543150

research-article2014

NROXXX10.1177/1073858414543150The NeuroscientistDavidson and others

Review

Post-Traumatic Brain Injury: Genetic Susceptibility to Outcome

The Neuroscientist 2015, Vol. 21(4) 424­–441 © The Author(s) 2014 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1073858414543150 nro.sagepub.com

Jennilee Davidson1,2, Michael D. Cusimano1, and William G. Bendena2

Abstract It is estimated that 2% of the population from industrialized countries live with lifelong disabilities resulting from traumatic brain injury (TBI) and roughly one in four adults are unable to return to work 1 year after injury because of physical or mental disabilities. TBI is a significant public health issue that causes substantial physical and economical repercussions for the individual and society. Electronic databases (PubMed, Web of Science, Google Scholar) were searched with the keywords traumatic brain injury, TBI, genes and TBI, TBI outcome, head injury. Human studies on non-penetrating traumatic brain injuries reported in English were included. To provide health care workers with the basic information for clinical management we summarize and compare the data on post-TBI outcome with regard to the impact of genetic variation: apolipoprotein E (APOE), brain-derived neurotrophic factor (BDNF), calcium channel, voltage dependent P/Q type, catechol-O-methyltransferase (COMT), dopamine receptor D2 and ankyrin repeat and kinase domain containing 1 (DRD2 and ANKK1), interleukin-1 (IL-1), interleukin-6 (IL-6), kidney and brain expressed protein (KIBRA), neurofilament, heavy polypeptide (NEFH), endothelial nitric oxide synthase 3 (NOS3), poly (ADPribose) polymerase-1 (PARP-1), protein phosphatase 3, catalytic subunit, gamma isozyme (PPP3CC), the serotonin transporter (5-HTT) gene solute carrier family 6 member (SLC6A4) and tumor protein 53 (TP53). It is evident that contradicting results are attributable to the heterogeneity of studies, thus further researches are warranted to effectively assess a relation between genetic traits and clinical outcome following traumatic injuries. Keywords head injury, traumatic brain injury (TBI), genes and TBI, outcome, apolipoprotein E

Introduction Traumatic brain injury (TBI) is a major public health concern. At least 1.7 million TBIs occur only in the United States every year (Faul and others 2010). TBI is responsible for more trauma deaths than injury to any other region of the body, accounting, in North America, for more than one third of all trauma deaths (Centers for Disease Control and Prevention 2003; Thurman and others 1999). TBIs are commonly attributed to motor vehicular accidents as well as sports injuries, even if many data have been accumulated about the war-related injuries sustained near an explosion (Hampton 2011). In industrialized countries such as Canada and the United States, it is estimated that 2% of the population live with lifelong disabilities resulting from TBI and roughly one in four adults with TBI are unable to return to work 1 year after injury because of physical or mental disabilities (Centers for Disease Control and Prevention 2003). Post-TBI outcome, above all in the case of mild TBI, is variable such as changes in physical, cognitive, social, behavioral, and emotional states, and only partially explained by known prognostic factors. The severity of

injury, as assessed by the Glasgow Coma Scale (GCS), correlates well with the phenotypic outcome (Teasdale and Jennett 1974). However, because of differences among individuals, a consistent prognosis is not always possible. Experimental and observational studies have attributed genetic factors to the spectrum of clinical outcomes that may occur as a result of TBI. Depending on the extent of neuronal damage, cellular responses may lead to neuroprotection or cell death. The cellular and molecular pathways that regulate neuron function are under complex polygenic control. This genetic complexity complicates the identification of mechanisms by which neuronal damage occurs. This article focuses on the genetic basis 1

Department of Surgery, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada 2 Deparment of Biology, Queen’s University, Kingston, Ontario, Canada Corresponding Author: Jennilee Davidson, Division of Neurosurgery, Department of Surgery, St. Michael’s Hospital, 30 Bond Street, Toronto, Ontario, M5B 1W8, Canada. Email: [email protected]

