The known regional abnormality of the dorsolateral prefrontal cortex (DLPFC) and its role in various neural circuits in mild cognitive impairment (MCI) has given prominence to its importance in studies on the disconnection associated with MCI. The purpose of the current study was to examine the DLPFC functional connectivity patterns during rest in MCI patients and the impact of regional grey matter (GM) atrophy on the functional results. Structural and functional MRI data were collected from 14 MCI patients and 14 age, gender-matched healthy controls. We found that both the bilateral DLPFC showed reduced functional connectivity with the inferior parietal lobule (IPL), superior/medial frontal gyrus and sub-cortical regions (e.g., thalamus, putamen) in MCI patients when compared with healthy controls. Moreover, the DLPFC connectivity with the IPL and thalamus significantly correlated with the cognitive performance of patients as measured by mini-mental state examination (MMSE), clock drawing test (CDT), and California verbal learning test (CVLT) scores. When taking GM atrophy as covariates, these results were approximately consistent with those without correction, although there may be a decrease in the statistical power. These results suggest that the DLPFC disconnections may be the substrates of cognitive impairments in MCI patients. In addition, we also found enhanced functional connectivity between the left DLPFC and the right prefrontal cortex in MCI patients. This is consistent with previous findings of MCI-related increased activation during cognitive tasks, and may represent a compensatory mechanism in MCI patients. Together, the present study demonstrated the coexistence of functional disconnection and compensation in MCI patients using DLPFC functional connectivity analysis, and thus might provide insights into biological mechanism of the disease.
Citation: Liang P, Wang Z, Yang Y, Jia X, Li K (2011) Functional Disconnection and Compensation in Mild Cognitive Impairment: Evidence from DLPFC Connectivity Using Resting-State fMRI. PLoS ONE 6(7): e22153. https://doi.org/10.1371/journal.pone.0022153
Editor: Yong He, Beijing Normal University, China
Received: December 10, 2010; Accepted: June 17, 2011; Published: July 21, 2011
Copyright: © 2011 Liang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Open Research Fund of the State Key Laboratory of Cognitive Neuroscience and Learning (No. CNKOPZD1001), China Postdoctoral Science Foundation funded project (No. 201003142), the State Key Program of National Natural Science of China (No. 30930029). and the Fund of Capital Medical University (No. 11JL24). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Alzheimer's disease (AD) is a progressive neurodegenerative disorder, which is characterized by global cognitive decline, including progressive loss of memory, reasoning and language. There has been much anatomical and functional evidence pointing toward AD as a kind of disconnection syndrome –. Some studies have shown disruption of a distributed network, especially the linkages between the hippocampus, the prefrontal cortex and other brain areas , , . Moreover, a number of studies have reported the compensatory mechanism in AD, as indicated by the increased functional connectivity within the prefrontal lobe or between prefrontal regions and other brain regions , , . Mild cognitive impairment (MCI) refers to a transitional stage between normal aging and early AD . MCI has a high probability of evolving toward AD at a rate of 10–15% per year . It is thus of theoretical and practical significance to examine if MCI show similar effects of functional disconnection and compensation with AD.
The study of resting state brain function is especially applicable to the study of patients because of the practical advantages of resting-state fMRI in terms of ease of clinical application. Functional connectivity during rest, measured by correlation in low frequency fluctuations (LFF) (<0.1 Hz) of the blood oxygen level-dependent (BOLD) signal, has been used to characterize intrinsic architecture of large-scale brain systems, including motor , visual , auditory , language , attention  and default mode network (DMN) ,  systems.
Several resting state fMRI studies have demonstrated the abnormalities of functional integrity in MCI patients –. For example, based on posterior cingulate cortex (PCC) functional connectivity analysis, Bai et al.  indicated regions of decreased functional connectivity were identified in aMCI patients, most notably in temporal cortex (BA 21, 37, 39), whilst significantly increased functional connectivity was mainly in frontal cortex (BA 9, 11, 47), as compared to healthy controls. Using independent component analysis (ICA) approach, Qi et al.  compared the default mode network (DMN) activity between MCI patients and healthy elderly, and showed that MCI patients exhibited decreased functional activity in the DMN regions, including the bilateral precuneus/PCC, right inferior parietal lobule, and left fusiform gyrus, while increased activity in the left prefrontal cortex (BA 8) and other parietal, temporal regions, as compared to the healthy elderly. Most of these studies showed that functional disconnection and compensation coexisted in MCI patients, however, they mainly focused on the DMN, i.e., the task-negative network.
