BDNF: New Views of an Old Player in Traumatic Brain Injury

Cognitive Function

Given its well-established role in synaptic plasticity, the effects of BDNF and the Val66Met SNP on cognition have been studied extensively. It is widely hypothesized that learning and memory impairments in Met allele carriers are due in part to differences in hippocampal morphology and function. In an initial study, Egan and others (2003) found that while genotype did not influence semantic or working memory in healthy individuals, deficits in delayed recall on the Wechsler Memory Scale–Revised suggested that episodic memory was significantly impaired in Met carriers. However, Dempster and others (2005) cited no significant difference between BDNF genotypes in episodic memory on this scale. An early fMRI study revealed that during the encoding and retrieval portions of a place recognition task, healthy Met carriers had decreased memory-related hippocampal activity when compared with healthy Val/Val participants. Of note, ~25% total variation in declarative memory performance was attributed to the interaction between the Met allele and hippocampal activity during the encoding phase (Hariri and others 2003). In addition to alterations in activity, several imaging studies have supported the findings of reduced hippocampal volume in healthy Met carriers (Bueller and others 2006; Hajek and others 2012; Pezawas and others 2004). Schofield and others (2009) utilized a computerized battery of neuropsychological tests to assess cognitive performance in healthy adults, finding that on a verbal list-learning task used to evaluate memory, Met/Met individuals had significantly higher errors than Val/Met and Val/Val individuals. These results were accompanied by lower hippocampal gray matter volume in Met carriers. Miyajima and others (2008) noted that in a healthy, elderly population, the presence of a Met allele was associated with significantly reduced processing speeds, diminished delayed recall, and lower general intelligence on a small cognitive battery; yet, no significant changes in hippocampal volume occurred.

Deficits in cognitive function associated with the Val66Met SNP have been well documented in populations with neuropsychiatric disorders. In a study examining individuals with schizophrenia, those with the Val/Val genotype had superior scores for visual working memory on the N-Back test as compared with the Val/Met and Met/Met genotypes (Rybakowski and others 2006). The same study examined the role of the SNP in individuals with bipolar disorder and revealed that Met carriers performed significantly worse than Val/Val carriers on domains of the Wisconsin Card Sorting Task, assessing working memory and executive function. This difference in cognition is in line with the hypothesis that the Val/Val genotype is associated with greater activity of the BDNF system when compared with Met carrier genotypes, thus resulting in altered neuronal function (i.e., superior cognitive performance; Rybakowski and others 2006). Similarly, Lu and others (2012) found that among antipsychotic-naive patients with schizophrenia, those with the Met allele had more persisting errors on the Wisconsin Card Sorting Task than those without the allele. These behavioral impairments are supported by neuroimaging studies that revealed decreases in gray matter volume and decreased activation within the dorsolateral prefrontal cortex of Met carriers (Pezawas and others 2004; Schofield and others 2009). Met carriers have also been reported to have significantly smaller hippocampal gray matter volumes in patients with major depression (Frodl and others 2007).

In studies evaluating the effects of the Val66Met SNP on cognitive function following TBI, the results are mixed. In an mTBI population, participants with the Val/Val genotype performed significantly better in attention, memory, executive function, and overall cognition on the Neuropsychological Assessment Battery over a 6-month time frame as compared with Met carriers (Narayanan and others 2016). McAllister and others (2012) found that Met carriers had significantly slower information-processing speeds 1 month post-mTBI. However, in a study investigating cognition at a chronic postinjury time point following a penetrating TBI, Met carriers had significantly higher scores in general intelligence, verbal comprehension, perceptual organization, working memory, and processing speed when compared with Val/Val carriers (Barbey and others 2014). Krueger and others (2011) also reported that the Met allele was associated with recovery of executive function following focal penetrating TBI. Another study demonstrated that at least 1 year after mTBI, Met carriers performed better than Val/Val carriers in the memory and executive functioning domains of a cognitive battery (Merritt and others 2020). As such, the current clinical literature, although preliminary, suggests that the Met allele may impair cognitive functioning acutely but be neuroprotective at chronic stages after TBI.

