Protective effect of APOE epsilon 2 on intrinsic functional connectivity of the entorhinal cortex is associated with better episodic memory in elderly individuals with risk factors for Alzheimer's disease

The apolipoprotein E (APOE) ε4 allele associates with accelerating the conversion from amnestic mild cognitive impairment (aMCI) to Alzheimer's disease (AD), whereas the protectiveAPOEε2 allele appears to be against the disease. Moreover, entorhinal cortex (ERC) is one of the earliest brain regions of AD pathology that disrupts the formation of episodic memory. To investigate the effects of APOE ε2 and ε4alleles on functional connectivity (FC) of ERC and cognition in aMCI. Methods The FC analyses of ERC were performed in 83 aMCI (9 ε2-carrier, 44 ε3ε3, and 30 ε4-carrier) and 88 healthy controls (HC, 15 ε2-carrier, 40 ε3ε3, and 33 ε4-carrier). Multiple linear regression model was performed between the altered ERC connectivities and cognition. In the ERC network, aMCI with ε4-carriers showed decreased FC in the bilateral middle temporal gyrus (MTG), right precuneus, and right precentral gyrus (PreCG), while ε2-carriers showed increased FC in these regions (except the right PreCG) compared to HC. The altered FC between ERC and right MTG correlated with episodic memory performance in aMCI carried ε2 and ε4 alleles. These results suggest that the effects ofAPOEon the ERC network are closely linked to the role of this gene on AD risk, which aMCI with ε4-carriers can accelerate the pathological progression of network-based mechanisms while ε2-carriers may play a protective role in contributing to a compensatory mechanism. It further suggests that APOE can appear to directly affect the ERC-MTG neural pathway associated with the impairment of episodic memory in aMCI.

electrophoresis to detect the alleles of rs429358 and rs7412. As a result, the APOE genotype was determined by the haplotype of rs429358 and rs7412. APOE ε2 allele was recognized by rs429358-T and rs7412-T, APOE ε3 allele was identified by rs429358-T and rs7412-C, and APOE ε4 allele was defined by rs429358-C and rs7412-C.

Assessment of susceptibility artifacts
Previous studies have indicated that brain areas in the MTL directly above the petrous bone especially tend to signal loss [4]. To assess the effects of susceptibility artifacts in our data, the signal-to-noise ratio (SNR) was computed for each voxel by averaging the signal intensity across all the target-atlas normalized BOLD runs and dividing it by the standard deviation (SD) over time [5]. To avoid variability in EPI timeseries cause of susceptibility artifact, the ERC seed was thresholded to exclude voxels with mean signal in the intensity-normalized EPI timeseries below 3000, corresponding approximately to a SNR of 20 [6]. Thresholds resulted in the rejection of no more than 5% of voxels in the ROI. Then the thresholded ROI was further used to perform FC analyses.

Gray matter loss effect
To avoid the interpretation on the differences of FC from the anatomical atrophies in the patients, voxelwise GM volumes were addressed as covariates in the further FC analysis. Voxel-based morphometry (VBM) analysis was performed using VBM8 toolbox in SPM8 (VBM8 toolbox, http://dbm.neuro.uni-jena.de/vbm). Briefly, the individual T1-weighted images were segmented into GM, WM and CSF, and then normalized to the MNI space. The normalized GM volume maps (modulated images) were resampled to the same grid as the functional image. Finally, the voxelwise GM values were regressed out as the nuisance regressor from the FC values to control the influence of GM volume on the FC strength. The voxewise GM volume correction was performed for each subject. Then, a two-sample t-test was performed to determine whether the GM was atrophied in aMCI, controlling for age, gender, and years of education.

Head motion effects
To minimize the influence of head motion both at the individual and at the group levels, three approaches were employed in QA measures. First, the head motion effects were regressed out, which were calculated as the root mean squared (rms) head displacement or rotation (in mm or °) derived from the motion-correction procedure [7]. Second, a 'scrubbing' procedure was carried out to scrub frames (volumes) with excessively high whole-brain rms signal change over time in the preprocessed rs-fcMRI data for each individual [8][9][10]. These frames were subsequently removed from rs-fcMRI analysis. The fraction of frames so removed was < 5 % in each group (no significant effect of group as factor on fraction of frames removed). Overall, 2 aMCI and 4 HC had a large proportion of high-noise frames (> 20% frames identified as contaminated) and were therefore excluded from the analysis. Third, additional QA measures included rms head displacement or rotation (in mm or °) and the voxel-wise time series SD averaged over the whole brain [11]. We referred to a prior study [12] to empirically determine the exclusion criteria for QA measure with the objective of achieving QA parameter distribution equivalence between groups while maximizing the number of included subjects. Overall, 4 HC subjects with a mean preprocessed rs-fcMRI signal 2.5% SD (after nuisance regression) or rms movement or rotation exceeding 2.0 mm or 2.0° and mean frame-to-frame rms movement or rotation more than 0.5 mm or 0.1° were also excluded. No significant differences between groups were observed in QA parameters (p > 0.05).

Group-level intrinsic connectivity analysis
A repeated-measure ANOVA with age, gender, years of education, and GM volumes treated as covariates was used to test the difference of FC patterns for APOE genotypes in HC subjects. The repeated-measure ANOVA have one between-subjects factor (APOE genotypes) and one within-subjects factor (target brain regions). A post-hoc Student's t-test with FDR correction for each two pairs between APOE genotypes was carried out to further investigate differences as any statistical significance for ANOVA. p < 0.05 was required for statistical significance.

The relationships between the altered ERC pattern and neuropsychological performance
To increase statistical power by reducing random variability, this study composited the neuropsychological tests into 4 cognitive domains and transformed the raw scores into 4 composite Z scores, as previously described [1,2,13]. First, for each neuropsychological test, the individual raw scores were transformed to Z scores, according to the mean and  Figure S1: Resting-state functional connectivity patterns within APOE genotypes and between APOE genotypes maps in the entorhinal cortex network in HC (A) and aMCI (B). The first and second columns show the within-APOE genotypes statistical maps for APOE2 carriers, APOE3, and APOE4, respectively, with statistical threshold set at p corrected < 0.01, corrected by FWE. Last column shows the between-APOE genotypes statistical maps, with statistical threshold set at p corrected < 0.05, corrected by false discovery rate (FDR). Warm and blue colors indicate decreased and increased functional connectivity in aMCI subjects compared to HC subjects. Color bar is presented with T score.