The dorsolateral prefrontal cortex (DLPFC), different from DMN regions (e.g., PCC/precuneus, medial prefrontal cortex, etc.), is an important region of task-positive. Numerious evidences have demonstrated that DLPFC plays crucial roles in various attention-demanded cognitive tasks such as working memory and executive function , episodic memory , attention , decision making , reasoning  and problem solving , and so on. Functional and structural studies also indicated that DLPFC is connected to a variety of brain areas, including the thalamus, basal ganglia, the orbitofrontal cortex and primary and secondary association areas of neocortex, including posterior temporal, parietal, and occipital areas , . Although abnormal activities of the DLPFC as well as functional disconnection involving DLPFC have been observed in MCI and AD , –, the precise and detailed connection patterns of the DLPFC in MCI are still unclear. Considering its local dysfunction and roles in various neural circuits relevant to the physiological mechanisms of cognitive impairment, the functional connectivity of the DLPFC in MCI should be especially emphasized. However, to our knowledge so far, no study has examined the DLPFC functional connectivity (DLPFC-FC) in MCI patients.
Using resting state functional connectivity and graph theory analysis, Dosenbach et al.  proposed that there are two commanders in the brain: fronto-parietal and cingulo-opercular circuits. The fronto-parietal circuit mainly comprises of premotor cortex (BA 6), DLPFC (BA 46, 47), IPL (BA 40) and ACC (BA 24/32). Two previous resting state fMRI studies observed the similar fronto-parietal network by using probabilistic independent component analysis (PICA)  or seed-based correlation analysis . Moreover, in an automatic meta-analysis of 825 neuroimaging articles representing 3402 experiments, Toro et al.  obtained a similar co-activation map of fronto-parietal network. The fronto-parietal network in these studies was called as “task-positive” network, which was activated during performance of attention-demanding cognitive tasks. The cingulo-opercular circuit is composed of the medial frontal cortex (BA 6, 8), insula (BA 13), anterior prefrontal cortex (aPFC; BA 11/10), and thalamus. This pathway was recruited in many cognitive tasks, e.g., cognitive control , learning  and planning , etc.
The first goal of the present study is then to investigate the resting-state functional connectivity patterns of the DLPFC in MCI patients. It was hypothesized that the effects of functional disconnection of the two circuits and compensation within lobes coexisted in AD can be detected in its early stage, i.e., MCI, by using DLPFC-FC analysis.
The second goal of this study is to examine the asymmetry of DLPFC-FC in MCI patients. The different role of the bilateral DLPFC is well-known in working memory , episodic memory  for healthy adults. We thus expect to observe the functional asymmetry of DLPFC-FC in healthy controls, and are interested to test whether the functional asymmetry of DLPFC-FC in MCI patients is altered as compared to healthy controls.
Materials and Methods
Twenty-eight right-handed subjects (14 MCI patients and 14 healthy elders) participated in this study. The MCI subjects were recruited from patients who had consulted a memory clinic for memory complaints at Xuanwu Hospital, Beijing, China. The criteria for identification and classification of subjects with MCI  was: (a) impaired memory performance on a normalized objective verbal memory delayed-recall test; (b) recent history of symptomatic worsening in memory; (c) normal or near-normal performance on global cognitive tests [Mini-Mental State Examination (MMSE) score >24], as well as on activities of daily living scale; (d) global rating of 0.5 on the CDR Scale, with a score of at least 0.5 on the memory domain; (e) absence of dementia. The healthy elderly controls were recruited from the local community by advertisements. The criteria for healthy elderly controls were as follows: (i) matched with MCI patients by the gender, age and education level; (ii) had no cognitive complaints and no positive sign in the neurological exam; (iii) MMSE score of 28 or higher.
The demographic and clinical data were shown in Table 1. The study was approved by the Research Ethics Committee of XuanWu Hospital, Capital Medical University and written informed consent was obtained from each participant.
MRI data acquisition was performed on a SIEMENS Trio 3-Tesla scanner (Siemens, Erlangen, Germany). Foam padding and headphones were used to limit head motion and reduce scanner noise. The subjects were instructed to hold still, keep their eyes closed and think nothing in particular. Functional images were collected axially by using an echo-planar imaging (EPI) sequence [repetition time (TR)/echo time (TE)/flip angle (FA)/field of view (FOV) = 2000 ms/40 ms/90°/24 cm, resolution = 64×64 matrix, slices = 28, thickness = 4 mm, gap = 1 mm, bandwidth = 2232 Hz/pixel]. The scan lasted for 478 s. 3D T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) sagittal images were collected by using the following parameters: TR/TE/inversion time (TI)/FA = 1900 ms/2.2 ms/900 ms/9°, resolution = 256×256 matrix, slices = 176, thickness = 1 mm.
