Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Association of Functional Polymorphisms from Brain-Derived Neurotrophic Factor and Serotonin-Related Genes with Depressive Symptoms after a Medical Stressor in Older Adults

  • Kerri S. Rawson ,

    rawsonk@psychiatry.wustl.edu

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • David Dixon,

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • Petra Nowotny,

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • William M. Ricci,

    Affiliation Orthopaedic Trauma Service, Washington University School of Medicine / Barnes-Jewish Hospital, Saint Louis, Missouri, United States of America

  • Ellen F. Binder,

    Affiliation Department of Internal Medicine, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • Thomas L. Rodebaugh,

    Affiliation Department of Psychology, Washington University in Saint Louis, Saint Louis, Missouri, United States of America

  • Leah Wendleton,

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • Peter Doré,

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

  • Eric J. Lenze

    Affiliation Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America

Correction

17 Apr 2015: The PLOS ONE Staff (2015) Correction: Association of Functional Polymorphisms from Brain-Derived Neurotrophic Factor and Serotonin-Related Genes with Depressive Symptoms after a Medical Stressor in Older Adults. PLOS ONE 10(4): e0126451. https://doi.org/10.1371/journal.pone.0126451 View correction

Abstract

Depressive symptoms are common in older adults after a disabling medical event and interfere with rehabilitation and recovery from the disability. This prospective study examined the role of genetic polymorphisms implicated in synaptic integrity and stress-associated depression as predictors of depressive symptoms after hip fracture. We recruited healthy comparisons from the community and participants with hip fracture after surgical fixation from Saint Louis, Missouri hospitals. We examined the valine (Val) to methionine (Met) polymorphism in brain-derived neurotrophic factor (BDNF), serotonin 1A receptor (5HT1a-rs6295) polymorphism, and the serotonin transporter-linked polymorphic region (5HTTLPR) interaction with the rs25531 A to G single nucleotide polymorphism (5HTTLPR-rs25531) as predictors of depressive symptoms. We also examined whether depressive symptoms mediate the influence of BDNF genotype on functional recovery. Among 429 participants with hip fracture, BDNF Met/Met carriers developed significantly more depressive symptoms than Val/Val carriers during a four-week period after the fracture (p=.012). BDNF genotype also predicted functional recovery over the ensuing year, mediated by its effects on depressive symptoms (CI: 0.07-3.37). Unlike prior studies of stressful life events, the S′ 5HTTLPR-rs25531 variant did not predict higher levels of depressive symptoms; instead, we report an exploratory finding of an epistatic effect between BDNF and 5HTTLPR-rs25531 whereby the compounded effects of two LA alleles and BDNF Met/Met genotype elevate risk of depressive symptoms after hip fracture (p=.006). No differences between 5HT1a genotypes were found. Our findings suggest plasticity-related genetic factors contribute to the neural mechanisms of mental and functional well-being after a disabling medical stressor.

Introduction

Disabling medical events, such as hip fracture, often trigger depressive symptoms. In fact, most of the total burden of depressive symptoms in our aging population occurs in the context of disabling medical events [13]. Hip fracture is a common [4] and severe life stressor [5,6]: it causes pain [7], fear [8], requires hospitalization and surgery [4,6], lengthy and intensive rehabilitation [9], and leads to increased risk of institutionalization and mortality [1012]. Depressive symptoms after hip fracture have a pernicious effect on the recovery process: they are associated with higher discharge rates to and increased length of stay in nursing homes [1315], poorer occupational and physical therapy participation, and poorer functional recovery [16,17]. Hip fracture patients with more depressive symptoms are less likely to return to pre-fracture levels in several areas of physical functioning (such as walking independently) as compared to patients with less depressive symptoms [16,18,19]. Currently, the ability to predict which persons are at-risk of depressive symptoms after a medical disability is surprisingly low. Identification of genetic risk factors has the potential to alert caregivers to monitor patients vulnerable for high depressive symptoms. Early-intervention and optimal management of depressive symptoms could then essentially mitigate their impact on functional recovery, thereby improving quality of life and functional independence in late-life.

The extent to which molecular processes contribute to the neurobiology of depressive symptoms and subsequent functional recovery after a medical stressor is not well understood. Past research indicates genes associated with synaptic integrity or those vulnerable to stressful events may influence depressive symptoms [2022]. In this study, we investigate genetic variants previously associated with depressive symptoms by examining their impact on both depressive symptoms and functional recovery in older adults who have experienced a medical stressor.

BDNF Val66Met polymorphism

Brain-derived neurotrophic factor (BDNF) is of the neurotrophin family [23] and has beneficial effects on health, particularly for promoting synaptic plasticity [20,24,25]. However, a valine (Val) to methionine (Met) substitution at nucleotide 66 (Val66Met; refSNP: rs6265) in the BDNF gene influences risk of depression in older adults [2629]. In vivo experiments demonstrate the Met allele results in a reduction of activity-dependent BDNF secretion and a loss of BDNF protein at the synapse due to abnormal trafficking patterns [30,31]. In addition, the Met substitution is associated with reduced hippocampal activity [32] and atrophy of the hippocampus in humans [22,33,34], a structure important for mood regulation. The reduction in hippocampal volume likely reflects a decrease in BDNF protein at the synapse and, consequently, a decrease in synapse formation [30,35,36].

Serotonin-related polymorphisms

In this study, we also assessed putative functional polymorphisms in serotonin (5HT), a hallmark neurotransmitter involved in depression. The gene SLC6A4 encodes for the serotonin transporter (5HTT) and this transporter’s efficiency is altered by both a genetic variation of the serotonin transporter gene-linked polymorphic region (5HTTLPR) and a functional A to G single-nucleotide polymorphism (SNP; refSNP: rs25531) [37,38]. 5HTTLPR is a 44-bp repeat insertion (long) or deletion (short) in the promoter region resulting in different levels of serotonin transporter expression [3941]. In depression research, the short (S) allele has been characterized as the susceptible allele due to its reduced ability to remove 5HT from the synaptic cleft compared to the long (L) allele [42,43]. Additional research has indicated the S and L alleles interact with the alleles of rs25531, such that the combination of the 5HTTLPR L allele and rs25531 G allele are equivalent to the S allele alone [38]. Although 5HTTLPR has been extensively studied in younger adults after stressful life events, few studies have examined it in late-life after a disabling medical event as the stressor. Older patients with myocardial infarction and at least one S allele had higher depressive scores than persons with two L alleles [40]. Similar results were found in older patients with a history of coronary disease [44], following a stroke [45], and in a small pilot study after hip fracture [46].

