Associating Leukocyte Telomere
Length with Functional Brain
Measures During Cognitive and
Emotional Processing
ABSTRACT
As we grow older we do not merely increase in chronological age, we are subjected to cellular aging as well, as indexed by leukocyte telomere length (LTL). While chronological aging is associated with cognitive decline, it remains largely unclear how LTL affects the neurobiological mechanisms underlying cognitive functioning. This research therefore associates LTL with brain activity during an emotional word encoding and recognition task, by using event-related fMRI on a large study sample of n = 167. In this task, participants had to encode, and after a retention interval, recognize words of different valences (positive, negative, neutral). During successful encoding, regardless of valence, brain activity in the left superior parietal lobe (BA 5) was inversely correlated with LTL. Specifically during successful encoding of positive words, brain activity in the left inferior parietal lobe (BA 40) was positively correlated with LTL. With regard to negative words, brain activity in the right middle frontal gyrus (BA 8/9) during successful encoding was positively correlated with LTL. In addition, decreased activity in this latter brain region (right middle frontal gyrus) associated with shorter LTL during encoding of negative words was also trend wise associated with poorer task performance (p = 0.056). No associations between brain activity and LTL were observed in the recognition phase of the task. These findings show that shorter LTL is primarily associated with decreased activity in a network encompassing cognitive and executive brain regions during encoding of emotional words. This suggests that cellular aging predominantly acts on the neural system that mediates memory formation, rather than recollection. Word count 257
L.K.M. Han
6175937
Lianne Schmaal
Word count 6596
12-12-2013
1
Introduction
As we grow older we do not merely increase in chronological age, the cells in our body are subjected to biological aging as well. Currently, leukocyte telomere length (LTL) is a commonly used biological marker to measure biological aging – also cellular aging -, and is expressed in base pairs (bp) (Dahse et al., 1997). Telomeres are simple DNA tandem repeats that cap off the ends of chromosomes, protecting them from DNA damage response machinery (O’Sullivan et al., 2010). Consequently, loss of telomeres facilitates increased genetic recombination and end-to-end chromosomal fusions, leading to chromosome, and, therefore, genome instability (Masutomi et al, 2010). Generally, leukocyte telomeres are considered a biological mechanism associated with cellular aging (Sibille et al., 2012; O’Donovan et al., 2012), because they become progressively shorter with every cell division (Wikgren et al., 2012a). As a result, telomeric DNA components can eventually become critically short, such that it leads to cellular senescence or apoptosis (Harris et al., 2006). However, the loss of telomeric repeats can be compensated by a cellular enzyme called telomerase. Telomerase counteracts telomere shortening by adding telomeric DNA, and is particularly active in stem-, germ-, and cancer cells (Lin et al., 2010; Hartmann et al., 2010). Other human cells largely express very low quantities of telomerase, and, correspondingly, have limited cell growth.
While cellular aging is a natural consequence of chronological aging in general, there are multiple determinants associated with the speed of cellular aging. For example, previous studies have shown that several chronic mood disorders such as major depressive disorder (MDD) are associated with shortened LTL, and thus accelerated cellular aging (Garcia-Rizo et al., 2012; Lung et al., 2007). Most recently, our own research group confirmed this finding, using data from the Netherlands Study of Depression and Anxiety (NESDA) (Penninx et al., 2008). In specific, our results demonstrated that MDD patients showed accelerated cellular aging according to a “dose-response” gradient. That is to say, patients with more severe and chronic MDD showed more LTL shortening (Verhoeven et al., in press). It is suggested that biochemical stressors such as oxidative and inflammatory stress are elevated in MDD patients, subsequently contributing to shortened LTL (Wolkowitz et al., 2011; O’Donovan et al., 2012). Several lines of evidence, including a study of our own, thus show that there is variation in LTL between patients suffering from MDD and healthy controls (HC) (Simon et al., 2006).
In addition, MDD is not only associated with accelerated cellular aging, but also with impairment in both cognitive and emotional processing (Woudstra et al, 2012; McDermott & Ebmeier, 2009). Many articles report findings of decreased cognitive functioning in different domains, such as executive functions, attention, memory and psychomotor speed, as well as a deficiency in processing emotional information (Hammer & Ardal, 2009; van Tol et al., 2012). It can be speculated that accelerated cellular aging, as indexed by shortened LTL, may potentially be – or at least partly – affecting these impairments.
As follows, there are numerous studies that have investigated LTL in relation to cognitive performance (Mather et al., 2010; Grodstein et al., 2008). However, research into LTL and cognition report mixed findings. On one hand, a previous study by Wikgren and colleagues shows an inverse correlation of LTL and episodic memory performance (Wikgren et al., 2012b). While, on the other hand, positive correlations of LTL with reaction time and working memory capacity (Valdes et al., 2010) and, similarly, a
2 positive correlation of LTL and scores on the modified mini mental state examination (3MS) (Yaffe et al., 2011) were found. Additional studies reported little or weak evidence for an association of LTL and cognition (Harris et al., 2006a; Harris et al, 2010b). Inconsistent findings may partly be due to small sample sizes, and/or the lack of full adjustment of (potential) influencing confounders. To date, the exact relationship between LTL and cognitive functioning remains to be elucidated.
As far as cognitive functioning is concerned, it is of great relevance to focus on the brain to gain more knowledge. This way, we may be able to reveal a pathway in which cellular aging is able to affect neural mechanisms underlying cognitive functioning. The neural correlates of cognitive functioning in relation to LTL can be studied in twofold: a) through structural measures, and b) through functional measures. Earlier studies have analyzed the relationship between LTL and structural brain volumes (Wikgren et al., 2012c; Grodstein et al., 2008; Thomas et al., 2008), in order to investigate the neuro-anatomical correlates of cellular aging. However, to our knowledge, no research has examined LTL in relation to functional brain measures as yet.
Moreover, controversial findings on the association of LTL and cognitive functioning clearly indicate a need for studying this issue in more detail. Little is known about how precisely changes in LTL affect cognition and emotion in the trajectory of chronological and cellular aging (Kljajevic, 2011). For this reason, the current study aims to investigate the potential relationship of LTL and the neural mechanisms underlying cognitive and emotional functioning. Also, because the available literature on this topic is inconsistent, the current cross sectional research with a large sample size can contribute to a significant enhancement of our understanding of the relationship between LTL and cognitive and emotional processing.
The goal of our study is therefore to associate LTL with functional brain measures. The current sample includes HC, patients diagnosed with MDD, and patients with a diagnosis of co morbid MDD and anxiety disorders (MDD+), in pursuance of extensive LTL variation. In order to obtain measures of LTL, we used DNA extracted from the participant’s blood. To further address the above issues, we used event-related functional magnetic resonance imaging (fMRI) to investigate to which extent LTL contributes to brain activity during, and performance on a task. Specifically, an emotional word encoding and recognition task is used to tap cognitive and emotional memory processing. Thus, this task measures memory formation and recollection, additionally taking into account emotional processing. Previous studies using the same (or a similar) task have shown that particularly the hippocampus, amygdala, prefrontal cortex (PFC), anterior cingulate cortex (ACC), insula, inferior frontal gyrus (IFG) and parietal lobes were implicated in performance of the task (van Tol et al., 2012; Daselaar et al., 2003; Woudstra et al., 2013; Spaniol et al., 2009) .