Unless otherwise stated, all analyses were conducted using a statistical parametric mapping software package (SPM5, http://www.fil.ion.ucl.ac.uk/spm). The first 10 volumes of the functional images were discarded for the signal equilibrium and participants' adaptation to the scanning noise. The remaining 229 fMRI images were first corrected for within-scan acquisition time differences between slices and then realigned to the first volume to correct for inter-scan head motions. No participant had head motion of more than 1.5 mm maximum displacement in any of the x, y, or z directions and 1.5° of any angular motion throughout the course of scan. The individual structural image was co-registered to the mean functional image after motion correction using a linear transformation. The transformed structural images were then segmented into gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF) by using a unified segmentation algorithm . The motion corrected functional volumes were spatially normalized to the Montreal Neurological Institute (MNI) space and re-sampled to 3 mm isotropic voxels using the normalization parameters estimated during unified segmentation. Subsequently, the functional images were spatially smoothed with a Gaussian kernel of 4×4×4 mm3 full width at half maximum (FWHM) to decrease spatial noise. Following this, temporal filtering (0.01 Hz<f<0.08 Hz) was applied to the time series of each voxel to reduce the effect of low-frequency drifts and high-frequency noise , ,  using Resting-State fMRI Data Analysis Toolkit (http://resting-fmri.sourceforge.net). To further reduce the effects of confounding factors, we also used a linear regression process to further remove the effects of head motion and other possible sources of artifacts : (1) six motion parameters, (2) whole-brain signal averaged over the entire brain, (3) linear drift.
Functional Connectivity Analysis
The bilateral DLPFC ROIs were generated using the free software WFU_PickAtlas (http://www.ansir.wfubmc.edu) . In the present study, DLPFC refers to the BA 46, which has been used in previous studies , . The BOLD time series of the voxels within each seed region were averaged to generate the reference time series for this seed region. For each subject and each seed region, a correlation map was produced by computing the correlation coefficients between the reference time series and the time series of all brain voxels. Correlation coefficients were converted to z values using Fisher's r-to-z transform to improve the normality . For verification, we also perform a coordinate-based DLPFC-FC and PCC functional connectivity (PCC-FC) analysis (See Supplementary Materials: 1).
Then, the individual z value was entered into a random effect one-sample t-test to determine brain regions showing significant connectivity to the left and right DLPFC within each group under a combined threshold of P<0.01 and cluster size > = 405 mm3. This yields a corrected threshold of P<0.05, determined by Monte Carlo simulation using the AlphaSim program with parameters: FWHM = 4 mm, within the GM mask (http://afni.nimh.nih.gov/pub/dist/doc/manual/AlphaSim.pdf). This procedure produced significant DLPFC-FC t-statistic maps for patients and controls. We then made a mask respectively for the bilateral DLPFC by combining the two DLPFC-FC t-statistic maps (i.e., MCI patients, healthy controls), and then used this mask for analyzing the group differences. Such an inclusion masking procedure can constrain our analyses within regions generating potentially meaningful low frequency oscillations. Voxels survived a combined threshold of P<0.01 and cluster size > = 216 mm3, which corresponds to a corrected threshold of P<0.05 (using the AlphaSim program with parameters: FWHM = 4 mm, with mask), were considered to show the group differences.
The z values were also entered into a random effect paired t-test to examine the DLPFC-FC asymmetry for healthy controls and MCI patients, respectively, and two-sample t-test to identify the DLPFC-FC asymmetry differences between MCI patients and healthy controls. Voxels survived a corrected threshold of P<0.05 (using the AlphaSim program with parameters: FWHM = 4 mm, with mask) were considered to show significant difference between the two groups.
Correlation Analysis of DLPFC-FC and Neuropsychological Measures
In order to explore whether DLPFC-FC varies with the disease progression and cognitive performances in MCI patients, a further correlation analysis between DLPFC-FC and neuropsychological performances was performed. First, averaged z-values of each cluster with the significant group differences were extracted. Then, Pearson's correlative analysis were performed to examine relationships between the z-values and neuropsychological performances (including CVLT: Immediate Recall, CVLT: Short Delayed Recall, CVLT: Long Delayed Recall, CDT and MMSE) in MCI patients using SPSS software (SPSS, Inc., Chicago, IL).
Controlling for Regional GM Atrophy
Brain atrophy may cause a partial volume effect in functional imaging techniques . Recent researches have demonstrated the potential impact of atrophy on the functional results in previous fMRI studies of AD , . To explore the possible effect, a voxel-based-morphometry (VBM) analysis of structural images was performed in this study.
