Genetic and Environmental Influences on Self-Reported Cognitive Functioning: Associations of Diverse Measures of Stress across the FMR1 CGG Repeat Range

Background : The FMR1 gene is essential for neural development and healthy synaptic function. The modal number of CGG repeats in FMR1 is 30, but the range is large with the reported copy number extending down to as few as 6 CGGs and up to over 200 CGGs, conferring fragile X syndrome. Prior work suggests that behavioral phenotypes, including cognitive function, may vary along the continuum of the FMR1 CGG repeat range. Stress may negatively influence cognitive function; however, it is not known whether FMR1 -related variability (i.e., CGG repeat length), in addition to stress, independently influences cognitive function across the CGG range. Methods : Participants included 1275 mothers who had CGGs ranging from 18 to 123 repeats. Participants completed self-report measures of executive function (BRIEF-A), memory, subjective stress (i.e., perceived stress), and objective stress (i.e., number of life events, parenting a child with a disability). Stress and FMR1 -related variability (i.e., CGG repeat length) were examined as predictors of self-reported executive function and memory difficulty. Results : Each measure of stress (i.e., perceived stress, life events, and parenting a child with a disability) significantly predicted greater self-reported difficulties in executive function and the likelihood of memory problems, net of age and level of education. Additionally, above and beyond stress effects, CGG repeat number significantly predicted executive functioning and memory difficulties. There was a linear association of CGG repeat number with executive functioning limitations. The association of CGGs with memory difficulties was curvilinear, with participants in the premutation range having the greatest likelihood of reporting such difficulties. Conclusions : These findings suggest that CGG repeat length confers independent contributions to self-reported executive function difficulty and memory problems over and above indices of stress, suggesting additive effects of genetic variation and environmental exposure.


Background
The fragile X mental retardation 1 (FMR1) gene is essential for healthy synaptic function and neural development, as well as the regulation of mRNA throughout the genome (1). Located on the X chromosome, an expansion of more than 200 cytosine-guanine-guanine (CGG) trinucleotide repeats in the 5' untranslated region of the FMR1 mRNA causes fragile X syndrome (FXS). Mothers of children with FXS are often carriers of the FMR1 premutation (PM; i.e., , which is relatively common in the general population (~ 1 in 150-209 females; ~1 in 430-470 males (2,3).
Prior work exploring genotype-phenotype associations has largely focused on individuals who carry CGG expansions (i.e., full mutations, premutations, intermediate expansions (4)(5)(6)(7)(8). However, the CGG repeat is highly polymorphic, ranging from as few as 6 repeats to over 200 (9)(10)(11)(12)(13). A small body of literature highlights the importance of evaluating the possibility of involvement of CGG repeats continuously across this range with respect to both biological (9,14,15) and behavioral phenotypes (16,17). Thus, consideration of the FMR1 CGG repeat number continuously across the range below FXS has the potential to elucidate phenotypic variation at the population-level. Of additional importance is the examination of environmental circumstances that may influence phenotypic variation. Prior work suggests that stress, such as parenting a child with a disability, may uniquely influence health outcomes across the CGG range (16). The present study aimed to characterize environmental and genetic predictors of self-reported cognitive function, specifically executive function and memory, across the CGG repeat range.

Cognitive Function and FMR1-related Variability
Executive function and memory encompass higher-order mental processes required for goal-directed behavior (e.g., planning, inhibition, performance monitoring) and retrieval (18). Although difficulties in these domains have been implicated as part of the PM phenotype (5,6,19), there remains some controversy regarding the extent to which variation in phenotypes may be due to CGG repeat expansions, or may be explained by ascertainment bias, namely the inclusion in research of PM carriers who also have children with FXS (20)(21)(22). To better tease apart these effects, one fruitful approach has been to examine the FMR1 CGG range (normal through premutation expansions) in relationship to variation in executive function and memory. In a large sample of both males and females under 50 years of age who had normal (≤ 40 repeats), intermediate (41-60 repeats), and premutation (61-~180) CGG repeats, Hunter and colleagues (22) examined whether FMR1 CGG repeat length predicted factor scores reflecting aspects of executive function and memory. In the full female sample (i.e., normal through premutation repeats), they found that greater CGG repeat length was predictive of higher factor scores indicative of self-reported inattention and impulsivity, as well as direct-assessment processing speed, but not other factors (e.g., memory, response fluency). Though these findings were interpreted as marginal after correcting for multiple comparisons, this work set the stage for continued exploration of executive function variability along the CGG repeat continuum.
Likewise, prior work from Debrey and colleagues (23) identified a subgroup of individuals with intermediate CGG repeats who demonstrated clinically significant memory problems. Building on this prior work, the present study examines cognitive functioning (i.e., self-reported executive function and memory) in individuals across the CGG repeat range (excluding the full mutation).