Davidson and others of susceptibility to post-TBI phenotypes by reviewing the limited number of studies that have assessed a genetic polymorphism in relation to TBI outcome. Studies that examined the association between postTBI outcome and genetic background were assessed to elucidate any notable trends for future research. All relevant human studies that have reported an association or lack of association to a polymorphism with a global TBIoutcome measure were considered; penetrating traumatic brain injury studies were not included, in attempt to keep the examined injuries consistent. For this purpose, global outcome refers to the physical, cognitive, social, behavioral, and emotional states observed to change, without focusing on the underlying causal mechanisms. Electronic databases (PubMed, Web of Science, Google Scholar) were used to identify and include relevant studies. These 48 original articles and 2 abstract reports of congresses’ proceedings were retrieved with search terms: traumatic brain injury, TBI, genes and TBI, TBI outcome, head injury. Trends in post-TBI outcomes emerged amongst the studies when assessed collectively. Contributing factors that may have influenced the difference in outcomes as well as future directions are discussed.

Genetic Studies Genetic loci assessed for association with clinical outcome after TBI include polymorphisms in apolipoprotein E (APOE), brain-derived neurotrophic factor (BDNF), calcium channel, voltage dependent P/Q type, catecholO-methyltransferase (COMT), dopamine receptor D2 and ankyrin repeat and kinase domain containing 1 (DRD2 and ANKK1), interleukin-1 (IL-1), interleukin-6 (IL-6), kidney and brain expressed protein (KIBRA), neurofilament, heavy polypeptide (NEFH), endothelial nitric oxide synthase 3 (NOS3), poly (ADP-ribose) polymerase-1 (PARP-1), protein phosphatase 3, catalytic subunit, gamma isozyme (PPP3CC), the serotonin transporter (5-HTT) gene solute carrier family 6 member (SLC6A4), and tumor protein 53 (TP53). Studies that met this article’s inclusion criteria are listed in Table 1. These are the human studies that investigated the polymorphisms of these genes and their role in post-TBI outcome.

Apolipoprotein E One of the most documented allelic variations affecting outcome following neurotrauma in humans is the apolipoprotein E gene (APOE) and will therefore be covered in greater detail in this article compared to the other genetic variations listed. APOE encodes a glycolipoprotein responsible for the transportation of lipids to regenerating neurons, promoting repair, and construction of new cell membranes, neurites, and synapses (Teasdale and others 1997). There are three APOE allelic variants (ε2, ε3, ε4), which encode three

425 isoforms of the protein (E2, E3, E4) (Teasdale and others 1997). APOE isoforms differ in amino acids at positions 112 and 158: E2 (cysteine/cysteine), E3 (cysteine/arginine), E4 (arginine/arginine). These isoforms vary amongst their structural stability and thus have different tendencies to assume domain interactions (Mahley and Huang 2012). It is the differences amongst these interactions that lead to multiple cellular pathways that may modify the response to injury; the ε4 allele is the most neurotoxic isoform and can induce neuropathology via proteolytic cleavage (Figure 1) (Mahley and Huang 2012). ε4 has been associated with reduced growth and branching of neurites (Mahley and Huang 2012), and has also been found present in individuals who experience poor recovery post-TBI. However, lack of correlation between outcome and presence of the e4 allele has also been reported (Alexander and others 2007; Ariza and others 2006; Brichtová and Kozák 2008; Chamelian and others 2004; Chiang and others 2003; Crawford and others 2002; Friedman and others 1999; Isoniemi and others 2006; Liberman and others 2002; Lichtman and others 2000; Miller and others 2010; Nathoo and others 2003; Noe and others 2010; Ost and others 2008; Ottenbacher and others 1996; Ponsford and others 2011; Pruthi and others 2010; Shadli and others 2011; Sorbi and others 1995; Teasdale and others 1997; Teasdale and others 2005). Human APOE exhibits genetic polymorphism with varying prevalence rates among different ethnic groups. The proportion of subjects with at least one or more APOE ε4 alleles in various populations is variable: 0% in tribal population of Koch, India (Singh and others 2006), 29.5% (Millar and others 2003) in Caucasian population; 40.9% in South African population (Nathoo and others 2003), 14.2% ε4 prevalence rate reported in a cohort from India (Pruthi and others 2010; Thelma and others 2001). Carrying the ε4 allele does not guarantee an accurately predicted outcome; many factors are involved including age, sex, race, history of previous TBI, severity of TBI, other existing injuries or diseases, and disorders of the central nervous system. In the studies that have identified a relationship between APOE genotype and a significant outcome, there appears to be a notable trend. Taken together, a trend appears whereby the APOE gene is related to rate of recovery, such that ε4 allele carriers with severe TBI (sTBI) have poorer outcomes following the trauma. Due to the variations in study inclusion and exclusion criteria, definition of outcome, heterogeneous populations, population characteristics, confounding diseases, and differences in treatments and analysis, this cannot be definitive. A meta-analysis from Zhou and others (2008) on 14 cohort studies found the ε4 allele not to be associated with initial TBI severity, but was associated with increased risk of poor long-term outcome at 6 months after injury. Carrying the ε4 allele after experiencing a TBI may contribute to cognitive abnormalities, behavioral

