants on the risk of end stage renal disease.

Absence of apolipoprotein E protects mice from cerebral malaria.

Cerebral palsy: causes, pathways, and the role of genetic variants.

Impact of child and family characteristics on cerebral palsy treatment.

Apolipoprotein C3 gene variants and the risk of coronary heart disease: A meta-analysis.

Apolipoprotein E gene polymorphism and risk of type 2 diabetes and cardiovascular disease.

Maternal Infections During Pregnancy and Cerebral Palsy in the Child.

Apolipoprotein E gene polymorphism and the risk of intracerebral hemorrhage: a meta-analysis of epidemiologic studies.

Association between apolipoprotein E gene polymorphism and the risk of recurrent pregnancy loss: a meta-analysis.

Associations between apolipoprotein E gene polymorphisms and Alzheimer's disease risk in a large Chinese Han population.

Primary genetic investigation of a hyperlipidemia model: molecular characteristics and variants of the apolipoprotein E gene in Mongolian gerbil.

The right interventions for each child with cerebral palsy.

The Distribution of Apolipoprotein E Gene Polymorphism and Apolipoprotein E Levels among Coronary Artery Patients Compared to Controls.

The association between sex-related interleukin-6 gene polymorphisms and the risk for cerebral palsy.

Maternal Prepregnancy BMI and Risk of Cerebral Palsy in Offspring.

Circulating Apolipoprotein E Concentration and Cardiovascular Disease Risk: Meta-analysis of Results from Three Studies.

Cerebral palsy risk factors and their impact on psychopathology.

Apolipoprotein E (Apo E) gene polymorphisms and recurrent pregnancy loss: a meta-analysis.

Genes determining the severity of cerebral palsy: the role of single nucleotide polymorphisms on the amount and structure of apolipoprotein E.

Interaction between ALOX5AP and CYP3A5 gene variants significantly increases the risk for cerebral infarctions in Chinese.

Phenotypes of apolipoprotein B and apolipoprotein E after liver transplantation.

The relationship between obstructive sleep apnea syndrome and apolipoprotein E genetic variants.

Association between Apolipoprotein E Variants and Obesity-Related Traits in Mexican School Children.

Risk of cerebral palsy among the offspring of immigrants.