The HTR1A gene encodes the serotonin 1A receptor (5HT1a; refSNP: rs6295). 5HT1a hinders the release of serotonin and the G allele of C(-1019)G polymorphism is associated with increased depressive symptoms [4750].

Hypotheses

Here we examined functional polymorphisms from BDNF and serotonin-related genes as predictors of depressive symptoms after hip fracture. We hypothesized BDNF Met/Met genotype, 5HTTLPR-rs25531 S′ allele, and the G allele of the 5HT1a polymorphism would be associated with more depressive symptoms, and if so, they would also predict poorer functional recovery as a result of depressive symptoms. Lastly, we explored the interaction of BDNF and 5HTTLPR-rs25531 on depressive symptoms.

Methods and Materials

Setting and sample

We recruited participants with hip fracture from admissions to eight local hospitals in Saint Louis, Missouri from 2008–2012 and healthy comparisons from the community via self-referrals, presentations, advertisements, and the university’s research registry. Washington University's Human Research Protection Office, Washington University School of Medicine Institutional Review Board, and the review boards at the applicable participating research site (The Christian Hospital Research Committee for Christian Hospital, DePaul Institutional Review Board for DePaul Health Center, Missouri Baptist Medical Center Institutional Review Board for Missouri Baptist Medical Center, Saint Anthony’s Medical Center Institutional Review Board for Saint Anthony's Medical Center, SSM Saint Louis Institutional Review Board for SSM Saint Clare Health Center, Saint Louis University Institutional Review Board for Saint Louis University Medical Center Campus and Saint Mary’s Health Center, and Washington University School of Medicine Institutional Review Board for Barnes-Jewish Hospital) approved this study and approved the informed written consent document that was signed by all participants. Eligibility for participants with hip fracture included being age 60 or older, primary diagnosis of hip fracture to be surgically repaired, and absence of delirium or dementia. Participants with hip fracture were excluded if they were unable to provide written informed consent after complete description of the study, were unable to cooperate with the protocol, had a current diagnosis of major depression, had metastatic cancer, were significantly impaired by language, visual or hearing barriers, lived more than one hour away, were taking interferon medications, or had an inoperable fracture. Exclusion criteria for healthy comparisons were the same but also included persons who had a hip fracture in the past year, a major health issue, or had been hospitalized in the last six months.

The study was 52 weeks with evaluation time points at baseline (typically two days after surgery), week 1, 2, 4, 8, 12, 26, and 52. We conducted baseline, weeks 4 and 52 in person, and weeks 1, 2, 8, 12, and 26 by phone.

Measures

Trained raters measured depressive symptoms with the Montgomery-Asberg Depression Rating Scale (MADRS) [51] at every time point. At baseline, MADRS scores are retrospective: participants described themselves for the week prior to hip fracture or entry to study (for healthy comparisons). Higher scores indicate more depressive symptoms. Major and Minor Depressive Disorder were assessed with the Structured Clinical Interview for Diagnostic and Statistical Manual of Medical Disorders-IV (SCID) [52] at baseline and any subsequent time point if MADRS > = 10 or if the depressed mood or anhedonia item was > = 2. Participants diagnosed with Major Depressive Disorder at baseline (reflecting current depression prior to hip fracture) were excluded. Test-retest reliability for MADRS was ICC = .84 (CI = 0.51–0.95), SCID past major depression κ = .52, and minor depression κ = .35.

Functional recovery was measured with the Functional Recovery Score (FRS) from the Hospital for Joint Diseases Geriatric Hip Fracture Research Group [53]. The FRS assesses Basic Activities of Daily Living, Instrumental Activities of Daily Living, and Mobility. A total score (0–100) is calculated from the three areas with higher scores indicating a greater ability to perform activities. Test-retest reliability was ICC = .75 (CI = 0.24–0.92).

Cognitive performance was assessed with the Short-Blessed Test [54] at baseline, week 4 and 52. Scores >10 warranted further evaluation to ensure absence of dementia or delirium.

Participants rated how stressful they have found their hip fracture as a measure of perceived stress. Ratings included: (1) not at all stressful, (2) somewhat stressful, or (3) very stressful [55]. Test-retest reliability was κ = .75.

Genotyping

DNA was extracted from saliva and blood samples using standard procedures. For genotyping of the polymorphisms in the serotonin 1A receptor (rs6295), BDNF (rs6265), and 69 single nucleotide polymorphisms (SNPs) for the population stratification analysis, we used Sequenom technology, following manufacturer’s protocol. For serotonin transporter polymorphisms, we followed a protocol which genotypes the SLC6A4 promoter haplotype [38]. In short, it is a triplex polymerase chain reaction protocol followed by double restriction endonuclease digestion, determining the phase-certain 5HTTLPR VNTR and rs25531 genotypes which are combined as SA, SG, LA, and LG haplotypes. For analyses, a composite of 5HTTLPR-rs25531 was reflected by the number of LA alleles: LA/LA, LA/S′, and S′/S′ [38].

Statistical Analysis

T-tests and chi-square were used to examine demographic and genotype differences between participants with hip fracture and healthy comparisons.

The main outcome, MADRS scores, had a skewed distribution. Thus, we used generalized estimating equations (GEE) with a Poisson distribution in SAS 9.3 procedure GENMOD (SAS Institute, Cary, NC) to examine the polymorphisms as predictors of depression symptoms. GEE is advised for instances where data will be correlated, such as repeated measurements, and for non-parametric data. GEE accounts for missed visits and dropouts using the all available pairs method. Within participants with hip fracture, we assessed main effects of each polymorphism for significance in separate models, with the most common homozygote as reference. We next ran separate GEE models to evaluate if age, gender, antidepressant use, education, or Short-Blessed Test were significant covariates. We then investigated the interaction between BDNF and 5HTTLPR-rs25531 to replicate prior findings. Significance level of p<.05 was set for main effects; however, due to testing for three polymorphisms, a probability value less than p<.0167 was considered significant.