Genetically modified rodents have enabled additional insights into the role of the BDNF genotype in cognitive function and health following TBI. For example, a mouse model of the BDNF Val66Met SNP was used to evaluate the effects of genotype on a fear-conditioning paradigm targeted at measuring hippocampus- and amygdala-dependent learning (Liu and others 2021). The authors reported significant context-dependent memory impairments in Met carrier mice as compared with wild type Val/Val mice. Met carrier mice also had significant decreases in hippocampal volume, up to nearly 14%, when compared with wild type mice (Liu and others 2021). More recently, Giarratana and others (2019) conducted a study with similar genetically modified mice but examined the effects of the SNP following rmTBI with a lateral fluid percussion model. The preliminary findings of this study suggest that transgenic rodent models of this SNP replicate similar observations characteristic of the genetic variant in humans, such as deficits in learning and memory following mTBI.

While most literature to date suggests an association between the Met allele and cognitive impairments, a few studies contradict this conclusion. In healthy populations, multiple studies have reported no relationship between the Val66Met polymorphism and cognition nor related brain morphology (Karnik and others 2010; Richter-Schmidinger and others 2011). Research examining short-term memory performance suggests that the Met allele is actually protective against cognitive decline and may decrease susceptibility to white matter hyperintensities in the elderly (Huang and others 2014). A number of confounding variables could be the cause of inconsistencies in current research. For example, results could be influenced by the chosen cognitive assessments, small sample size, and sample population demographics, including age and ethnicity. The differences in brain morphology across genotypes could be the result of a combination of changes resulting from altered BDNF levels within the region, including decreased dendritic complexity, as well as increased cell death or decreased neurogenesis during development or over the life span. More targeted research is needed to determine the cause of these discrepancies.

Neuropsychiatric Symptoms

BDNF is widely considered an effective biomarker of various neuropsychiatric disorders and may play an instrumental role in the presentation and pathology of depression (Buchmann and others 2013; Lang and Borgwardt 2013), anxiety (Moreira and others 2015), and schizophrenia (Niitsu and others 2014; Notaras and others 2015). Meta-analyses evaluating depression (Molendijk and others 2014), schizophrenia (Green and others 2011), and bipolar disorder (Fernandes and others 2014) in the context of BDNF have revealed that low plasma and serum BDNF levels are associated with an increased risk of developing these disorders. Considering that Val66Met has been associated with decreased circulating BDNF levels, one can hypothesize that this SNP may modulate mood-related symptoms and pathology associated with a number of neuropsychiatric disorders.

In a population of healthy individuals, Met carriers have reported significantly stronger feelings of anxiety, nervousness, and insecurity using a visual analog scale ranging from 0 to 10 during the cold pressor test, as well as generally higher levels of social anxiety on the Liebowitz Social Anxiety Scale, as compared with Val/Val carriers (Colzato and others 2011). In addition, healthy Met/Met individuals had significantly higher rates of “anticipated worry” on the Temperament and Character Inventory when compared with healthy Val/Met and Val/Val individuals (Montag and others 2010). In a Han Chinese population, Met carriers who were schizophrenic reported significantly higher severity scores for depression and anxiety symptoms on the Positive and Negative Syndrome Scale (Sun and others 2013). Montag and others (2008) found that in a sample of healthy females, Met carriers demonstrated stronger amygdala activation while viewing pleasant and unpleasant emotional stimuli when compared with Val/Val carriers. The Met allele appears to have a similar effect on emotional processing in adolescents diagnosed with anxiety and/or depressive disorders in that Met carriers have increased activation in the amygdala as well as in the anterior hippocampus in response to emotional cues (Lau and others 2010). Soliman and others (2010) reported that when compared with healthy Val/Val individuals, healthy Met carriers demonstrated significant impairment in extinguishing a conditioned fear response. Impaired extinction learning was also accompanied by decreased activity in the ventromedial prefrontal cortex and increased activity in the amygdala, alterations commonly seen in depression and anxiety disorders. While these studies indicated significant alterations in neural activation, many reported a lack of association between genotype and brain volume in these areas (Frodl and others 2007; Kambeitz and others 2012).