Given that cognitive functioning and LTL both decrease with chronological age (Salthouse, 2009; Butler et al., 1998), it is hypothesized that cellular aging is associated with abnormal brain activity during, and poorer performance on a task tapping both cognitive and emotional processes. In specific, regarding previous studies, we hypothesized that alterations in brain activity within the hippocampus, amygdala, IFG extending into the insula, inferior and superior parietal lobes (IPL & SPL), medial frontal gyrus (MedFG) extending into the ACC, and the superior and middle frontal gyrus (SFG & MFG) were associated with cellular aging, as indexed by shorter LTL.
Ideally, this present study exposes a pathway in which telomeres are able to exert influence on cognitive performance through underlying neural mechanisms, showing
3 that LTL is not only associated with performance on a cognitive task, but also with brain functioning.
Methods and Materials
Participants
Participants were recruited from NESDA; a large-scale, longitudinal, multi-center, observational cohort study (Penninx et al., 2008) designed to investigate the long term course and consequences of depression and anxiety disorders. The NESDA study took place among 2981 participants including patients with a current or lifetime diagnosis of MDD and/or anxiety disorder and healthy controls, ranging from 18 through 65 years old. Participants were recruited from the general population, general practices and mental health organizations. During recruitment two main exclusion criteria were used: a) for patients; a primary clinical diagnosis other than MDD, Panic Disorder or social anxiety disorder (except generalized anxiety disorder) lifetime, and b) for healthy controls; had to be currently free of, and had never met criteria for, depressive or anxiety disorders or any other axis-I disorder and were not taking any psychotropic drugs. Additionally, for both patients and healthy controls; participants who were not fluent in Dutch were excluded (language problems would impair the validity and reliability of collected data).
A subsample of the NESDA cohort underwent magnetic resonance imaging (MRI) in the Academic Medical Center (AMC), Leiden University Medical Center (LUMC), or University Medical Center Groningen (UMCG). Both structural, as well as functional MRI was performed using a 3T MRI scanner. The study was approved by the Medical Ethics Committees of all three centers. After receiving written information, all subjects provided written informed consent.
For the present study, a total sample of 220 participants was selected from the NESDA neuroimaging substudy. Exclusion criteria were: presence of MRI contraindications, dependence or recent abuse of alcohol and/or drugs, hypertension, major internal and/or neurological disorders, and use of psychotropic medication other than stable use of a selective serotonin reuptake inhibitor (SSRI) or infrequent use of benzodiazepines. We further excluded participants from the current analysis because of missing performance data (n = 7), missing telomere data (n = 3), and/or insufficient task performance (n = 20). Task performance data was considered insufficient if participants had > 40 non-responses and/or a discriminant power (d’) < 0.1 (d’ is defined as the proportion (prop) hits – the propFalse Alarms (FA)), indicating unreliable task involvement. Additional participants were excluded because of technical problems during scanning, such as movement artifacts > 3mm (n = 2), and/or a lack of coverage of the whole brain and loss of voxels in the first-level mask (n = 21). The remaining data from 167 participants with a DSM-IV diagnosis of MDD (n = 49), MDD+ (n = 68), and HC (n = 50) were used in statistical analyses. Further group characteristics of our study population are shown in Table 1.
4
Table 1. Group characteristics of the study participants
Full sample
n 167
Age, years (SD) 36.6 (10.4)
Gender, male/female ratio; n (% male) 56/111 (33.5%) Leukocyte telomere length, bp (SD) 5556 (683) Education level, years attained (SD) 12.7 (3.1) Scan center, AMC/LUMC/UMCG; n 54/70/43 Number of chronic diseases under
treatment, 0/1/2/3/5; n 114/39/12/1/1 Alcohol intake, no/mild/heavy; n 30/125/12 Smoking status, never/former/current; n 64/53/50 Physical activity, MET-minutes per week
(SD) 3636 (3530)
Body Mass Index (BMI),
underweight/normal/overweight/obese; n 4/88/50/25 Use of antidepressants (AD), yes/no; n 38/129 All values are means unless otherwise stated. SD = standard deviation.
Leukocyte telomere length
All baseline assessments took place in the morning between 8:30 and 9:30. This allowed the draw of an overnight fasting blood sample. The blood samples were transferred to a local laboratory; however most of it was processed and stored at -85 °C for later assaying. Actual LTL measures were extracted from the DNA in the blood sample, and determined at the laboratory of Telome Health Inc. (Menlo Park, CA, USA), using quantitative polymerase chain reaction (qPCR), described in detail elsewhere (Cawthon, 2002). Telomere sequence copy number in each participant’s sample (T) was compared to a single-copy gene copy number (S), relative to a reference sample. The resulting T/S ratio is proportional to mean LTL (Aviv et al., 2011; Cawthon, 2002). Subsequently, a conversion formula was composed in order to compare T/S ratios to telomere restriction fragments (TRF). TRF are part of another LTL measure reported by studies using Southern Blot analyses, as opposed to qPCR. The following steps were used to obtain a conversion formula: published work from the Blackburn lab at the University of California San Francisco (UCSF) used a formula of base pairs = 3274 + 2413 *T/S, based on comparison of T/S ratios and TRF analyses of a series of genomic DNA samples from the human fibroblast cell line IMR90 (Lin et al., 2010). Comparison of the T/S ratios of the eight control DNA samples derived from the Blackburn lab and Telome Health lab generated the following formula: T/S (UCSF)=(T/S (Telome Health)-0.0545)/1.16. Therefore, the final formula used to convert T/S ratios to base pairs is: base pairs = 3274 + 2413 x ((T/S-0.0545)/1.16).
Emotional word encoding and recognition task
An event-related, subject-paced, implicit word-encoding and recognition paradigm was used (van Tol et al., 2012), programmed in E-prime software (Psychological Software Tools, Pittsburgh, PA). The task consisted of two parts, namely an encoding and a recognition part. During the encoding part, participants classified 40
5 positive, 40 negative, and 40 neutral words according to their valence. Words were presented together with 40 baseline trials in 20 blocks of eight words, in a pseudo-randomized fashion. In addition, words were presented with a minimum interstimulus interval of 1018 ms and a maximum of 1035 ms. Each block contained two negative words, two positive words, two neutral words and two baseline trials, all presented in randomized order. All words (negative, positive, neutral) were matched for length (three to twelve letters) and frequency of occurrence in the Dutch language. The task was subject-paced, but each word was presented with a maximal duration of 5 seconds. To indicate whether participants assumed that the word presented was negative, positive or neutral to them, they had to press the corresponding button displayed at the bottom of the screen. During baseline trials, participants had to indicate the direction of arrows (<<left, <<middle>>, right>>). To control for primacy and recency effects, three filler words (1 negative, 1 positive, 1 neutral) were presented at the start and end of the encoding part. These filler words were not subsequently tested in the recognition part. The interval between the encoding part and recognition part was 10 minutes. After this retention interval, participants completed a word recognition task. This task was comprised of the 120 old target words of the encoding part, and 120 new distracter words, as well as 40 baseline words. Words were presented pseudo-randomized in 20 blocks of 14 words, each block containing two old and two new negative words, two old and two new positive words, two old and two new neutral words, and two baseline trials. Old and new words were matched on complexity, word length, and emotional intensity. Participants had to indicate whether they had “seen” (i.e. remembered) the words previously, “probably seen” (know), or “not seen” (rejection). Both responses and response times were recorded with two magnet-compatible button boxes. No feedback as to whether their answer was correct was provided.