We performed Kaplan-Meier Estimates of Survival Function to examine if the polymorphisms predicted Major and Minor Depressive Disorder. Due to low numbers of participants with hip fracture developing Major Depressive Disorder, we combined the two diagnoses. Additionally, we expanded the timeframe to eight weeks post-fracture to account for extra time needed for a clinical interview to make a SCID diagnosis after a MADRS score warranted further evaluation.

Lastly, mediation analysis using MEDIATE [56,57] in IBM SPSS Statistics 21.0 (Armonk, NY) was performed to examine BDNF effects on functional recovery through depressive symptoms. Using Helmert coding to address the multicategorical BDNF variable [57], estimation of coefficients for the first contrast compares Val/Val and Val/Met (Val+) relative to Met/Met while the second contrast compares Val/Val relative to Val/Met carriers. MADRS scores, averaged from weeks 1 through 4 to reflect high points of depressive symptoms post-fracture, served as the mediator. FRS was calculated as percent of baseline at week 12, 26, and 52. Indirect effects of BDNF on FRS at week 12, 26, and 52 were conducted using stratified bootstrap sampling based on 10,000 bootstrap samples. Significance of indirect effects was determined with biased-corrected confidence intervals (90%) not including zero [58]. Mediation analysis uses listwise deletion for participants missing data on variables included in each model.

We conducted population structure analysis with Structure Software Version 2.3 [59]. Thirty-five SNPs were selected from a panel of 69 SNPs that were in Hardy-Weinberg equilibrium. Structure parameters included the admixture model assuming independent allele frequencies with 10,000 burn-in period and 50,000 of MCMC reps. Project was run K 1–5 for five iterations. Results were imported into Structure Harvester Web v0.6.93 to detect the ideal number of clusters [60].

Results

Our initial sample included 558 white (93%), 41 black (7%) and 3 Asian adults (<1%). Investigation of population stratification revealed two populations: those that self-identified as black and as white. Thus, to reduce variance explained by genetic diversity, data is presented for white participants only in the analyses [61,62]. At baseline (Table 1), participants with hip fracture were more likely to be female, to have less education, poorer functional scores, and worse Short-Blessed Test scores than healthy comparisons. We found no differences between participants with hip fracture and comparisons for age, MADRS depressive scores, or antidepressant use. Genotype frequencies from BDNF-rs6265, 5HTTLPR-rs25531, and 5HT1a-rs6295 were in Hardy-Weinberg equilibrium.

thumbnail
Table 1. Baseline demographic variables and summary of frequencies for BDNF Val66Met, 5HTTLPR-rs25531, and C(-1019)G 5HT1a (Caucasian only).

https://doi.org/10.1371/journal.pone.0120685.t001

Our time course data shows hip fracture to be depressogenic from week 1 to 4, followed by a decline in depressive symptoms (Fig. 1). Therefore, to examine the candidate polymorphisms, we focused on the four-week period post-fracture where an increase in depressive scores among participants with hip fracture was observed relative to comparisons.

thumbnail
Fig 1. Observed Montgomery-Asberg Depression Rating Scale (MADRS) scores over time for participants with hip fracture and healthy comparisons.

The number of participants with hip fracture and healthy comparisons are listed below the figure.

https://doi.org/10.1371/journal.pone.0120685.g001

Main genetic effects on depressive symptoms

Observed MADRS scores by BDNF and 5HTTLPR-rs25531 polymorphisms for participants with hip fracture are in Fig. 2. To examine differences in depression scores between genotypes, we entered BDNF, 5HTTLPR-rs25531, and C(-1019)G 5HT1a into separate GEE models that included time to account for repeated MADRS measures (Table 2). BDNF Met/Met carriers had significantly higher MADRS scores than Val/Val carriers (p = .012). For 5HTTLPR-rs25531, both S′/S′ (p = .032) and LA/S′ (p = .016) carriers had lower MADRS scores compared to LA/LA carriers. The 5HT1a model was not significant. Inclusion of all races/ethnicities in GEE models produced similar results (not shown).

thumbnail
Fig 2. Observed Montgomery-Asberg Depression Rating Scale (MADRS) scores for participants with hip fracture.

https://doi.org/10.1371/journal.pone.0120685.g002

thumbnail
Table 2. Parameter estimates (log) and empirical standard error estimates with time for four GEE models predicting MADRS depressive scores post-fracture.

https://doi.org/10.1371/journal.pone.0120685.t002

We next examined which covariates were predictive of MADRS scores. Gender, age, education, and Short-Blessed Test scores were not significant. Antidepressant use significantly predicted MADRS scores (b = 0.24, SE = .08, p = .005). The polymorphisms were therefore re-examined after including time and antidepressant use (S1 Table). Antidepressant use remained significant in all models, indicating participants on antidepressants had higher MADRS scores. Like the model without antidepressant use, the same trends were seen for BDNF, 5HTTLPR-rs25531, and 5HT1a.

Exploring an interaction between BDNF and 5HTTLPR-rs25531

Table 2 also summarizes the significant interaction between BDNF and 5HTTLPR-rs25531. As displayed in Fig. 3, Met/Met carriers had significantly higher MADRS scores than Val/Val carriers among participants with two LA alleles, whereas this distinction was not observed among S′ carriers. The interaction remained significant after including time and antidepressant use (p = .014; S1 Table).

thumbnail
Fig 3. Predicted values for the interaction effect between brain-derived neurotrophic factor (BDNF) and serotonin transporter gene-linked polymorphic region, 5HTTLPR-rs25531

. Within hip fracture participants with two 5HTTLPR-rs25531 LA alleles (LA/LA), contrast results for the GEE analysis indicate BDNF Met/Met carriers had significantly higher depressive symptoms than Val/Val carriers (χ2 = 4.37(1), p = .037) and Val/Met carriers (χ2 = 3.86(1), p = .05). Sample sizes are listed below each graph. Abbreviations: MADRS, Montgomery-Asberg Depression Rating Scale; GEE, Generalized estimating equations.

https://doi.org/10.1371/journal.pone.0120685.g003

Genetic results for participants with high-perceived stress

We further analyzed results by including only participants who perceived the hip fracture as stressful. Five percent of participants who rated the hip fracture as “not at all stressful” had lower MADRS scores than 68% of participants who rated it “very stressful” (p = .005) and 27% of participants who rated it “somewhat stressful” (p<.001). Analyses were re-examined by excluding those who reported the event as being “not at all stressful” (S2 Table). As in the aforementioned results, we found the same trends for the main effects of the at-risk polymorphisms using uncorrected p-values in this high-perceived stressed subsample. The interaction between BDNF and 5HTTLPR-rs25531 remained significant (p = .011).