Stress And Cognitive Function
Heightened levels of stress, both subjective and objective, have been reported in observational studies to be associated with executive dysfunction across multiple populations, including individuals with disabilities, individuals exposed to trauma, older adults, and healthy adult controls (18,(24)(25)(26)(27)(28)(29).
Greater perceived stress, a measure of subjective stress in response to external stressors, has been found to be associated with poorer cognitive functioning including attentional control, processing speed, and working memory (24,26). Likewise, objective indices of stress (e.g., life events, parenting a child with a disability) may prompt similar aberrations in executive function across diverse populations (28,29). These objective stressors have also been found to be associated with cognitive dysfunction (including poorer episodic memory) (30), decreased well-being (31), poor mental health (32,33), as well as changes to biological mechanisms, such as dampened cortisol-awakeningresponses (34)(35)(36), and alterations to the neural structures that underlie stress-related responses (25,26,37).
Parenting stress, or adverse psychological responses to parenting obligations (38), may be observed at increased rates in parents of children with disabilities due to unique and chronic caregiving demands. Meta-analyses (39,40) suggest that parents of a child with a developmental disability experience higher rates of parenting stress than do parents of typically developing children. Parenting stress may also be observed at increased rates in parents of children with other health conditions. For instance, having a child with a chronic physical (e.g., epilepsy) or mental health (e.g., bipolar disorder) condition has been associated with higher parenting stress compared to the parenting stress of raising typically developing children (41)(42)(43). Parents of adult children with developmental or mental health conditions have been exposed to this unique stressor for many years (32) and have been shown to have particularly high levels of parenting stress.
Past research has suggested that mothers may be more negatively affected by parenting stress than fathers (44)(45)(46). Therefore, the present study focused on mothers and examined the effects of multiple indices of stress on executive function and memory in mothers across the CGG repeat range (below the full mutation). In addition to stress, it is possible that other factors may contribute to variability in executive function and memory, namely age and education (47)(48)(49). These individual factors are therefore incorporated into the study as covariates to account for sociodemographic features that may influence variation in cognitive function. We hypothesized that subjective and objective stressors, as well as increased FMR1 CGG repeat number, would each independently predict executive function and memory difficulty, net of age and education.

Participants and Procedures
Participants included 1275 mothers with CGG repeats ranging from 18 to 123.
The majority of these participants (n = 1152) were drawn from the Marshfield Clinic Personalized Medicine Research Project (PMRP) (50), a 20,000-person population-based biobank. Over half of the PMRP members were female (n = 11,556) and DNA was available for 99.7% of them. For a previous investigation (51), the DNA samples of all 20,000 PMRP members were screened for FMR1 CGG repeats, which formed the basis for the current study's sampling plan. We invited all PMRP females who had CGG expansions (41 or greater CGGs) as well as all of those who had at least one allele below the normal range (defined here as 25 or fewer CGGs) to participate in the present study.
Additionally, based on a power analysis, a random sample of females with normal-range CGGs was selected for inclusion in the present research. Thus, by design, the recruited sample included all females in the population biobank who had expanded or low numbers of CGGs but fewer females in the normal range than in the full population biobank. The response rate of the recruited females was 77.4%. We further restricted the current analysis to data obtained from mothers who had at least one biological or adopted child.
The CGGs of the participants from PMRP ranged from 18-100 repeats, although per IRB they were not aware of their CGG repeat number. To extend the range of FMR1 CGG repeat length, clinicallyascertained mothers of children diagnosed with FXS were included in the present analysis (n = 123, with 67-123 CGGs). Participants from the clinically-ascertained samples were recruited from fragile X clinics, via local media, newsletters, brochures, and disability registries (52,53). All participants (PMRP and clinically-ascertained) completed a questionnaire that provided information on whether they had a child with a developmental or mental health condition (see Table 1), as well as all other non-genetic measures for the current study.