Image data acquisition
The task-related fMRI acquisition was part of a larger fixed imaging protocol within NESDA. Further functional neuroimaging methods have been described in-depth elsewhere (Demenescu et al., 2011; van Tol et al., 2011; Woudstra et al., 2012). In short, imaging data were acquired using a Philips 3-Tesla MRI system (Best, The Netherlands) located at the LUMC, AMC, and UMCG, equipped with SENSE-8 (LUMC and UMCG) and SENSE-6 (AMC) channel head coils. For each subject, echo-planar images were obtained using a T2*-weighted gradient echo sequence (repetition time = 2300 msec; echo time = 30 msec [UMCG: 28 msec], matrix size: 96x96 [UMCG: 64x64], 35 axial slices [UMCG: 39], interleaved acquisition, 2.29x2.29 mm in-plane resolution [UMCG: 3x3mm], 3 mm slice thickness). Anatomic imaging included a sagittal 3D gradient -echo T1- weighted sequence (repetition time = 9 msec, echo time = 3.5 msec; matrix 256x256; voxel size: 1x1x1 mm; 170 slices).
Data preprocessing
Imaging data were preprocessed with SPM8 software (Statistic Parametric Mapping, http://www.fil.ion.ucl.ac.uk/spm/), implemented in Matlab version 7.11.0 (The Mathworks, Natick, Massachusetts, USA). The following processing steps were applied: reorientation, slice time correction, image realignment, co registration of the T1-scan and the functional images, spatial normalization to Montreal Neurological Institute (MNI) space as defined by the SPM8 T1-template, reslicing to 3x3x3 mm voxels, and spatial smoothing using a Gaussian kernel (8mm FWHM).
6
Statistical analyses
Behavioral and performance data were analyzed with SPSS 20 (SPSS Inc., Chicago, Illinois, USA). Functional imaging data were analyzed using SPM8 software (Statistical Parametric Mapping, http://www.fil.ion.ucl.ac.uk/spm) implemented in Matlab version 7.11.0 (The MathWorks, Natick, Massachusetts, USA).
Behavioral and performance data
Responses were recorded and propHits, propFA, and old/new discriminant accuracy (d’ = propHits – propFA) were calculated, overall and per valence (positive, negative, neutral) (see Table 2 for an overview). A linear regression analysis was used to determine the predictive model of LTL and explore the relationship between LTL and task performance variables. LTL was included in the model as a regressor of interest, with age, gender and education level as covariates.
In addition, bivariate correlation (Pearson’s) analyses were performed to assess the relationships between activities in brain regions associated with LTL and performance data. The statistical significance threshold was set at p < 0.05.
Table 2. Overview of computed task performance variables
Prop, proportion; neg, negative; pos, positive; neu, neutral; Hits, correctly recognized words; FA, False Alarms; d’, discriminant accuracy.
Imaging data
Data were analyzed in the context of the General Linear Model (GLM) (Friston et al., 1995). In a first level, single-subject fixed effects analysis, regressors were constructed by convolving the onsets of the word presentation with a canonical hemodynamic response function and modulated using response times. The model included the following regressors for encoding and recognition parameters: a)
Encoding: Subsequent_Hits_positive, Subsequent_Hits_negative, Subsequent_Hits_neutral, Subsequent_Misses_positive, Subsequent_Misses_negative,
Subsequent _Misses_neutral, baseline trials. (Subsequent refers to whether the word was correctly recognized or missed in the subsequent recognition phase), and b) Recognition: Hits_positive, Hits_negative, Hits_neutral, CorrectRejections_positive, CorrectRejections_negative, CorrectRejections_neutral, FalseAlarms_positive, FalseAlarms_negative, FalseAlarms_neutral, Misses_positive, Misses_negative, Misses_neutral, baseline trials. Additionally, filler words, error- and no-response trials were included as a regressor of no interest. To account for low-frequency signal drift a
Task performance variables
propHits_all Hits_all/amount of target words (120)
propHits_neg Hits_neg/amount of target words (120)
propHits_pos Hits_pos/ amount of target words (120)
propHits_neu Hits_neu/ amount of target words (120)
propFA_all FA_all/ amount of target words (120)
propFA_neg FA_neg/ amount of target words (120)
propFA_pos FA_pos/ amount of target words (120)
propFA_neu FA_neu/ amount of target words (120)
d’_all propHits_all – propFA_all
d’_neg propHits_neg – propFA_neg
d’_pos propHits_pos – propFA_pos
7 high-pass filter (1/128 Hz) was applied. On the grounds that participants rarely responded with a “know” response in the recognition trials, these responses were treated as “remembered” and added to either “Hits” or “False Alarms”. Next, contrast images for “Subsequent_Hits_negative > Subsequent_Hits_neutral”, “Subsequent_Hits_positive > Subsequent_Hits_neutral”, “Subsequent_Hits_all > Encoding_baseline”, resulting from the encoding phase and “Hits_negative > Hits_neutral”, “Hits_positive > Hits_neutral”, “Recognition_all > Recognition_baseline” resulting from the recognition phase, were computed for each subject on a voxel-by-voxel basis. The task-related contrast images were entered in a second-level random effects analysis using a multiple regression design, including LTL as a predictor variable.
Basic covariates include age, gender, education level and dummy variables for the different scan centers. Additionally, all analyses were fully adjusted for (potential) influential confounders such as other chronic diseases, alcohol intake, smoking, physical activity, Body Mass Index (BMI) and the use of antidepressants (AD) (Valdes et al., 2005; Cherkas et al., 2008; Epel et al., 2004; Savolainen et al., 2012).
The covariates were measured as follows: gender, age and education level were assessed during the baseline interview. The number of chronic diseases (i.e. heart disease, epilepsy, diabetes, osteoarthritis, stroke, cancer, chronic lung-disease, thyroid disease, liver disease, chronic fatigue syndrome, intestinal disorders and ulcers) was also assessed and a disease was considered present if participants received medical treatment for it. Alcohol intake was divided into categories of non-drinkers (0 drinks), moderate drinkers (female < 14 and male < 21 drinks/week) or heavy drinkers (female ≥ 14 and male ≥ 21 drinks/week). Smoking status was categorized into current, former or never smoker. Physical activity was assessed using the International Physical Activity Questionnaire (IPAQ) and expressed as one’s resting metabolic rate multiplied by minutes of physical activity per week (MET-minutes per week) (Craig et al., 2003). BMI was calculated as measured weight divided by length2 and subsequently divided into underweight (< 18.5), normal (18.5–24.9), overweight (25.0–30.0) and obese (> 30.0). BMI and alcohol consumption were added as categorical covariates because they were not linearly associated with LTL. The use of AD was assessed at the baseline interview.
We plotted the fitted responses against LTL with SPM8 to extract the effect sizes (y-parameter estimates of activity) of significant activity within the regions of interest (ROIs) associated with LTL obtained by the second-level analyses. The effect size measures were correlated with task performance variables in order to investigate the relationship of brain activity associated with LTL, and task performance.