Genetic risk for Major and Minor Depressive Disorder

We examined if the polymorphisms predicted incident depression diagnosis during the eight weeks post-fracture. The log rank tests from Kaplan-Meier survival estimates indicated no significant differences between genotypes for BDNF Val66Met (χ2 = 0.81(2), p = .67), 5HTTLPR-rs25531 (χ2 = 0.31(2), p = .86), or 5HT1a (χ2 = 2.36(2), p = .31).

Mediation test of BDNF on FRS through depressive symptoms

The indirect effects from the mediation analysis were significant and indicated BDNF genotype affects functional recovery as a result of depressive symptoms at weeks 12, 26, and 52 (S3 Table). Total, direct, and indirect effects for the Val+:Met/Met contrast at week 12 are provided in Fig. 4 for illustration. The total effect indicated Val+ carriers had significantly better FRS than Met/Met carriers at week 12. The direct effects indicated Val+ carriers had significantly lower MADRS scores relative to Met/Met carriers and participants with lower MADRS scores had higher FRS, regardless of BDNF genotype, at all 3 time points. Indirect effects indicated MADRS scores are a significant mediator for the Val+:Met/Met contrast. That is, Val+ carriers scored an average 1.728 units higher on FRS at week 12 relative to Met/Met carriers due to the effect of BDNF genotype on depressive symptoms. Similar results were seen using FRS as the outcome at week 26 and 52. Neither total, direct, nor indirect effects for the Val/Val:Val/Met contrast were significant. There was no evidence that BDNF genotype influenced FRS independent of its effect on depressive symptoms at week 12, 26, or 52 for either contrast.

thumbnail
Fig 4. Mediation model using Hayes’ (2013) multicategorical independent variable method

. Mediation test of the relationship between brain-derived neurotrophic factor (BDNF) Val66Met polymorphism and the Functional Recovery Score (FRS) at week 12 as a result of the Montgomery-Asberg Depression Rating Scale (MADRS) depression scores. Results are shown for the BDNF Val+:Met/Met contrast. The Val/Val:Val/Met contrast was not significant.

https://doi.org/10.1371/journal.pone.0120685.g004

aIndependent of MADRS scores, FRS was not influenced by the direct effect of BDNF Val+ relative to Met/Met (p = .09).

Discussion

Depressive symptoms are commonplace after a disabling medical event, can increase risk and length of stay in nursing homes, and impede functional recovery [13,16,17]. In this study, we sought to identify responsible genetic mechanisms underlying the neurobiological mental and functional consequences that arise after a medical stressor.

This longitudinal analysis provided several noteworthy findings. Results indicate hip fracture participants with two BDNF Met alleles have higher depressive symptoms than Val/Val carriers in the first weeks post-fracture. Analyses also indicated BDNF Met/Met carriers were more likely to have poorer functional recovery scores for a year after hip fracture as a result of BDNF genotype on depressive symptoms. We did not find the expected association between the 5HTTLPR-rs25531 short allele and depressive symptoms; however, we report an interaction between BDNF and 5HTTLPR-rs25531. Although the interaction needs to be further explored in a more suitably sized cohort, our current sample indicated BDNF Met/Met carriers have higher depressive symptoms than Val/Met and Val/Val carriers in the presence of two LA alleles.

Understanding the neurobiological consequences of genes arising after a medical stressor is not only important for identifying older adults at risk of higher depressive symptoms, but secondarily, as our data (and others’) suggests [63,64], also significant for their functional recovery. BDNF generally plays an important role in synaptic plasticity [20], however, the Met allele at the Val66Met polymorphism results in abnormal neurotrophin trafficking patterns in neuronal cells. More specifically, the BDNF Val protein is localized more in the axon terminal, whereas expression of the BDNF Met protein has most localization in the cell body [30,31]. In consequence, a loss of BDNF at the axonal terminal among Met/Met carriers may compromise formation of new synapses [30,31,33,35,36]. Considering previous studies have indicated that atrophy of the hippocampus [22,33] and reduced hippocampal activity [32] are associated with the Met allele, BDNF Met/Met carriers may be vulnerable to the development of depressive symptoms due to inadequate synaptic formation in the hippocampus, a structure important for mood regulation [26,34,35,65,66].

The serotonergic system has long been a focus of research on the genetic origins of depression. Studies, like Caspi’s [42], investigating 5HTTLPR and depression, have been particularly arguable in the context of gene-environment effects [67,68]. While several previous studies have observed the 5HTTLPR short allele is related to depression due to less efficient transcription of serotonin transporters [43], others have found no association of 5HTTLPR and depression [69]. The different conclusions may be due to prior observations having smaller sample sizes [46], examining a history of life events rather than an acute medical event, having younger subjects, or not accounting for the influence of rs25531 A>G on 5HTTLPR [70]. Our current observations suggest the phenomenon may be pathophysiologically different after a medical stressor, or as Taylor and colleagues [71] findings indicate, neurostructural differences are evident between early and late-onset depression. Among older adults classified as having depression after age 50, results showed those who were homozygous with the L/L genotype had smaller hippocampal volume than those with early-onset depression. Future efforts should include the concerted genetic mechanisms among serotonin-related polymorphisms as well as compare genetic differences between persons with progressive medical events (i.e. stroke, heart disease) to those with injury-related medical events as investigated here.