426 m, M m, M

m, M m, M m, M, s

m, M, s m, M, s

m, M, s

m, M, s sTBI

Severe enough for inpatient rehabilitation Postmortem M, s

m, M, s

m, M, s

ε4

ε4

ε3 or ε4

ε4

ε4

ε4

ε4

ε4

ε4, ε3/3

ε4

ε4

ε4

ε4

ε4

Teasdale and others 2005

Brichtová and Kozák 2008 Lo and others 2009

Blackman and others 2005

Quinn and others 2004 Crawford and others 2002

Hiekkanen and others 2009

Ponsford and others 2011

Nathoo and others 2003 Chiang and others 2003

Chamelian and others 2004 Pruthi and others 2010 Teasdale and others 1997

Zhou and others 2008 Shadli and others 2011 Liberman and others 2002

ε4

TBI (m, mild; M, moderate; s, severe) m, M, s

Gene/Allele

APOE ε4

Authors

648

33

110

106

71

65

70

983

100

110

93

98

90

80

19

2527

n

GOSE

HISC, GOSE

CVLT, verbal fluency measures

Diffuse brain swelling

WeeFIM

CPP, consciousness at discharge, GOS

GOS

GOS

GOS

GOS

GOS

Grooved pegboard, simple reaction time, choice reaction time, PASAT, number vigilance, dual attention, Stroop, word recall, word recognition, picture presentation, memory scanning GHQ, SCI for DSM-IV, GOS, RHIFQ, RPCSQ GOS, DVT, DSST, CFT, Token Test

AVLT, TMT-B, COWA, WCST-64

Meta-analysis

Evaluation

Table 1.  Evaluations Used to Assess Genetic Effect on Traumatic Brain Injury (TBI) Outcome.

Mean 1.9 years

1 year

≤6 months

On admission to rehabilitation and at discharge Postmortem

6 months

1 year

6 months

6 months

6 months minimum

6 months

6 months

6 months

3, 6 weeks

6 weeks, 6 months

6 months

Post-TBI testing

(continued)

No significant differences between diffuse brain swelling and possession of an ε4 allele. ε4 carriers had poorer memory as seen by the learning slope (P = 0.036), amount of learning (P = 0.033), level of short-delay recall (P = 0.019), level of long-delay recall (P = 0.022). No significant differences between ε4 carriers and non-carriers, but after adjusting for age, the presence of TAI lesions and duration of PTA were predictive of 1-year outcome (P = 0.001). Greater susceptibility in older females for worse GOSE scores, and ε4 may be associated with poorer long-term outcome (P < 0.001), but not injury severity.

ε4 carriers more likely to have an unfavorable outcome at 6 months post-TBI (P = 0.006, after adjustment for age, GCS, CT: P = 0.024), and proportion of patients with lower GCS (