We focused our analyses on a priori ROIs. Based on previous studies (Daselaar et al., 2003; Spaniol et al., 2009; van Tol et al., 2012; Woudstra et al., 2013), the following ROIs were defined: hippocampus, amygdala, IFG extending into the insula, IPL & SPL, MedFG extending into the ACC, and the SFG & MFG. The six ROIs were selected from the Nielsen and Hansen’s volumes of interest defined in the Brainmap database (Nielsen & Hansen, 2004). For an image of the ROI masks see figure 1. Statistical tests were thresholded at p<0.05 family-wise error (FWE), voxel wise corrected for the spatial extent of the ROI by using small volume correction as implemented in SPM8 (Worsley et al., 1996), in order to be regarded significant.
8
Figure 1. The six ROI masks from the Nielsen and Hansen’s volumes of interest defined in the Brainmap database. A: The hippocampus (blue), amygdala (red) and the
IFG extending into the insula (green). B: The MedFG extending into the ACC (yellow), and the SFG & MFG (cyan). C: The IPL & SPL (violet).
9
Results
First, Pearson correlation coefficients were computed and we found an expected negative correlation between LTL and chronological age (r = -0.299, p < 0.01). This indicates that with chronological aging, LTL indeed decreases. With that, we validate the use of chronological aging as a covariate, in order to examine the association of cellular aging and functional brain activity in specific.
Second, a one way independent ANOVA analysis was used to determine the differences between groups for LTL measures. However, we could not replicate the findings of shortened LTL, and thus accelerated cellular aging in MDD and/or MDD+ patients, compared to HC ( p = 0.58), as reported by Verhoeven et al., (in press), Garcia-Rizo et al. (2012), Lung et al., (2007), Wolkowitz et al., (2011) and O’Donovan et al., (2012). Therefore, we chose to do all analyses over the entire sample of MDD and MDD+ patients, and HC.
Behavioral task performance & leukocyte telomere length
To determine the relationship between behavioral task performance variables and LTL, linear regression analyses were performed. However, the results did not confirm any significant age, gender, and education level-adjusted associations between LTL and performance data (see table 3).
Table 3. Relationship between behavioral task performance variables and leukocyte telomere length (bp).
Dependent variable: Leukocyte Telomere Length (bp). Age, gender and education level were included in the model as covariates. Prop, proportion; neg, negative; pos, positive; neu, neutral; Hits, correctly recognized words; FA, False Alarms; d’, discriminant accuracy.
Unstandardized coefficients B Std. Error p-value propHits_all -124,04 416,93 0,77 propHits_neg -5,06 369,61 0,99 propHits_pos 5,73 378,63 0,99 propHits_neu -206,98 310,39 0,51 propFA_all 164,06 719,92 0,82 propFA_neg 36,74 509,90 0,94 propFA_pos 688,50 534,82 0,20 propFA_neu -1181,78 861,31 0,17 d’_all -179,78 417,62 0,67 d’_neg -31,42 419,75 0,94 d’_pos -288,21 350,01 0,41 d’_neu -52,56 304,24 0,86
10
fMRI results
The results of the current study are presented in two sections. First, we summarize the associations of LTL and brain activity during memory encoding and recognition across emotional conditions. Second, we examine the interaction effect of emotion on the association between memory and LTL, by investigating brain activation during encoding and recognition of negative and positive words, compared to neutral words.
Memory
Encoding Figure 2 shows activation in the brain area inversely correlated with
LTL, demonstrating that shorter LTL was associated with higher brain activity during successful encoding, than during baseline. That is to say, the left superior parietal lobe/postcentral gyrus (Brodmann area (BA) 5); x = -21mm, y = -40mm, z = 55mm; Z = 4.73; p < 0.05 FWE small-volume corrected; cluster size = 24; coordinates in MNI space) was more active during encoding of words, regardless of valence, in subjects with shorter LTL.
Recognition No main effect of LTL on memory recognition was observed at the
set threshold.
Figure 2. BOLD increase associated with shorter leukocyte telomere length for the comparison of successful encoding more than baseline. The left superior parietal lobe
(BA 5) was inversely correlated with leukocyte telomere length and showed higher activity during successful encoding, than during baseline. Color bar indicates t value. The threshold of the activation maps was set at p < 0.001 uncorrected for display purposes.
11
Interaction of memory with emotion
Encoding During successful encoding of positive words compared with neutral
words, we observed an association between activity in the left inferior parietal lobe (BA 40) and LTL (x = -54mm, y = -37mm, z = 25mm; Z = 4.16; p < 0.05 FWE small volume corrected; cluster size = 26; coordinates in MNI space), see figure 4. LTL was positively correlated with brain activity in the left inferior parietal lobe; indicating decreased activity in this brain area during successful encoding of positive words in participants with shorter LTL.
Figure 4. The association of leukocyte telomere length and successful encoding of positive words, more than neutral words. The left inferior parietal lobe (BA 40) was
positively correlated with leukocyte telomere length and showed decreased activity during successful encoding of positive words in participants with shorter LTL. Color bar indicates t value. The threshold of the activation maps was set a p < 0.001 uncorrected for display purposes.
We found a positive correlation between brain activation in the right middle frontal gyrus (BA 8/9) during successful encoding of negative compared to neutral words and LTL (x = 42mm, y = 8mm, z = 43mm; Z = 4.08; p < 0.05 FWE small volume corrected; cluster size = 22; coordinates in MNI space), as can be seen in figure 3. In other words, there was decreased activation in the right middle frontal gyrus during successful encoding of negative words (compared to neutral words) in participants with shorter LTL.
Figure 3. The association between leukocyte telomere length and brain activation during successful encoding of negative words more than neutral words. The right
middle frontal gyrus was positively correlated with leukocyte telomere length and showed decreased activity during successful encoding of negative words in participants with shorter LTL. Color bar indicates t value. The threshold of the activation maps was set a p < 0.001 uncorrected for display purposes.
Recognition We did not find any significant associations of LTL and the
12
Relationship between brain regions associated with LTL & task performance
To further investigate the relationships between brain activity in regions associated with LTL, and behavioral task performance, we extracted parameter estimates of significant activity in the left superior parietal lobe (BA 5) (main effect of encoding), the left inferior parietal lobe (BA 40)(successful encoding of positive words), and the right middle frontal gyrus (BA 8/9) (successful encoding of negative words), and correlated these with task performance variables.
Pearson correlation coefficients were computed and we found a positive correlation between the effect sizes in the right middle frontal gyrus and discriminant accuracy of negative words (r = 0.15, p = 0.056), showing a trend towards significance. A scatterplot summarizes the results (see Figure 5). This indicates that decreased brain activity associated with LTL in the right middle frontal gyrus was correlated with a diminished ability to discriminate between negative target words and negative distractor words.
There were no significant correlations between brain activity in the other areas associated with LTL and task performance variables.
13 -2 -1 0 1 2 3 4 5 0,1 0,3 0,5 0,7 0,9 Ef fe ct S iz es in th e r ig ht m id dl e f ro nt al g yr us
Discriminant Accuracy of Negative Words (d'_neg)
Figure 5. The relationship between brain activity in the right middle frontal gyrus (BA 8/9) associated with leukocyte telomere length and task performance. On the top, an
image of brain activity associated with leukocyte telomere length in the right middle frontal gyrus is shown. On the bottom, we plotted a near significant positive correlation of the effect sizes in the middle frontal gyrus positively associated with LTL and discriminant accuracy of negative words.