We also explored a gene-gene interaction due to existing reports linking neurotrophins with serotonin in depression literature [72,73] and research suggesting compounded effects of certain genes can heighten risk of psychiatric symptoms [74]. In this study, the interaction between BDNF and 5HTTLPR-rs25531 was specific to LA/LA carriers. Particularly, significant differences in depressive symptoms were found between Met/Met and Val+ carriers among LA/LA carriers only; the interaction was not significant among S′ carriers. Thus, the short allele appears to contribute protection against depressive symptoms in persons with the at-risk BDNF genotype. While an inclusive model explaining the neurotrophin-serotonin association is unknown, past research indicates a dynamic relationship between the two signaling systems. The BDNF protein contributes to the survival and plasticity of serotonergic neurons as well as plays a role in serotonin secretion via BDNF’s TrkB receptors. Additional studies have also demonstrated stress reduces BDNF and serotonin levels [72,73]. Conceivably, and like most bodily systems, a homeostatic equilibrium would be achieved between the two signaling systems under normal conditions regardless of one’s BDNF and 5HTTLPR-rs25531 alleles. However, it could be speculated that LA/LA carriers with normal functioning 5HT transcription, combined with the suboptimal BDNF Met substitution, are more vulnerable during a stressful situation. That is, a further reduction of BDNF due to stress could impact BDNF’s ability to contribute to the survival and functioning of serotoninergic neurons, whereas the less functional serotonin system among S′ carriers has learned to adapt with neurotrophins over time and may be more resistant to stressors. Further studies, specifically ones that reveal how the neurotrophin-serotonin systems adapt to maintain homeostasis over time and how they interact as a result of a stressful experience at the biological level, are needed.

Similar results have also been found for neuroticism and cortisol levels: BDNF Met carriers homozygous for 5HTTLPR L allele have increased neuroticism scores [75], and, in children, BDNF Met carriers with at least one 5HTTLPR L allele have higher cortisol levels [76]. In contrast, studies investigating stressful life events found higher depression among Korean older adults [28] and maltreated children [77] with BDNF Met and 5HTTLPR S′ allele. As opposed to limiting investigations to a single gene, further studies examining net effects of BDNF and serotonin-related polymorphisms may better explain stress-associated depression risk factors [62]. This is particularly true in consideration of the complexity underlying depression pathophysiology.

In contrast to the high levels of depressive symptoms in this sample, a low number of participants with hip fracture received a clinical diagnosis of depression. Likely, due to these low rates, we did not find that our candidate polymorphisms were related to a clinical diagnosis. However, depressive symptoms, whether or not they meet criteria for a clinical diagnosis of depression, are still dangerous for patient’s health and are a barrier to successful recovery.

This was a study specifically designed to test for plasticity-related genetic effects on depressive symptoms after a disabling medical stressor. Depression is a heterogeneous disease, but in this project we utilized hip fracture as a sort of natural experiment to test candidate genetic factors explicitly. This study included immediate recruitment of participants after hip fracture and careful characterization of depressive symptoms.

Limitations included a small number of BDNF Met/Met carriers. The interaction finding is preliminary and requires replication. A replication study could be accomplished in several ways including studying BDNF levels, downstream functioning, or a sample with a higher number of BDNF Met/Met carriers. Prior studies suggest the Met allele is more common in certain Asian populations [78] which may be a more ideal population to further examine Met/Met effects.

In conclusion, our data indicates plasticity-related polymorphisms are associated with higher depressive symptoms after a medical stressor in late-life and, subsequently, can negatively affect recovery. This knowledge is crucial for our aging population’s need to maintain independence. In addition, our data suggests a neurotrophin-serotonin link. As Kendler [79] argues, it is unlikely we will uncover an exclusive genetic pattern revealing a heterogeneous illness such as depression; but it is possible to identify subsets of genetic interactions that delineate particular molecular pathways susceptible under certain genetic and environmental conditions. The interaction identified here highlights the need to further study and understand the neurobiological outcomes resulting from neurotrophin and neurotransmitter genetic interactions in stress-associated depression.

Supporting Information

S1 Table. Parameter estimates (log) and empirical standard error estimates with time and antidepressant use entered as covariates for four GEE models predicting MADRS depressive scores post-fracture.

Abbreviations: 5HT1a, Serotonin 1A receptor; 5-HTTLPR, serotonin transporter gene-linked polymorphic region; BDNF, brain-derived neurotrophic factor; BL, baseline; GEE, generalized estimating equation; MADRS, Montgomery-Asberg Depression Rating Scale.

https://doi.org/10.1371/journal.pone.0120685.s001

(DOCX)

S2 Table. Parameter estimates (log) and empirical standard error estimates in the high-perceived stress subsample post-fracture (excluding five percent of participants who rated the hip fracture as “not at all stressful”) with time and antidepressant use entered as covariates.

Four separate GEE models predicting MADRS depressive scores post-fracture. Abbreviations: 5HT1a, Serotonin 1A receptor; 5HTTLPR, serotonin transporter gene-linked polymorphic region; BDNF, brain-derived neurotrophic factor; BL, baseline; GEE, generalized estimating equation; MADRS, Montgomery-Asberg Depression Rating Scale.

https://doi.org/10.1371/journal.pone.0120685.s002

(DOCX)

S3 Table. Mediation analysis examining the effect of the BDNF Val66Met polymorphism on Functional Recovery Score (FRS) as a result of MADRS Depressive Scores.

Abbreviations: BDNF, brain-derived neurotrophic factor; MADRS, Montgomery-Asberg Depression Rating Scale; FRS, Functional Recovery Score.

https://doi.org/10.1371/journal.pone.0120685.s003

(DOCX)

S1 File. Supporting Information eReferences.

https://doi.org/10.1371/journal.pone.0120685.s004

(DOCX)

Acknowledgments

We are grateful for the helpful comments and review of manuscript from Steven Mennerick, Ph.D., Department of Psychiatry, Washington University School of Medicine, Saint Louis, Missouri, United States of America.