Measures
Stress.
Perceived Stress Scale. The Perceived Stress Scale (PSS) (54) is a 10-item, self-report measure that quantifies an individual's appraisal of stressful experiences from the past month. Examples include "In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?" and "In the last month how often did you feel nervous or stressed?" Each item was scored on a scale of 0 (never) -4 (very often); four positively-stated items were reverse-coded. The total score represents a sum across items; higher numbers indicate a greater degree of perceived stress (i.e., subjective stress). Age-and gender-based norms were previously established in prior work  Table 2 for definitions). Each item was rated from 1 (never) to 3 (often). Raw scores for each domain were converted into t-scores, with higher scores suggestive of greater executive difficulties in daily life. T-scores that exceeded 65 on any domain indicated clinically-significant executive dysfunction in that area. In order to ensure that respondents did not indicate excessively negative self-perception about their own executive function, the Negativity scale was examined to ensure that no participant met or exceeded a total score of six (57). Table 2 Standard definitions of the BRIEF-A domains (57,58,72).

Domain Definition Inhibit
The Inhibit domain assesses inhibitory control and impulsivity. This can be described as the ability to resist impulses and the ability to stop one's own behavior at the appropriate time.

Shift
The Shift domain assesses the ability to move with ease from one situation, activity, or aspect of a problem to another as the circumstances demand. Key aspects of shifting include the ability to (a) make transitions; (b) tolerate change; (c) problemsolve flexibly; (d) switch or alternate attention; and (e) change focus from one mindset or topic to another.

Emotional Control
The Emotional Control domain measures the impact of executive function problems on emotional expression and assesses an individual's ability to modulate or control his or her emotional responses.

Self-Monitor
The Self-Monitor domain assesses aspects of social or interpersonal awareness. It captures the degree to which an individual perceives himself as aware of the effect that his or her behavior has on others. Initiate The Initiate domain reflects an individual's ability to begin a task or activity and to independently generate ideas, responses, or problem-solving strategies.

Working Memory
The Working Memory domain measures "on-line representational memory;" that is, the capacity to hold information in mind for the purpose of completing a task, encoding information, or generating goals, plans, and sequential steps to achieving goals. Working memory is essential to carry out multistep activities, complete mental manipulations such as mental arithmetic, and follow complex instructions. Plan/Organize The Plan/Organize domain measures an individual's ability to manage current and future-oriented task demands. The domain consists of two components: plan and organize. The Plan component captures the ability to anticipate future events, to set goals, and to develop appropriate sequential steps ahead of time in order to carry out a task or activity. The Organize component refers to the ability to bring order to information and to appreciate main ideas or key concepts when learning or communicating information.

Task Monitor
The Task Monitor domain reflects the ability to keep track of one's problem-solving success or failure, and to identify and correct mistakes during behaviors.

Organization of Materials
The Organization of Materials domain measures orderliness of work, living, and storage spaces (e.g., desks, rooms).
The BRIEF-A was previously standardized on a representative population sample of 1136 adults (age 18-90) with Cronbach α coefficients ranging from .93-.96 and test-retest reliability ranging from .93-.94 across domains, with utility demonstrated in both clinical and non-clinical samples (57)(58)(59)(60).
The BRIEF-A has been shown to correlate significantly with direct-assessment measures of executive function (e.g., go/no go and trail making tests) in healthy adults (61) and in individuals with disorders associated with executive dysfunction (62)(63)(64). The present study used the GEC t-score as the indicator of executive functioning.
A subset of participants in our sample were over 85 years old (n = 42). Given that the BRIEF-A was standardized on participants up to age 90, we checked all findings excluding participants over 85 (n = 30) and over 90 years (n = 12), which did not change results. Thus, the findings reported below include all participants.
Self-reported Memory Problems. Participants answered the question: Do you have problems with memory? This question was rated as 0 (no problems with memory), 1 (undiagnosed problems with memory), or 2 (diagnosed memory problems). Due to skewness in the data (as very few mothers reported diagnosed memory problems), all "2" responses were collapsed to 1.
FMR1 -related variation. DNA samples from all participants were analyzed for CGG repeats in FMR1.
Assays were completed at Kimball Genetics, Inc., the Wisconsin State Laboratory of Hygiene, and the Rush University Medical Center Molecular Diagnostics Laboratory. PM mothers who were mosaic for the full mutation were excluded from analyses given previously reported differences in cognitive function in this subsample of PM carriers (53).