In the context of TBI, Met carriers had significantly higher anxiety and depression scores at 1 week after mTBI (Wang and others 2018). Gabrys and others (2019) found that Met carriers with a history of mTBI had higher levels of brooding rumination and depressive symptoms. In addition, in children, Met carriers demonstrated a consistent trend of poorer behavioral adjustment as compared with Val/Val carriers after TBI (Treble-Barna and others 2022). Contrary to these findings, Gagner and others (2021) reported that pediatric Val/Val carriers showed significantly more internalizing behaviors 6 months post-mTBI. At 18 months postinjury, all children with mTBI presented more internalizing symptoms; however, genotype did not have an influence on behavior in this study. Adolescent Met carriers also reported having a better quality of life 6 months post-mTBI (Tuerk and others 2020). These results suggest that the Val66Met polymorphism may have a protective effect on neuropsychiatric symptoms after early childhood mTBI but exacerbate symptoms in adulthood.

Mood-related behaviors have been reported in rodent models of the Val66Met SNP. Chen and others (2006) used transgenic mice and found that when compared with wild type mice, Met/Met mice exhibited increased anxiety-like behaviors in response to environmental stressors. These anxiety-like behaviors did not normalize in response to treatment with the antidepressant fluoxetine. Met/Met carriers within the study also showed impairments in neuronal secretion of BDNF. These results were replicated by Yu and others (2012), who determined that the Met allele was associated with elevated depressive-like behaviors on the forced swim task and sucrose preference task. By using multiple tasks to assess anxiety-like behavior in Val66Met mice, stressed Met carriers spent significantly less time exploring the open arms in the elevated plus maze and demonstrated an increased latency to drink in a stressful environment on a novelty-induced hypophagia task (Dincheva and others 2017). Similarly, Soliman and others (2010) reported that Met carrier mice have significant impairments in extinction learning when compared with Val/Val controls. More recently, Hill and others (2020) reported that in an adolescent mouse model, the Met/Met genotype interacts with stress, resulting in reduced cell density and dendrite spine density in the dorsal hippocampus.

While many experts maintain that the Val66Met polymorphism is associated with a variety of mood-related symptoms comorbid with TBI, the evidence is not conclusive. The effects of this SNP on symptomatology could vary among neurologic insults and disorders owing to differences in BDNF expression in various brain regions. Similar to the current literature on cognitive function, the inconsistencies in these results may also be due to differences in study methodology and populations and attributed to small sample sizes. Further research is necessary to determine the effect of the Val66Met SNP on mood-related symptoms in preclinical and clinical samples.

Motor Function

Though few published studies exist on the topic, current literature suggests that the Val66Met SNP may influence motor function by altering sensorimotor cortex activation. Much of this research exists within the field of stroke. As compared with Val/Val patients, Met carriers have been reported to have a greater degree of disability and loss of independence (Braun and others 2020), as well as overall worse functional outcome following stroke (Kim and others 2013; van der Vliet and others 2017). A recent study used fMRI to reveal that the presence of the Met allele was associated with decreased ipsilesional primary sensorimotor cortex activation in patients after a stroke (Kim and others 2016). McHughen and others (2010) found that on a driving-based motor task, healthy Val/Val individuals demonstrated a nonsignificant trend of higher levels of short-term motor learning and retention as compared with healthy Val/Met counterparts. In the same study, Val/Val individuals had significantly greater sensorimotor cortex activation when compared with Val/Met individuals in response to index finger training. Using a complex motor learning program (i.e., a 5-day basketball-shooting exercise), Deveci and others (2020) stated that the Met allele was associated with slightly lower learning performance as compared with the Val/Val genotype in healthy adults. The SNP was also associated with a sensory-motor integration pattern in which higher afferent facilitation was shown at the early stage of motor learning after the first day of basketball exercises but not maintained to the final day of basketball exercises. These changes may be the result of altered synaptic plasticity of the primary motor cortex.

Research into the effects of the Val66Met SNP on motor function is scarce in preclinical models. Chen and others (2006) noted that locomotor activity did not differ between Met carrier and wild type mice in an open field behavioral assay. Fritsch and others (2010) found that as compared with wild type littermates, Met/Met knock-in mice demonstrated significantly impaired motor skill acquisition on an accelerating rotarod task. In a chronic stroke recovery model, Met carrier mice had significantly greater motor impairment when compared with wild type mice on the rotarod task during the acute recovery phase. However, the Met allele accelerated motor function recovery during subacute and chronic stages, with function returning to sham-related levels nearly 6 months postinjury (Qin and others 2014). In the context of TBI, rmTBI has been shown to cause a significant deficit in vestibular motor function in injured Met carrier mice relative to injured Val/Val mice (Giarratana and others 2019). While this research is still informative, further studies need to be conducted within the context of TBI to develop a better understanding of how this SNP might affect motor function postinjury.