14
Discussion
In this study, we investigated the neural correlates associated with cellular aging during cognitive and emotional processing. Using fMRI and a large sample including MDD and MDD+ patients, and HC (in pursuance of extensive LTL variation), we associated LTL with brain activity during an emotional word encoding and recognition task. Analyses were fully adjusted for (potential) influencing confounders including age, gender, education level, number of chronic diseases, alcohol intake, smoking status, physical activity, BMI and the use of antidepressants (Valdes et al., 2005; Cherkas et al., 2008; Epel et al., 2004; Savolainen et al., 2012). We identified several specific cognitive, attentional and executive brain regions that were associated with cellular aging during cognitive and emotional functioning. In addition, our results showed that decreased activity in an executive brain region associated with shorter LTL, correlates with poorer task performance (in a borderline statistical significant fashion with p = 0.056). More specific, the present study yielded four main findings that will be further discussed.
First, concerning memory regardless of emotion, we found an inverse correlation between brain activity in the left superior parietal lobe (BA 5) and LTL, during successful encoding. In other words, participants with shorter LTL showed higher brain activity in the left superior parietal lobe in order to achieve successful encoding. Previous studies show that the superior parietal lobe is often activated during memory retrieval, rather than during memory encoding (Spaniol et al., 2009; Vilberg & Rugg, 2008). Nevertheless, a large number of other studies show that the superior parietal lobe is involved in top-down attentional orienting, as well as in the support of cognitive and executive control functions, important for goal-directed behavior (Shomstein, 2012; Uncapher & Rugg, 2005; Cabeza et al., 2008). Needless to say, memory and attention are highly intertwined during encoding, because top-down attention improves perception, acts as a filter and maintains memory during delay periods (Carrasco et al., 2004; Olson & Berryhill, 2009). Because LTL was not directly associated with task performance, we suggest that increased superior parietal lobe activation in participants with shorter LTL reflects a compensatory mechanism. This compensation is postulated to be required for the direction of sufficient top-down attention, essential for successful encoding. Similarly, a study carried out by Krauel et al. (2007), showed that patients suffering from attention deficit hyperactivity disorder (ADHD) recruited attentional compensation by also increasing superior parietal lobe activation during encoding of pictures. It is therefore speculated that the superior parietal lobe facilitates memory encoding by means of an attentional account.
Furthermore, the present study showed that the right middle frontal gyrus (BA 8/9) was positively correlated with LTL and thus showed decreased activity during encoding of negative words (compared to neutral words) in participants with shorter LTL. This effect was mainly driven by the negative encoding condition, as became evident when looking at brain activity during successful encoding of negative words, compared to a low-level baseline. In this contrast, the same brain area (the left middle frontal gyrus (BA 8/9)) was found to be less active in participants with shorter LTL, whereas this result was absent in the comparison of successfully encoded neutral words with the baseline condition. Moreover, brain activity associated with LTL in this specific brain region was trend wise correlated with task performance. That is, decreased right middle frontal gyrus (BA 8/9) activation that was associated with shorter LTL was correlated with a diminished ability to discriminate between negative target words and negative distractor words in the recognition part of the task. In this light, we showed
15 that diminished brain activity during encoding of negative words is not only associated with shorter LTL, but also with poorer behavioral task performance. Furthermore, BA 9 in the middle frontal gyrus contributes to the dorsolateral prefrontal cortex (DLPFC). Previous studies have reported that DLPFC activity correlates with successful encoding (Blumenfeld et al., 2011). Moreover, it is suggested that the DLPFC contributes to successful encoding through maintenance and mental manipulation of memory contents (Ptak, 2012). Additional studies show that the DLPFC promotes long term memory formation through its role in the organization of information in working memory, and is also involved in the self-initiation of elaborative cognitive processing and memory strategy use (Blumenfeld & Ranganath, 2006; Hawco et al., 2013). Our results suggest that cellular aging is associated with decreased DLPFC activity, and that this decreased activity is subsequently correlated to poorer task performance through diminished DLPFC functions that facilitate successful encoding.
We further observed an association of shorter LTL and decreased brain activity in the left inferior parietal lobe during successful encoding of positive words. The inferior parietal lobe is involved in various cognitive functions, and its functional diversity includes, amongst others, attention, memory, higher social cognition, sustaining attention, stimulus saliency and other integrative processes (Singh-Curry & Husain, 2009; Müller et al., 2013). The fact that participants with shorter LTL activate the inferior parietal lobe to a lesser extent during successful encoding of positive words may be due to decreased ability to engage in more effective encoding strategies. In addition, inferior parietal activation is associated with capture of attention by salient stimuli (Cabeza et al., 2008). We suggest that bottom-up attention, the ability to integrate processes and to engage in effective encoding strategies provides a facilitary role for memory encoding, through the exertion of cognitive control.
In addition, considering that we found cellular aging to be generally associated with cognitive, attentional and executive brain areas, we suggest that emotional and memory processing associated with LTL should be placed in a much broader framework. It seems that the results are more readily explained from a wider network perspective, encompassing a distributed network of brain regions. Clearly, a disperse network of attentional and executive regions, and its dynamics are associated with LTL during memory encoding. Similar to other theoretical accounts, we propose that cognitive and emotional functioning associated with cellular aging can be explained in terms of a more general impairment in associative, attentional and integrative processing (Craik & Bird, 1982). The superior and inferior parietal lobes have extensive reciprocal connections with BA 8/9 in the middle frontal gyrus (Miller & Cohen, 2001; Schmahmann & Pandya, 2006), and all regions seem to be important for engaging in effective encoding strategies. Moreover, these are the brain regions that, among others, are frequently classified as areas supporting task-general executive and control processes (Uncapher & Rugg, 2005). In this light, cellular aging seems to be associated with a network encompassing a diverse functionality. This network mainly controls cognitive, attentional and executive functions, and correspondingly facilitates memory formation.
Most importantly, it is of great interest that all of the results were found in the encoding condition, as opposed to recognition. This strongly suggests that cellular aging, as indexed by LTL, is primarily associated with the neural system that mediates memory formation. This is in line with the idea that episodic memory impairments associated with chronological aging mostly appear to be derived from difficulties in encoding, with recollection playing a role to a lesser extent (Morcom et al., 2003; Daselaar et al., 2003).
16 This is of particular relevance to the development of therapeutic cognitive training that, to some extent, can counteract and/or reduce age-related losses in encoding efficiency.
In contrast to structural MRI studies, we did not find LTL to be associated with more conventional memory encoding regions such as the hippocampus (Wikgren et al., 2012; Grodstein et al., 2008). Apparently, when looking at cellular aging in relation to functional brain measures, different neural systems seem to be associated. To our knowledge, we are the first study to show that LTL is associated with distinct cognitive, attentional and executive regions. However, participants with shorter LTL showed increased superior parietal lobe activation, whereas on the other hand decreased activations in the inferior parietal lobe and middle frontal gyrus were found. It is therefore of great interest for future studies to focus on mapping functional connectivity between regions encompassing this memory encoding network associated with cellular aging. Besides, functional connectivity mapping may also potentially reveal how brain areas associated with LTL found in the current study are connected to conventional memory encoding areas, such as the hippocampus.