Statistical Analyses
Statistical analyses were performed using IBM SPSS Statistics, version 26 (65). Descriptive statistics and Pearson correlations among all study variables are presented in Table 3. Maternal age and education were controlled in all subsequent analyses. To control for potential effects of the second FMR1 allele (as females carry two X chromosomes), the "shorter" allele (i.e., the allele with the lower number of CGG repeats) was included as a covariate in all regression analyses.

Descriptive Findings
As shown in Table 3 Correlations among study variables are depicted in Table 3. Notably, the stress measures were

Multivariate Findings
The key predictions of the study were that measures of stress would influence executive function and memory function, and that an effect of FMR1 CGG repeat length would be associated with these measures of cognitive function above and beyond indices of stress.
Executive Function. Curvilinear CGG effects were not significant in any model of executive function (pvalues > .589); thus, linear effects are reported below. As presented in Table 4, perceived stress (PSS) significantly predicted variance in the global executive composite (GEC) of the BRIEF-A (b = .866, p < .001), such that higher levels of perceived stress was associated with greater problems in executive functioning. There was an additional significant effect of CGG repeat length (b = .036, p = .011; see

Estimated regression lines indicate the independent influences of CGG repeat length and increased
number of life events on executive function difficulty. Trend lines represent a subset of participants with high life events (> 1 SD above the mean) and low life events (< 1 SD below the mean).
Estimated regression lines indicate the independent influences of CGG repeat length and parenting a child with a disability on executive function difficulty.
Memory Problems. Second, we predicted that stress and FMR1 CGG repeat length would independently predict the likelihood of self-reported memory problems. The curvilinear term was tested and found to be significant in all models reported below. As shown in  Covariates. Both age and education were significant predictors of executive function difficulty across all models (p-values ≤ .022). Age and education were significant predictors of self-reported memory problems in all models that included CGG repeat length (p-values ≤ .034).

Discussion
The present study evaluated the influence of distinct dimensions of stress, and the independent effects of FMR1 CGG repeat length (up to but excluding the full mutation), on self-reported cognitive functioning (i.e., executive function and memory). Importantly, CGG repeat length accounted for small, but statistically significant elevations in executive function difficulty and memory problems, above and beyond stress, age and level of education. To date, this study represents the largest sample in which the association between cognitive function and FMR1 CGG repeat length has been studied. By taking a continuous approach to evaluating FMR1-related effects on cognitive function, and by assessing mothers of non-disabled children as well as children with a diverse range of disability conditions, this study advances understanding of how both environmental and genetic factors influence self-reported cognitive functioning.
Historically, examination of behavioral phenotypes associated with FMR1-related variability (e.g., CGG repeat length) have largely focused on individuals with full mutation fragile X syndrome or the premutation, with some exceptions (16,17,22,66). Many prior assessments of cognition associated with the FMR1 gene involved group comparisons, typically between PM carriers and those with modal numbers of CGG repeats (4,6,7,22). With consideration of the continuous nature of the CGG repeat range, as in the present study, the interpretation of the relationship between FMR1-related variation and phenotypic expression can be advanced.
Our findings revealed linear effects of CGG repeat length on executive functioning and curvilinear effects of CGG repeat length on memory problems. Higher incidence of memory problems was evident at approximately 80 CGG repeats, suggesting that repeats in the premutation range may be driving this effect. The divergent results observed between limitations in executive functioning and memory problems suggest that these constructs may represent distinct aspects of cognition. Higher scores on the BRIEF-A reflect day-to-day difficulties with a broad array of tasks other than memory problems, including difficulties sitting still and waiting, the propensity to make untactful remarks, and the tendency to complete tasks in a hurried manner.
One question that remains is why increasing numbers of CGG repeats across the full range are associated with problems with cognitive functioning. As noted previously, there is evidence to suggest that executive dysfunction is more common among PM carriers and possibly among those in the intermediate range (4,6,17,67,68). However, our findings are in contrast with other research literature suggesting that FMR1 translation may be most efficient at ~ 30 CGGs, with less efficient translation of FMRP at both higher and lower repeat numbers (9,69). Replication of the current findings is necessary, and research examining the basic biological functions of the FMR1 CGG repeat is needed to fully understand these effects.
In this study, we observed associations between multiple measures of stress and cognitive functioning, both objective and subjective, with each measure of stress providing unique insights about how stress is associated with cognitive functioning. For example, the life events that were endorsed from the past year encompassed a variety of events that are not necessarily "negative", such as the birth of a child, increased income, and moving to a new home. Luhmann and colleagues In the present analysis, age and education each significantly contributed to variance in cognitive function in addition to stress and CGG repeat effects, as suggested in prior research (47,48). Prior work suggests that age-related cognitive problems are most pronounced for individuals with lower levels of education (49). The present research suggests that studies of the relationships between variation in the FMR1 gene and behavioral phenotypes should continue to consider additional individual and environmental factors to accurately evaluate FMR1-related influences.