Nociception

Posttraumatic headache, which commonly presents as a migraine- or tension-like headache, is one of the most common consequences of TBI. Posttraumatic headache is also frequently reported in conjunction with somatic, psychiatric, and cognitive symptoms following TBI (Ashina and others 2019). BDNF is widely distributed in the CNS, including key regions involved in pain regulation, such as the hippocampus, amygdala, hypothalamus, and sensory and trigeminal ganglia (Marziniak and others 2008). BDNF stimulates dopaminergic and serotonergic neurons involved in the pathogenesis of migraine and in activity-dependent plasticity in nociceptive pathways, which may lead to pain progression/persistence (Marziniak and others 2008). It has been suggested that BDNF has a role in migraine pain, due to its interaction with calcitonin gene–related peptide, a key vasodilating neuropeptide implicated in migraine pathogenesis (Terrazzino and others 2017). Therefore, it may play a role in the development and severity of posttraumatic headache.

In humans, elevated BDNF levels in serum and cerebrospinal fluid have been observed in individuals with migraine and headache (Blandini and others 2006; Fischer and others 2012; Tanure and others 2010). Serum BDNF levels are also significantly elevated during migraine attacks (Fischer and others 2012). Similar increases in CSF BDNF levels have been observed in rats with mechanical allodynia (Feliciano and others 2014). Multiple meta-analyses have revealed an association between the Val66Met polymorphism and migraine in adults, with the presence of the Met allele increasing susceptibility (Cai and others 2017; Terrazzino and others 2017). Yet, in a recent study, Koute and others (2021) noted that the presence of the SNP was associated with a decreased risk of headache and migraine in a sample of healthy pediatric participants. In spinal cord injury research, BDNF expression has been associated with the development of neuropathic pain (Garraway and Huie 2016). Elevated BDNF levels promoted functional recovery in the acute postinjury phase (Hutchinson and others 2004). However, chronic overexpression of BDNF can lead to maladaptive plasticity and hypersensitivity to pain (Boyce and others 2012; Garraway and others 2014). Although the effect of the Val66Met SNP on nociceptive response has not been directly investigated in the context of TBI, based on the role that BDNF plays in nociception, it is hypothesized that the presence of the Met allele would increase an individual’s risk for developing posttraumatic headache after TBI.

Age

Age is one factor that may influence the effect of the Val66Met SNP on TBI outcomes. For example, in young adult populations, the presence of the Met allele is related to significant emotional disturbance and cognitive impairments following TBI (Gabrys and others 2019; Marziniak and others 2008; Wang and others 2018). On the contrary, in studies with elderly participants, the Met allele preserved executive functioning after injury (Barbey and others 2014; Krueger and others 2011). Consistent with these age-related findings, Failla and others (2016) reported that after sustaining a TBI, Met carriers <45 years of age had a significantly higher risk of mortality, while the opposite effect was observed in elderly Met carriers.

This interaction between genotype and age may be explained by changes in cell signaling. As previously mentioned, TrkB expression plays a crucial role in neuronal survival. In the normal aging process, TrkB expression is negatively correlated with age (Webster and others 2006). However, sortilin, which works in conjunction with the p75 receptor to initiate apoptosis and long-term depression, is upregulated in old age (Al-Shawi and others 2008). Based on these shifts in BDNF signaling, reduced secretion of BDNF associated with the Met allele may be deleterious throughout adulthood but become beneficial in the advanced stages of life.