Furthermore, it should be noted that telomere length of cells derived from the central nervous system can only be studied post-mortem. As a result, telomere length obtained from leukocytes is indirectly used as a measure for telomere length in the brain. While telomere lengths can differ in different tissues within the same individual, there is substantial support for synchronicity across tissues (Okuda et al., 2002; Butler et al., 1998; Youngren et al., 1998; Takubo et al., 2002). That is to say, individuals with relatively long telomeres in the periphery (leukocytes), have relatively long telomeres in the brain (Wikgren, 2011). However, telomeres derived from the central nervous system are more often longer than leukocyte telomeres, due to limited cell proliferation in the brain.
In addition, we also have several remarks on limitations regarding our study. First, similar to Hartmann and colleagues (2010), and Hoen and colleagues (2012), we were not able to replicate a higher age dependent telomere shortening rate in MDD and MDD+ patient groups, compared to HC. Even more, we were not able to replicate the aforementioned finding within the same dataset as Verhoeven and colleagues (in press) did. Nevertheless, this can be explained by the fact that the current study adhered to even more strict inclusion criteria for the MRI substudy, resulting in a relatively small sample size of n = 167 (vs. n = 2407), limiting statistical power. Second, our study was cross sectional in nature, but in order to fully capture how telomeres are able to exert influence on cognitive and emotional brain processes, we should follow telomere length over a considerable length of time in individual participants. In such manner, we will be able to track brain changes over time associated with LTL, and with that infer some causality on the relationship between LTL and abnormalities in brain activations. Lastly, we did not have data on telomerase activity in any of the participants. Inclusion of telomerase data would enable us to investigate the interplay of the LTL/telomerase maintenance system, and more importantly, how disturbances in this balance affect brain functioning. Nevertheless, a major strength concerning the current study is the large imaging sample, lending confidence to our findings.
In conclusion, our findings delineate how cellular aging is associated with brain functioning in cognitive and emotional processing, showing that LTL is primarily associated with cognitive, attentional and executive brain regions such as the left superior parietal lobe (BA 5), right middle frontal gyrus (BA 8/9) and left inferior parietal lobe (BA 40). To our knowledge, this is the first study to show that LTL is indeed associated with brain activity in these brain regions, while the direct relationship
17 between LTL and task performance was not necessarily evident. However, our results do indicate that participants with shorter LTL are less able to activate the middle frontal gyrus (BA 8/9) during successful encoding of negative words, and that this diminished activation is associated with poorer task performance. Furthermore, what is most striking is that LTL is specifically associated with brain activity during the encoding part of the task, and not during recognition. These results strongly suggest that cellular aging predominantly acts on the neural system that mediates memory formation, rather than recollection. Taken together, these findings contribute to an enhanced understanding of memory encoding and recognition in cellular aging. A future research goal should be to investigate functional connectivity between brain regions associated with cellular aging. Additionally, neural mechanisms should be studied longitudinally in individuals, in order to monitor how precisely differences in LTL affect brain functioning during the trajectory of both chronological, as well as cellular aging.
18
References
Blumenfeld, R.S., Parks, C.M., Yonelinas, A.P. & Ranganath, C. (2011). Putting the pieces together: the role of dorsolateral prefrontal cortex in relational memory encoding. Journal of Cognitive Neuroscience, 23(1), 257–65. doi:10.1162/jocn.2010.21459
Butler, M.G., Tilburt, J., DeVries, A., Muralidhar, B., Aue, G., Hedges, L., Atkinson, J. & Schwartz, H. (1998). Comparison of chromosome telomere integrity in multiple tissues from subjects at different ages. Cancer Genetics and Cytogenetics, 105, 2, 138-1344.
Cabeza, R., Ciaramelli, E., Olson, I. R. & Moscovitch, M. (2008). The parietal cortex and episodic memory: an attentional account. Nature reviews. Neuroscience, 9(8), 613– 25. doi:10.1038/nrn2459
Carrasco, M., Ling, S. & Read, S. (2004). Attention alters appearance. Nature Neuroscience, 7(3), 308–13. doi:10.1038/nn1194
Cawthon, R.M. (2002). Telomere measurement by quantitative PCR. Nucleic Acids Research, 30, 10, 47.
Cherkas, L.F., Hunkin, J.L., Kato, B.S., Richards, J.B., Gardner, J.P., Surdulescu, G.L., Kimura, M., Lu, X., Spector, T.D. & Aviv, A. (2008). The association between physical activity in leisure time and leukocyte telomere length. Archives of Internal Medicine, 168, 2, 154-158.
Craig, C. L., Marshall, A.L., Sjöström, M., Bauman, A.E., Booth, M.L., Ainsworth, B.E., Pratt, M., Ekelund, U., Yngve, A., Sallis, J.F. & Oja, P. (2003). International physical activity questionnaire: 12-country reliability and validity. Medicine and Science in Sports and Exercise, 35(8), 1381–95. doi:10.1249/01.MSS.0000078924.61453.FB
Dahse, R., Fiedler, W. & Ernst, G. (1997). Telomeres and telomerase: biological and clinical importance. Clinical Chemistry, 43, 5, 708-714.
Damoiseaux, J.S. & Greicius, M.D. (2009). Greater than the sum of its parts: a review of studies combining structural connectivity and resting-state functional connectivity. Brain Structure & Function, 213(6), 525–33. doi:10.1007/s00429-009-0208-6
Daselaar, S.M., Veltman, D.J, Rombouts, S.A.R.B., Raaijmakers, J.G.W. & Jonker, C. (2003). Neuroanatomical correlates of episodic encoding and retrieval in young and elderly subjects. Brain, 126(1), 43–56. doi:10.1093/brain/awg005
Demenescu, L.R., Renken, R., Kortekaas, R., Van Tol, M.-J., Marsman, J.B.C., Van Buchem, M.A. & Van der Wee, N.J.A. (2011). Neural correlates of perception of emotional facial expressions in out-patients with mild-to-moderate depression and anxiety. A multicenter fMRI study. Psychological Medicine, 41(11), 2253–64. doi:10.1017/S0033291711000596
19 Epel, E.S., Blackburn, E.H., Lin, J., Dhabhar, F.S., Adler, N.E., Morrow, J.D. & Cawthon, R.M. (2004). Accelerated telomere shortening in response to life stress. PNAS, 101, 49, 17312-17315.
Friston, K. J., Holmes, A. P., Poline, J.-B., Grasby, P. J., Williams, S.C.R., Frackowiak, R.S.J. & Turner, R. (1995). Analysis of fMRI time-series revisited. NeuroImage , 2, 45–53.
Garcia-Rizo, C., Fernandez-Egea, E., Miller, B. J., Oliveira, C., Justicia, A., Griffith, J. K., Heaphy, C. M., Bernardo, M. & Kirkpatrick, B. (2013). Abnormal glucose tolerance, white blood cell count, and telomere length in newly diagnosed, antidepressant-naïve patients with depression. Brain, Behavior, and Immunity, 28, 49–53. doi:10.1016/j.bbi.2012.11.009
Grodstein, F., van Oijen, M., Irizarry, M.C., Rosas, H.D., Hyman, B.T., Growdon, J.H. & De Vivo, I. (2008). Shorter telomeres may mark the risk of early risk of dementia: preliminary analysis of 61 participants from the nurses’ health study. PLoSone, 3, 2, e1590.
Hammar, A. & Ardal, G. (2009). Cognitive functioning in major depression – a summary. Frontiers in Human Neuroscience, 3, 26.