Author Contributions

Conceived and designed the experiments: EJL. Performed the experiments: LW PN. Analyzed the data: KSR DD TLR PD. Contributed reagents/materials/analysis tools: PN. Wrote the paper: KSR DD WMR PN EFB TLR LW PD EJL. Interpretation of data: WMR EFB. Database Management: PD.

References

  1. 1. Bisschop MI, Kriegsman DMW, Deeg DJH, Beekman ATF, van Tilburg W. The longitudinal relation between chronic diseases and depression in older persons in the community: the Longitudinal Aging Study Amsterdam. J Clin Epidemiol. 2004;57: 187–194. pmid:15125629
  2. 2. Kennedy GJ, Kelman HR, Thomas C. The emergence of depressive symptoms in late life: the importance of declining health and increasing disability. J Community Health. 1990;15: 93–104. pmid:2141337
  3. 3. Prince MJ, Harwood RH, Blizard RA, Thomas A, Mann AH. Impairment, disability and handicap as risk factors for depression in old age: the Gospel Oak project V. Psychol Med. 1997;27: 311–321. pmid:9089824
  4. 4. Marks R, Allegrante JP, MacKenzie C, Lane JM. Hip fractures among the elderly: causes, consequences and control. Ageing Res Rev. 2003;2: 57–93. pmid:12437996
  5. 5. Kain H. Care of the older adult following hip fracture. Holist Nurs Pract. 2000;14: 24–39. pmid:12119649
  6. 6. Zuckerman JD. Hip fracture. N Engl J Med. 1996;334: 1519–1525. pmid:8618608
  7. 7. Williams CS, Tinetti ME, Kasl SV, Peduzzi PN. The role of pain in the recovery of instrumental and social functioning after hip fracture. J Aging Health. 2006;18: 743–762. pmid:16980638
  8. 8. Parker M, Johansen A. Hip fracture. BMJ. 2006;333: 27–30. pmid:16809710
  9. 9. Holmes J, House A. Psychiatric illness predicts poor outcome after surgery for hip fracture: a prospective cohort study. Psychol Med. 2000;30: 921–929. pmid:11037100
  10. 10. Cree M, Soskolne CL, Belseck E, Hornig J, McElhaney JE, Brant R, et al. Mortality and institutionalization following hip fracture. J Am Geriatr Soc. 2000;48: 283–288. pmid:10733054
  11. 11. Hannan EL, Magaziner J, Wang JJ, Eastwood EA, Silberzweig SB, Gilbert M, et al. Mortality and locomotion 6 months after hospitalization for hip fracture: risk factors and risk-adjusted hospital outcomes. JAMA. 2001;285: 2736–2742. pmid:11386929
  12. 12. Magaziner J, Lydick E, Hawkes W, Fox KM, Zimmerman SI, Epstein RS, et al. Excess mortality attributable to hip fracture in white women aged 70 years and older. Am J Public Health. 1997;87: 1630–1636. pmid:9357344
  13. 13. Givens JL, Sanft TB, Marcantonio ER. Functional recovery after hip fracture: the combined effects of depressive symptoms, cognitive impairment, and delirium. J Am Geriatr Soc. 2008;56: 1075–1079. pmid:18422945
  14. 14. Hershkovitz A, Kalandariov Z, Hermush V, Weiss R, Brill S. Factors affecting short-term rehabilitation outcomes of disabled elderly patients with proximal hip fracture. Arch Phys Med Rehabil. 2007;88: 916–921. pmid:17601474
  15. 15. Gialanella B, Prometti P, Monguzzi V, Ferlucci C. Neuropsychiatric symptoms and rehabilitation outcomes in patients with hip fracture. Am J Phys Med Rehabil. 2014;
  16. 16. Mossey JM, Mutran E, Knott K, Craik R. Determinants of recovery 12 months after hip fracture: the importance of psychosocial factors. Am J Public Health. 1989;79: 279–286. pmid:2916712
  17. 17. Lenze EJ, Munin MC, Dew MA, Rogers JC, Seligman K, Mulsant BH, et al. Adverse effects of depression and cognitive impairment on rehabilitation participation and recovery from hip fracture. Int J Geriatr Psychiatry. 2004;19: 472–478. pmid:15156549
  18. 18. Magaziner J, Simonsick EM, Kashner TM, Hebel JR, Kenzora JE. Predictors of functional recovery one year following hospital discharge for hip fracture: a prospective study. J Gerontol. 1990;45: M101–M107. pmid:2335719
  19. 19. Marottoli R, Berkman L, Cooney L. Decline in physical function following hip fracture. J Am Geriatr Soc. 1992;40: 861–866. pmid:1512379
  20. 20. Lipsky RH, Marini AM. Brain-derived neurotrophic factor in neuronal survival and behavior-related plasticity. Ann N Y Acad Sci. 2007;1122: 130–143. pmid:18077569
  21. 21. Krishnan V, Nestler EJ. The molecular neurobiology of depression. Nature. 2008;455: 894–902. pmid:18923511
  22. 22. Campbell S, MacQueen G. The role of the hippocampus in the pathophysiology of major depression. J Psychiatry Neurosci. 2004;29: 417–426. pmid:15644983
  23. 23. Leibrock J, Lottspeich F, Hohn A, Hofer M, Hengerer B, Masiakowski P, et al. Molecular cloning and expression of brain-derived neurotrophic factor. Nature. 1989;341: 149–152. pmid:2779653
  24. 24. Tapia-Arancibia L, Aliaga E, Silhol M, Arancibia S. New insights into brain BDNF function in normal aging and Alzheimer disease. Brain Res Rev. 2008;59: 201–220. pmid:18708092
  25. 25. Hayley S, Anisman H. Neurotrophic paths in the treatment of depression. J Psychiatry Neurosci. 2013;38: 291–293. pmid:23972006
  26. 26. Taylor WD, Züchner S, McQuoid DR, Steffens DC, Speer MC, Krishnan KRR. Allelic differences in the brain-derived neurotrophic factor val66met polymorphism in late-life depression. Am J Geriatr Psychiatry. 2007;15: 850–857. pmid:17911362
  27. 27. Taylor WD, McQuoid DR, Ashley-Koch A, MacFall JR, Bridgers J, Krishnan RR, et al. BDNF val66met genotype and 6-month remission rates in late-life depression. Pharmacogenomics J. 2011;11: 146–154. pmid:20195291