Sex

Sex-dependent associations to the Val66Met SNP have been reported in neuropsychiatric and neurodegenerative research. For example, in a meta-analysis, Verhagen and others (2010) reported that the presence of the Met allele significantly increased the risk of major depressive disorder in males but not females. Conversely, the Met allele has been associated with an increased stress response in females but not males (Jiang and others 2017). The Met allele also significantly reduced cognitive functioning in normal aging (Laing and others 2012) and increased the risk of Alzheimer disease (Fukumoto and others 2010; Lin and others 2014) in females but not males. Although the interaction between sex and the Val66Met SNP in TBI research is very limited, the Met allele has been associated with increased depressive symptoms in females (Gabrys and others 2019) and increased anxiety in males (Wang and others 2018).

The interaction between sex and the Val66Met polymorphism can likely be attributed to sex hormones. Chan and Ye (2017) identified four mechanisms in which estrogens modulate the expression of BDNF. First, BDNF expression can be directly induced by estrogen activation of the estrogen receptor. Second, estrogen can modify the activity of BDNF gene promoters. Third, the estrogen receptor modulates the activity of CREB, a key transcription factor of BDNF. Fourth, estrogen inhibits BDNF expression via neuronal activity by reducing GABA production. It has been shown that circulating BDNF levels peak when estradiol and progesterone levels are at their highest during the menstrual cycle (Begliuomini and others 2007). Based on the differing actions of sex hormones between males and females, it is reasonable to deduce that the effect of the Val66Met polymorphism may differ between males and females, especially within the context of TBI.

Ethnicity

One of the major limitations in studying the Val66Met polymorphism in humans is the substantial diversity of allelic frequency that exists among populations globally. In fact, according to the Allele Frequency Aggregator—a groundbreaking project recently developed by the US National Center for Biotechnology Information that provides aggregated allele frequencies from its database of genotypes and phenotypes across 12 diverse populations—the Met allele is present in approximately 19.1% of the global population (Phan and others 2020). As illustrated in Figure 3, the presence of the BDNF Met allele considerably varies across populations. The Met allele is reported to be >10 times more prevalent in East Asian populations (41.1%) when compared with African populations (4.3%; Failla and others 2016). It is possible that current studies are not accounting for this variability, leading to insufficient sample sizes when examining the SNP across multiple ethnic groups. Several meta-analyses have shown ethnicity-dependent associations to the Val66Met SNP in neurodegenerative and neuropsychiatric disorder outcomes, such as Alzheimer disease (Lin and others 2014), Parkinson disease (Lee and Song 2014; Q Wang and others 2019), obsessive-compulsive disorder (S Wang and others 2019), schizophrenia (Kheirollahi and others 2016), suicidal behavior (González-Castro and others 2017), and antidepressant response (Yan and others 2014; Zou and others 2010). However, it is important to note that meta-analyses examining these interactions have not been conducted in the context of TBI, likely due to the limited literature. Of the studies investigating the Val66Met polymorphism in TBI, the majority have included only Caucasian populations and therefore lack generalizability.

OPEN IN VIEWER

Figure 3.

Val66Met allele frequencies by population. The presence of the BDNF Met allele considerably varies across populations. The Met allele is reported to be >10 times more prevalent in East Asian populations (41.1%) as compared with African populations (4.3%). These differences in allelic frequencies may have a role in the availability of BDNF across populations. Met = methionine; Val = valine.

To date, there are only two known studies evaluating the interaction between the Val66Met polymorphism and TBI in non-Caucasian individuals. In Asian populations, the studies show that as compared with Val/Val carriers, Met carriers demonstrate impaired cognitive functioning (Narayanan and others 2016) and increased feelings of anxiety and depression (Wang and others 2018). These results mirror findings in Caucasian samples with TBI at similar postinjury time points (Gabrys and others 2019; Marziniak and others 2008). Based on the current literature, it appears that Val66Met does not significantly influence TBI outcomes with regard to ethnicity. However, given the overall lack of literature, especially involving non-Caucasian populations, it is still unclear exactly how ethnicity may alter TBI outcomes.

Although there is currently no evidence to suggest that ethnicity interacts with the Val66Met polymorphism after TBI, there are some possible explanations for variability in genetic associations among ethnicities. For example, differences in genetic mechanisms among populations may cause one population to become more vulnerable to the molecular and behavioral dysfunction associated with the SNP while another population may be protected. It is important to consider other environmental factors that may be specific to certain ethnic groups and investigate their potential interactions with the Val66Met polymorphism.