Harris, S.E., Deary, I.J., MacIntyre, A., Lamb, K.J., Radhakrishnan, K., Starr, J.M., Whalley, L.J. & Shiels, P.G. (2006a). The association between telomere length, physical health, cognitive ageing, and mortality in non-demented older people. Neuroscience Letters, 406(3), 260–264. doi:10.1016/j.neulet.2006.07.055
Harris, S.E., Martin-Ruiz, C., von Zglinicki, T., Starr, J.M. & Deary, I.J. (2010b). Telomere length and aging biomarkers in 70-year-olds: the Lothian Birth Cohort. Neurobiological Aging, 33, 7, 1486.e3-1486.e8.
Hartmann, N., Boehner, M., Groenen, F. & Kalb, R. (2010). Telomere length of patients with major depression is shortened but independent from therapy and severity of the disease. Depression and Anxiety, 27, 12, 1111-1116.
Hawco, C., Berlim, M.T. & Lepage, M. (2013). The Dorsolateral Prefrontal Cortex Plays a Role in Self-Initiated Elaborative Cognitive Processing during Episodic Memory Encoding: rTMS Evidence. PloS one, 8(9), e73789. doi:10.1371/journal.pone.0073789
Hoen, P.W., de Jonge, P., Na, B.Y., Farzaneh-far, R., Epel, E., Lin, J., Blackburn, E. & Whooley, M.A. (2012). Depression and leukocyte telomere length in patients with coronary heart disease: data from The Heart and Soul study. Psychosomatic Medicine, 73(7), 541–547. doi:10.1097/PSY.0b013e31821b1f6e.
Kao, Y.-C., Davis, E.S. & Gabrieli, J.D.E. (2005). Neural correlates of actual and predicted memory formation. Nature Neuroscience, 8(12), 1776–83. doi:10.1038/nn1595
Kljajevic, V. (2011). From cell to cognition: can changes in telomere length indicate patterns of cognitive aging? Clinical Science, 121, 7, 313-314.
20 Krauel, K., Duzel, E., Hinrichs, H., Santel, S., Rellum, T. & Baving, L. (2007). Impact of emotional salience on episodic memory in attention-deficit/hyperactivity disorder: a functional magnetic resonance imaging study. Biological Psychiatry, 61(12), 1370–9. doi:10.1016/j.biopsych.2006.08.051
Lin, J., Epel, E., Cheon, J., Kroenke, C., Sinclair, E., Bigos, M., Wolkowitz, O., Mellon, S. & Blackburn, E. (2010). Analyses and comparisons of telomerase activity and telomere length in human T and B cells: Insights for epidemiology of telomere maintenance. Journal of Immunological Methods, 352, 1-2, 71-80.
Lung, F.-W., Chen, N.C. & Shu, B.-C. (2007). Genetic pathway of major depressive disorder in shortening telomere length. Psychiatric Genetics, 17, 3, 195-199.
Masutomi, K., Yu, E.Y., Khurts, S., Ben-Porath, I., Currier, J.L., Metz, G.B., Brooks, M.W., Kaneko, S., Murakami, S., DeCaprio, J.A., Weinberg, R.A., Stewart, S.A. & Hahn, W.C. (2003). Telomerase maintains telomere structure in normal human cells. Cell, 114, 2, 241-253.
Mather, K.A., Jorm, A.F., Anstey, K.J., Milburn, P.J., Easteal, S. & Christensen, H. (2010). Cognitive performance and leukocyte telomere length in two narrow age-range cohorts: a population study. BMC Geriatrics, 10, 62. doi:10.1186/1471-2318-10-62
McDermott, L.M. & Ebmeijer, K.P. (2009). A meta-analysis of depression severity and cognitive function. Journal of Affective Disorders, 119, 1-3, 1-8.
Miller, E.K. & Cohen, J.D. (2001). An integrative theory of prefrontal cortex function. Annual Review of Neuroscience, 24, 167–202. doi:10.1146/annurev.neuro.24.1.167
Morcom, A.M., Good, C.D., Frackowiak, R.S.J. & Rugg, M.D. (2003). Age effects on the neural correlates of successful encoding. Brain, 126, 1, 213-229.
Müller, V.I., Cieslik, E.C., Laird, A.R., Fox, P.T. & Eickhoff, S.B. (2013). Dysregulated left inferior parietal activity in schizophrenia and depression: functional connectivity and characterization. Frontiers in Human Neuroscience, 7, 268. doi:10.3389/fnhum.2013.00268
Nielsen, F.A. & Hansen, L.K. (2004). Finding related functional neuroimaging volumes. Artificial Intelligende in Medicine, 30, 2, 141-151.
O’Donovan, A., Tomiyama, A.J., Lin, J., Puterman, E., Adler, N.E., Kemeny, M., Wolkowitz, O.M., Blackburn, E.H. & Epel, E.S. (2012). Stress appraisals and cellular aging: a key role for anticipatory threat in the relationship between psychological stress and telomere length. Brain, Behavior, and Immunity, 26(4), 573–9. doi:10.1016/j.bbi.2012.01.007
Olson, I.R. & Berryhill, M. (2009). Some surprising findings on the involvement of the parietal lobe in human memory. Neurobiology of Learning and Memory, 91(2), 155– 65. doi:10.1016/j.nlm.2008.09.006
21 O’Sullivan, R.J. & Karlseder, J. (2010). Telomeres: protecting chromosomes against genome instability. Nature Reviews Molecular Cell Biology, 11, 171-181.
Park, H., Kennedy, K.M., Rodrigue, K.M., Hebrank, A. & Park, D.C. (2013). An fMRI study of episodic encoding across the lifespan: changes in subsequent memory effects are evident by middle-age. Neuropsychologia, 51(3), 448–56. doi:10.1016/j.neuropsychologia.2012.11.025
Park, H. & Rugg, M.D. (2009). Neural correlates of successful encoding of semantically- and phonologically – mediated inter-item associations. NIH Public Access Neuroimage, 43(1), 165–172. doi:10.1016/j.neuroimage.2008.06.044.Neural
Penninx, B.W.J.H., Beekman, A.T.F., Smit, J.H., Zitman, F.G., Nolen, W.A., Spinhoven, P. & Cuijpers, P.I.M. (2008). The Netherlands Study of Depression and Anxiety (NESDA): rationale, objectives and methods. International Journal of Methods in Psychiatric Research, 17(3), 121–140. doi:10.1002/mpr
Ptak, R. (2012). The frontoparietal attention network of the human brain: action, saliency, and a priority map of the environment. The Neuroscientist, 18, 5, 502-515.
Salthouse, T.A. (2009). When does age-related cognitive decline begin? Neurobiology of Aging, 30, 4, 507-514.
Sambataro, F., Murty, V.P., Callicott, J.H., Tan, H.-Y., Das, S., Weinberger, D.R. & Mattay, V.S. (2010). Age-related alterations in default mode network: impact on working memory performance. Neurobiology of Aging, 31(5), 839–52. doi:10.1016/j.neurobiolaging.2008.05.022
Savolainen, K., Raikkonen, K. & Kananen, L. (2012). History of mental disorders and leukocyte telomere length in late adulthood: the Helsinki Birth Cohort Study. Journal of Psychiatric Research, 46, 10, 1346-1353.