  28. 28. Kim J, Stewart R, Kim S, Yang S, Shin I, Kim Y, et al. Interactions between life stressors and susceptibility genes (5-httlpr and BDNF) on depression in Korean elders. Biol Psychiatry. 2007;62: 423–428. pmid:17482146
  29. 29. Pei Y, Smith AK, Wang Y, Pan Y, Yang J, Chen Q, et al. The brain-derived neurotrophic-factor (BDNF) val66met polymorphism is associated with geriatric depression: a meta-analysis. Am J Med Genet B Neuropsychiatr Genet. 2012;159B: 560–566. pmid:22610920
  30. 30. Egan MF, Kojima M, Callicott JH, Goldberg TE, Kolachana BS, Bertolino A, et al. The BDNF val66met polymorphism affects activity-dependent secretion of BDNF and human memory and hippocampal function. Cell. 2003;112: 257–269. pmid:12553913
  31. 31. Chen Z, Patel PD, Sant G, Meng C, Teng KK, Hempstead BL, et al. Variant brain-derived neurotrophic factor (BDNF) (met66) alters the intracellular trafficking and activity-dependent secretion of wild-type BDNF in neurosecretory cells and cortical neurons. J Neurosci. 2004;24: 4401–4411. pmid:15128854
  32. 32. Hariri AR, Goldberg TE, Mattay VS, Kolachana BS, Callicott JH, Egan MF, et al. Brain-derived neurotrophic factor val66met polymorphism affects human memory-related hippocampal activity and predicts memory performance. J Neurosci. 2003;23: 6690–6694. pmid:12890761
  33. 33. Bueller JA, Aftab M, Sen S, Gomez-Hassan D, Burmeister M, Zubieta J. BDNF val66met allele is associated with reduced hippocampal volume in healthy subjects. Biol Psychiatry. 2006;59: 812–815. pmid:16442082
  34. 34. MacQueen G, Frodl T. The hippocampus in major depression: evidence for the convergence of the bench and bedside in psychiatric research? Mol Psychiatry. 2011;16: 252–264. pmid:20661246
  35. 35. Duman RS. Neuronal damage and protection in the pathophysiology and treatment of psychiatric illness: stress and depression. Dialogues Clin Neurosci. 2009;11: 239–255. pmid:19877493
  36. 36. Martinowich K, Manji H, Lu B. New insights into BDNF function in depression and anxiety. Nat Neurosci. 2007;10: 1089–1093. pmid:17726474
  37. 37. Heils A, Teufel A, Petri S, Stöber G, Riederer P, Bengel D, et al. Allelic variation of human serotonin transporter gene expression. J Neurochem. 1996;66: 2621–2624. pmid:8632190
  38. 38. Wendland JR, Martin BJ, Kruse MR, Lesch K, Murphy DL. Simultaneous genotyping of four functional loci of human SLC6A4, with a reappraisal of 5-HTTLPR and rs25531. Mol Psychiatry. 2006;11: 224–226. pmid:16402131
  39. 39. Heils A, Teufel A, Petri S, Seemann M, Bengel D, Balling U, et al. Functional promoter and polyadenylation site mapping of the human serotonin (5-HT) transporter gene. J Neural Transm. 1995;102: 247–254. pmid:8788073
  40. 40. Nakatani D, Sato H, Sakata Y, Shiotani I, Kinjo K, Mizuno H, et al. Influence of serotonin transporter gene polymorphism on depressive symptoms and new cardiac events after acute myocardial infarction. Am Heart J. 2005;150: 652–658. pmid:16209960
  41. 41. Lesch K, Mössner R. Genetically driven variation in serotonin uptake: is there a link to affective spectrum, neurodevelopmental, and neurodegenerative disorders? Biol Psychiatry. 1998;44: 179–192. pmid:9693390
  42. 42. Caspi A, Sugden K, Moffitt TE, Taylor A, Craig IW, Harrington H, et al. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science. 2003;301: 386–389. pmid:12869766
  43. 43. Lesch K, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S, et al. Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science. 1996;274: 1527–1531. pmid:8929413
  44. 44. Otte C, McCaffery J, Ali S, Whooley MA. Association of a serotonin transporter polymorphism (5-HTTLPR) with depression, perceived stress, and norepinephrine in patients with coronary disease: the Heart and Soul study. Am J Psychiatry. 2007;164: 1379–1384. pmid:17728423
  45. 45. Kohen R, Cain KC, Mitchell PH, Becker K, Buzaitis A, Millard SP, et al. Association of serotonin transporter gene polymorphisms with poststroke depression. Arch Gen Psychiatry. 2008;65: 1296–1302. pmid:18981341
  46. 46. Lenze EJ, Munin MC, Ferrell RE, Pollock BG, Skidmore E, Lotrich F, et al. Association of the serotonin transporter gene-linked polymorphic region (5-HTTLPR) genotype with depression in elderly persons after hip fracture. Am J Geriatr Psychiatry. 2005;13: 428–432. pmid:15879594
  47. 47. Albert PR, François BL, Millar AM. Transcriptional dysregulation of 5-HT1A autoreceptors in mental illness. Mol Brain. 2011;4: 21–34. pmid:21619616
  48. 48. Kishi T, Yoshimura R, Fukuo Y, Okochi T, Matsunaga S, Umene-Nakano W, et al. The serotonin 1A receptor gene confer susceptibility to mood disorders: results from an extended meta-analysis of patients with major depression and bipolar disorder. Eur Arch Psychiatry Clin Neurosci. 2013;263: 105–118. pmid:22752684
  49. 49. Lemonde S, Turecki G, Bakish D, Du L, Hrdina PD, Bown CD, et al. Impaired repression at a 5-hydroxytryptamine 1a receptor gene polymorphism associated with major depression and suicide. J Neurosci. 2003;23: 8788–8799. pmid:14507979
  50. 50. Lenze EJ, Shardell M, Ferrell RE, Orwig D, Yu-Yahiro J, Hawkes W, et al. Association of serotonin-1A and 2A receptor promoter polymorphisms with depressive symptoms and functional recovery in elderly persons after hip fracture. J Affect Disord. 2008;111: 61–66. pmid:18334271