Study Strengths, Limitations, And Future Directions
This study had several notable strengths. First, the availability of DNA and FMR1 CGG repeat assays across the full range of CGG repeats (up to the full mutation) enabled robust examination of the effects of FMR1-related variability on self-reported cognitive function. Second, we had a large sample size, which drew, in large part, from a population-based sample of participants. Third, this study was strengthened by consideration of multiple measures of stress. Additionally, the inclusion of child disabilities was broad, further contributing to the generalizability of study findings.
This study also had some limitations. Although the sample was diverse with regards to age and the range of FMR1-related variation, the participants in the sample were racially and ethnically homogenous. Additionally, many prior reports of associations between cognitive function and FMR1 expansions have included direct-assessment measures, whereas the present study relied on selfreport. The study's large sample size precluded direct testing of > 1200 individuals. It has been suggested that PM carriers may over-report symptoms not evidenced on neurological exam (70).
However, there is also extant literature to suggest significant associations between cognitive function and CGG repeat length using both direct-assessment and self-report measures across the CGG range (22,71), suggesting the validity of self-reported results. Hunter and colleagues (22) have discussed the possibility that individuals who participate in research may be less likely to have cognitive difficulties. Thus, while it is possible that the present findings could reflect potential ascertainment bias in recruited individuals (who are less likely to experience cognitive difficulties), the high response rate from our sample recruitment (77.4%) is indicative of a sample that is largely representative of the population from which participants were drawn.
Another limitation of the present study is that the only FMR1-related biomarker available for the study participants was CGG repeat number. Inclusion of activation ratio, mRNA, and FMRP levels would greatly enhance understanding of the processes investigated here. Additionally, interpretation of these findings can only be extrapolated to females. Given prior work suggestive of age-related cognitive differences in male PM carriers, most obvious in those with motor-related symptoms or signs of FXTAS, it may be that executive function difficulties and memory problems manifest differently in males and females across the CGG range. Future work should examine cognitive function across the CGG range in males, particularly with consideration as to how external stressors may affect profiles, as there may be particular stressors that may be more salient for males than females. This is an important consideration, particularly for male PM carriers, as stress could play an important role in the onset of FXTAS symptoms (72), but further work is needed to investigate this possibility.
Finally, the participants in the present study were recruited using multiple methods, including drawing from a 20,000-person population-based biobank and via a national sample of premutation carriers who were identified clinically after a child was diagnosed with FXS. Although this approach made it possible to include participants with repeats ranging from 18 to 123 CGGs, in future research, it would be advantageous to use a single method of recruitment across diverse samples, but that would require access to much larger population biobanks.

Consent for publication
Consent for publication is not applicable.

Availability of data and materials
Data are available from the corresponding author upon reasonable request.

Competing interests
MM is the Chair of the Scientific Advisory Board of the John Merck Fund.
EBK has received funding from Seaside Therapeutics, Novartis, Roche, Alcobra, Neuren, Cydan, All funding to EBK is directed to Rush University Medical Center to support rare disease programs.
EBK receives no personal funds.
The remaining authors declare they have no competing interests.

Funding
Research reported in this publication was supported by R01 HD082110 (    Curvilinear association between CGG repeat length and the probability of self-reported memory problems. Trend lines represent estimated probability for participants with high PSS (>1 SD above the mean) and low PSS scores (<1 SD below the mean).

Figure 5
Curvilinear association between CGG repeat length and the probability of self-reported memory problems. Trend lines represent estimated probability for participants with high life events (>1 SD above the mean) and low life events (<1 SD below the mean).

Figure 6
Curvilinear association between CGG repeat length and the probability of self-reported memory problems by parenting status.