The results of this study demonstrate that relative to controls, men with an FMR1
premutation allele had smaller amygdala volumes and reduced amygdala activation during an emotion-matching task. In contrast to our prior study (18
), in this investigation we were able to determine that aberrant amygdala function was more strongly associated with reduced FMRP (the cause of FXS) than abnormal elevation of FMR1
mRNA (the hypothesized toxic mRNA gain-of-function model underlying FXTAS). To our knowledge this the first evidence of the impact of reduced FMRP without methylation in FMR1
premutation carriers. Standardized clinical assessment indicated that men with the premutation had more significant communication and reciprocal social behavior symptoms of autism spectrum disorder compared to controls; however, on a group level the severity of these symptoms did not reach clinical significance. Exploratory analyses showed that the FMRP-mediated blunted amygdala response may be correlated with deficits in social information processing, a component of the broader autism phenotype. Since no evidence of transcriptional silencing was observed in any individual with a premutation included in this study, who do show levels of FMR1
mRNA substantially higher than normal levels (24
), the mild reductions in FMRP with increasing CGG repeat length (45
) are due primarily to reduced efficiency of translational initiation (42
). Although the full range of FMRP levels in individuals with the full mutation, premutation and general population controls has not been fully elucidated, it is important to emphasize that the relatively lower FMRP levels observed in some premutation carriers in this study are likely to be quite modest compared to most males with the full mutation and FXS who express essentially no or very little FMRP and who demonstrate more substantial behavioral and cognitive deficits.
Overall, the study indicates that neurodevelopmental and psychiatric symptoms previously reported among younger men and boys with the premutation may be due to the same molecular genetic mechanism leading to FXS, but to a milder degree. Previous findings, including our own (16
) highlighting the potential negative impact of elevated mRNA on these symptoms suggests that both mRNA- and protein-level effects may be operating in the premutation range, with protein decrements perhaps playing a larger role, at least in amygdala-mediated functions.
In the current cohort of participants, we found reduced amygdala volumes in premutation carriers, whereas in our prior study of a different group of participants (18
), we did not find this effect. A number of factors might explain this difference. For example, the more powerful 3T scanner used here provided higher signal-to-noise ratio, enabling higher levels of image quality, speed and resolution, perhaps making the group effect easier to detect. Also, somewhat larger sample sizes in the present study provided additional data and greater power to detect differences.
Prior fMRI studies in FXS using social stimuli (human faces showing direct or averted gaze) (49
) have documented greater
sensitization in the left amygdala with successive exposure to direct gaze in a sample of adolescent boys. Given that patients with FXS and some individuals with the premutation have reduced FMRP (albeit to a different degree), we might expect premutation carriers to have a similarly increased amygdala response. The explanation for this inconsistency is unclear but could be related to the nature of the tasks, the added complexity of mildly reduced FMRP and elevated message in premutation carriers, or due to developmental differences. Nevertheless, this research highlights the critical role of amygdala function in perception of and response to human social stimuli and in social behavior in individuals affected by FMR1
In our previous report, we suggested that the altered limbic system function observed in men with the premutation could reflect early prodromal brain changes associated with later-onset FXTAS. However, given that the primary molecular genetic mechanism underlying FXTAS neuropathology is thought to involve abnormal elevation of the expanded-CGG-repeat FMR1 mRNA, our current results suggest that lowered FMRP contributes to amygdala dysfunction, and that the latter effect could be both neurodevelopmental in origin and independent of FXTAS progression.
The reduced amygdala activation in premutation carriers observed here cannot be explained by decreased visual fixation on stimuli, as we observed similar fixation patterns between groups. Recruitment biases are unlikely to have affected our findings because participants were selected at random from pedigrees and were not clinic referred. The blunted amygdala response cannot be explained by reduced amygdala volume as the ROI method controls for individual differences in amygdala size and, additionally, we did not find a significant correlation between volume and activation (data not shown). Interestingly, although differences in brain activation were observed during the emotion-matching task, we found no difference in performance, indicating that brain activation patterns may be more sensitive to effects of the premutation than observable behavior.
There were important limitations to the study. Although we took several steps to minimize recruitment bias, this form of bias cannot be completely eliminated because individuals make decisions whether to participate based on many factors, some of which may be associated with effects of the premutation being studied here. Although men with the premutation were rated as having higher rates of autism symptoms on clinical exam, it is not possible to rule out effects of experimenter bias. Many participants were assessed blind to group; however, blinding was not possible in all cases due to personal information revealed during the interview portions of the ADOS. Approximately one-third of participants in each group were taking psychoactive medication which may have influenced our results, and somewhat detracts from the generalizability of the findings. It is important to emphasize that the FMR1 molecular genetic measures were ascertained from blood samples and may not necessarily reflect what would be found in brain tissue. Finally, we were unable to derive valid FMRP values for all participants in the study; as such the results pertaining to the association between FMRP level and brain function must be considered preliminary.
Our findings may have treatment implications for some individuals with the premutation who present with significant symptoms of FXS or ASD. The “mGluR theory” of FXS states that mental impairment and phenotypic behaviors associated with FXS arise, at least in part, from constitutive activation of translational pathways normally controlled by group 1 metabotropic glutamate receptor (mGluR1 and mGluR5) activity (50
). Accordingly, down regulation of mGluR5 activity, both pharmacologically and genetically, has rescued many phenotypes examined in the animal models of FXS (51
). Phase II clinical trials of mGluR5 antagonists in patients with FXS are now ongoing, and it is intriguing to consider whether premutation carriers with significantly lowered FMRP would benefit from these treatments as well.