Schmahmann, J. D. & Pandya, D.N. (2006). Fiber Pathways of the Brain. Oxford
(UK): Oxford University Press. 81–87.
doi:10.1093/acprof:oso/9780195104233.001.0001
Shomstein, S. (2012). Cognitive functions of the posterior parietal cortex: top-down and bottom-up attentional control. Frontiers in Integrative Neuroscience, 6, 38. doi:10.3389/fnint.2012.00038
Sibille, K.T., Langaee, T., Burkley, B., Gong, Y., Glover, T.L., King, C., Riley, J.L., Leeuwenburgh, C., Staud, R., Bradley, L.A. & Fillingim, R.B. (2012). Chronic pain, perceived stress, and cellular aging: an exploratory study. Molecular Pain, 8(1), 12. doi:10.1186/1744-8069-8-12
Singh-Curry, V. & Husain, M. (2009). The functional role of the inferior parietal lobe in the dorsal and ventral stream dichotomy. Neuropsychologia, 47(6), 1434–48. doi:10.1016/j.neuropsychologia.2008.11.033
22 Spaniol, J., Davidson, P.S.R., Kim, A.S.N., Han, H., Moscovitch, M. & Grady, C.L. (2009). Event-related fMRI studies of episodic encoding and retrieval: meta-analyses using activation likelihood estimation. Neuropsychologia, 47(8-9), 1765–79. doi:10.1016/j.neuropsychologia.2009.02.028
Thomas, P., O’Callaghan, N.J. & Fenech, M. (2008). Telomere length in white blood cells, buccal cells and brain tissue and its variation with ageing and Alzheimer’s disease. Mechanisms of Ageing and Development, 129, 183-190.
Uncapher, M.R. & Rugg, M.D. (2005). Effects of divided attention on fMRI correlates of memory encoding. Journal of Cognitive Neuroscience, 17(12), 1923–35. doi:10.1162/089892905775008616
Valdes, A.M., Andrew, T., Gardner, J.P., Kimura, M. Oelsner, E., Cherkas, L.F., Aviv, A. & Spector, T.D. (2005). Obesity, cigarette smoking, and telomere length in women. The Lancet, 366, 9486, 662-664.
Valdes, A.M., Deary, I.J., Gardner, J., Kimura, M., Lu, X., Spector, T.D., Aviv, A. & Cherkas, L.F. (2010). Leukocyte telomere length is associated with cognitive performance in women. Neurobiology of Aging, 31, 6, 986-992.
Vannini, P., Hedden, T., Sullivan, C. & Sperling, R.A. (2013). Differential functional response in the posteromedial cortices and hippocampus to stimulus repetition during successful memory encoding. Human Brain Mapping, 34(7), 1568–78. doi:10.1002/hbm.22011
Van Tol, M.J., Van der Wee, N.J.A., Demenescu, L.R., Nielen, M.M.A., Aleman, A., Renken, R. & Van Buchem, M.A. (2011). Functional MRI correlates of visuospatial planning in out-patient depression and anxiety. Acta Psychiatrica Scandinavica, 124(4), 273–84. doi:10.1111/j.1600-0447.2011.01702.x
Van Tol, M.-J., Demenescu, L.R., Van der Wee, N.J.A., Kortekaas, R., Nielen, M.M.A., den Boer, J.A., Renken, R.J., van Buchem, M.A., Zitman, F.G., Aleman, A. & Veltman, D. (2012). Functional magnetic resonance imaging correlates of emotional word encoding and recognition in depression and anxiety disorders. Biological Psychiatry, 71(7), 593– 602. doi:10.1016/j.biopsych.2011.11.016
Verhoeven, J.E., Révész, D., Epel, E.S., Lin, J., Wolkowitz, O.M. & Penninx, B.W.J.H. (in press). Major depressive disorder and accelerated cellular aging: results from a large psychiatric cohort study. Journal of Molecular Psychiatry.
Vilberg, K.L. & Rugg, M.D. (2008). Memory retrieval and the parietal cortex: a review of evidence from a dual-process perspective. Neuropsychologia, 46(7), 1787–99. doi:10.1016/j.neuropsychologia.2008.01.004
Whitfield-Gabrieli, S., & Ford, J.M. (2012). Default mode network activity and connectivity in psychopathology. Annual Review of Clinical Psychology, 8, 49–76. doi:10.1146/annurev-clinpsy-032511-143049
23 Wikgren, M. (2011). Telomeres and the brain: an investigation into the relationships of leukocyte telomere length with functional and structural attributes of the brain (Doctoral dissertation, Umeå University).
Wikgren, M., Maripuu, M., Karlsson, T., Nordfjäll, K., Bergdahl, J., Hultdin, J., Del-Favero, J., Roos, G., Nilsson, L.-G., Adolfsson, R. & Norrback, K.-F. (2012a). Short telomeres in depression and the general population are associated with a hypocortisolemic state. Biological Psychiatry, 71(4), 294–300. doi:10.1016/j.biopsych.2011.09.015
Wikgren, M., Karlsson, T., Nilbrink, T., Nordfjäll, K., Hultdin, J., Sleegers, K., Van Broeckhoven, C., Nyberg, L., Roos, G., Nilsson, L.-G., Adolfsson, R. & Norrback, K.-F. (2012b). APOE ε4 is associated with longer telomeres, and longer telomeres among ε4 carriers predicts worse episodic memory. Neurobiology of Aging, 33(2), 335–344. doi:10.1016/j.neurobiolaging.2010.03.004
Wikgren, M., Karlsson, T., Lind, J., Nilbrink, T., Hultdin, J., Sleegers, K., van Broeckhoven, C., Roos, G., Nilsson, L.-G., Nyberg, L., Adolfsson, R. & Norrback, K.-F. (2012c). Longer leukocyte telomere length is associated with smaller hippocampal volume among non demented APOE ε3/ε3 subjects. PLoSone, 7, 4, e34292.
Wolkowitz, O.M., Epel, E.S., Reus, V.I. & Mellon, S.H. (2010). Depression gets old fast: do stress and depression accelerate cell aging? Depression and Anxiety, 27(4), 327– 38. doi:10.1002/da.20686
Wolkowitz, O.M., Mellon, S.H., Epel, E.S., Lin, J., Dhabhar, F.S., Su, Y., Reus, V.I., Rosser, R., Burke, H.M., Kupferman, E., Compagnone, M., Nelson, J.C. & Blackburn, E. (2011). Leukocyte telomere length in major depression: correlations with chronicity, inflammation and oxidative stress – preliminary findings. PLoSone, 6, 3, e17837.
Worsley, K.J., Marrett, S., Neelin, P., Vandal, A.C., Friston, K.J. & Evans, A.C. (1996). A unified statistical approach for determining significant signals in images of cerebral activation. Human Brain Mapping, 4, 58-73.
Woudstra, S., Bochdanovits, Z., Van Tol, M.-J., Veltman, D.J., Zitman, F.G., Van Buchem, M.A. & Van der Wee, N.J. (2013). Piccolo genotype modulates neural correlates of emotion processing but not executive functioning. Translational Psychiatry, 2(4), e99. doi:10.1038/tp.2012.29
Yaffe, K., Lindquist, K., Kluse, M., Cawthon, R., Harris, T., Hsueh, W.-C., Simonsick, E.M., Kuller, L., Li, L., Ayonayon, H.N., Rubin, S.M. & Cummings, S.R. (2011). Telomere length and cognitive function in community-dwelling elders: Findings from the Health
ABC Study. Neurobiology of Aging, 32(11), 2055–2060.