  51. 51. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry. 1979;134: 382–389. pmid:444788
  52. 52. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I), Clinician Version, Administration Booklet. American Psychiatric Pub; 2012.
  53. 53. Zuckerman J, Koval K, Aharonoff G, Hiebert R, Skovron M. A functional recovery score for elderly hip fracture patients: i. development. J Orthop Trauma January 2000. 2000;14: 20–25. pmid:10630798
  54. 54. Katzman R, Brown T, Fuld P, Peck A, Schechter R, Schimmel H. Validation of a short Orientation-Memory-Concentration Test of cognitive impairment. Am J Psychiatry. 1983;140: 734–739. pmid:6846631
  55. 55. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav. 1983;24: 385–396. pmid:6668417
  56. 56. Hayes AF. Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press; 2013.
  57. 57. Hayes AF, Preacher KJ. Statistical mediation analysis with a multicategorical independent variable. Br J Math Stat Psychol. 2013;
  58. 58. Hayes AF. Beyond Baron and Kenny: statistical mediation analysis in the new millennium. Commun Monogr. 2009;76: 408–420.
  59. 59. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155: 945–959. pmid:10835412
  60. 60. Earl DA, vonHoldt BM. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv Genet Resour. 2011;
  61. 61. Petryshen TL, Sabeti PC, Aldinger KA, Fry B, Fan JB, Schaffner SF, et al. Population genetic study of the brain-derived neurotrophic factor (BDNF) gene. Mol Psychiatry. 2010;15: 810–815. pmid:19255578
  62. 62. Murdoch JD, Speed WC, Pakstis AJ, Heffelfinger CE, Kidd KK. Worldwide population variation and haplotype analysis at the serotonin transporter gene SLC6A4 and implications for association studies. Biol Psychiatry. 2013;74: 879–889. pmid:23510579
  63. 63. Prina AM, Deeg D, Brayne C, Beekman A, Huisman M. The association between depressive symptoms and non-psychiatric hospitalisation in older adults. PLoS ONE. 2012;7: e34821. pmid:22496867
  64. 64. Katon WJ. Epidemiology and treatment of depression in patients with chronic medical illness. Dialogues Clin Neurosci. 2011;13: 7–23. pmid:21485743
  65. 65. Yu H, Chen Z. The role of BDNF in depression on the basis of its location in the neural circuitry. Acta Pharmacol Sin. 2011;32: 3–11. pmid:21131999
  66. 66. Chen Z, Bath K, McEwen B, Hempstead B, Lee F. Impact of genetic variant BDNF (val66met) on brain structure and function. Novartis Found Symp. 2008;289: 180–195. pmid:18497103
  67. 67. Risch N, Herrell R, Lehner T, Liang KY, Eaves L, Hoh J, et al. Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk of depression: a meta-analysis. JAMA. 2009;301: 2462–2471. pmid:19531786
  68. 68. Karg K, Burmeister M, Shedden K, Sen S. The serotonin transporter promoter variant (5-HTTLPR), stress, and depression meta-analysis revisited: evidence of genetic moderation. Arch Gen Psychiatry. 2011;68: 444–454. pmid:21199959
  69. 69. Munafò MR, Durrant C, Lewis G, Flint J. Gene × environment interactions at the serotonin transporter locus. Biol Psychiatry. 2009;65: 211–219. pmid:18691701
  70. 70. Porcelli S, Drago A, Fabbri C, Gibiino S, Calati R, Serretti A. Pharmacogenetics of antidepressant response. J Psychiatry Neurosci. 2011;36: 87–113. pmid:21172166
  71. 71. Taylor WD, Steffens DC, Payne ME, MacFall JR, Marchuk DA, Svenson IK, et al. Influence of serotonin transporter promoter region polymorphisms on hippocampal volumes in late-life depression. Arch Gen Psychiatry. 2005;62: 537–544. pmid:15867107
  72. 72. Martinowich K, Lu B. Interaction between BDNF and serotonin: role in mood disorders. Neuropsychopharmacology. 2007;33: 73–83. pmid:17882234
  73. 73. Mattson MP, Maudsley S, Martin B. BDNF and 5-HT: a dynamic duo in age-related neuronal plasticity and neurodegenerative disorders. Trends Neurosci. 2004;27: 589–594. pmid:15374669
  74. 74. Bogdan R, Hyde LW, Hariri AR. A neurogenetics approach to understanding individual differences in brain, behavior, and risk for psychopathology. Mol Psychiatry. 2013;18: 288–299. pmid:22614291
  75. 75. Terracciano A, Tanaka T, Sutin AR, Deiana B, Balaci L, Sanna S, et al. BDNF val66met is associated with introversion and interacts with 5-HTTLPR to influence neuroticism. Neuropsychopharmacology. 2010;35: 1083–1089. pmid:20042999
  76. 76. Dougherty LR, Klein DN, Congdon E, Canli T, Hayden EP. Interaction between 5-HTTLPR and BDNF Val66Met polymorphisms on HPA axis reactivity in preschoolers. Biol Psychol. 2010;83: 93–100. pmid:19914329
  77. 77. Kaufman J, Yang B, Douglas-Palumberi H, Grasso D, Lipschitz D, Houshyar S, et al. Brain-derived neurotrophic factor–5-HTTLPR gene interactions and environmental modifiers of depression in children. Biol Psychiatry. 2006;59: 673–680. pmid:16458264
  78. 78. Shimizu E, Hashimoto K, Iyo M. Ethnic difference of the BDNF 196G/A (val66met) polymorphism frequencies: the possibility to explain ethnic mental traits. Am J Med Genet B Neuropsychiatr Genet. 2004;126B: 122–123. pmid:15048661
  79. 79. Kendler KS. What psychiatric genetics has taught us about the nature of psychiatric illness and what is left to learn. Mol Psychiatry. 2013;18: 1058–1066. pmid:23628988