GM intensity maps were obtained by the unified segmentation algorithm  as described in the Data Preprocessing section. After spatially smoothed with a Gaussian kernel of 10 mm FWHM, a two-sample t-test was performed on the smoothed GM intensity maps between healthy controls and MCI patients. The statistical threshold was set at P<0.001 and cluster size >324 mm3, which corresponded to a corrected P<0.05 (using the AlphaSim program with parameters: FWHM = 10 mm, within the GM mask).
Furthermore, we re-analyzed the DLPFC-FC results (i.e., two-sample t-tests and correlation analysis) by taking voxel-wise GM volume as covariates . The statistical threshold was also set at a corrected P<0.05.
There were no significant differences between the two groups in gender, age, and years of education, but the MMSE, CVLT and CDT scores were significantly different (P<0.05) between the two groups.
DLPFC-FC: Within-group Results
Figure 1 and Figure 2 respectively show the bilateral DLPFC-FC maps for MCI patients and healthy controls. Visual inspection indicated that the two groups showed the similar patterns of functional connectivity in the bilateral DLPFC. The DLPFC showed strong functional connectivity to a distributed regions including middle frontal gyrus (BA 6, 11, 46, 47), medial frontal gyrus (BA 6, 8), inferior parietal lobule (IPL; BA 40), insula (BA 13), anterior cingulate cortex (ACC; BA 24/32), thalamus, putamen/caudate, etc.
The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at P<0.01 and cluster size > = 405 mm3, which corresponded to a corrected P<0.05. Left is the left.
The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at P<0.01 and cluster size > = 405 mm3, which corresponded to a corrected P<0.05. Left is the left.
The left DLPFC-FC and the right DLPFC-FC were statistically contrasted for healthy controls and MCI patients respectively. Figure 3 showed the results of within-group DLPFC-FC asymmetry.
DLPFC-FC: Between-group Results
Figure 4(A) and Table 2 show the regions with group differences between MCI patients and healthy controls for the left DLPFC-FC. Comparing with healthy controls, decreased functional connectivity with the left DLPFC was observed in MCI patients in the right superior frontal gyrus (BA 6), left inferior/superior parietal lobule (IPL/SPL; BA 39/7, 40), and left thalamus, while increased functional connectivity with the left DLPFC was observed in MCI patients in the right middle frontal gyrus (BA 9).
For the details of the regions, see Table 4 and Table 5. Areas in hot color indicate brain regions with significantly decreased functional connectivity with the left/right DLPFC in MCI patients compared to healthy controls. Areas in cold color indicate brain regions with significantly increased functional connectivity with the left/right DLPFC in MCI patients compared to healthy controls. The threshold was set at p<0.01 with cluster size > = 216 mm3, which corresponded to a corrected threshold of p<0.05. Of note, we showed the two-sample t-test results were attained within a corresponding mask by combining the one sample t-test maps of the two groups (Figure 1 and Figure 2). Left is the left.
Figure 4(B) and Table 3 show the regions with the right DLPFC-FC differences between MCI patients and healthy controls. Comparing with healthy controls, decreased functional connectivity with the right DLPFC was observed in MCI patients in the left middle/inferior frontal gyrus (BA 11/47), right middle frontal gyrus (BA 6), right medial frontal gyrus (BA 9), left IPL/SPL (BA 39/7), left supramarginal gyrus (BA 40), right IPL/SPL/angular gyrus (BA 39, 7, 40), and right putamen, while no regions showed increased connectivity with the right DLPFC in MCI patients.
Figure 5 shows the regions with the DLPFC-FC asymmetry differences between MCI patients and healthy controls. As contrast to healthy controls, MCI patients showed decreased left lateralized DLPFC-FC in the left superior frontal gyrus (BA 10), middle frontal gyrus (BA 6) and right inferior occipital gyrus (BA 18), while decreased right lateralized DLPFC-FC in the right IPL (BA 40), PCu (BA 7), superior frontal gyrus (BA 6, 8) and middle frontal gyrus (BA 9). In addition, compared with healthy controls, MCI patients also showed increased left lateralized DLPFC-FC in the left middle frontal gyrus (BA 8) and IPL (BA 40).
Areas in hot color indicate significant decreased lateralized DLPFC-FC asymmetry, and areas in cold color indicate significant increased lateralized DLPFC-FC asymmetry in MCI patients compared to healthy controls. The threshold was set at p<0.01 with cluster size > = 162 mm3, which corresponded to a corrected threshold of p<0.05. Left is the left.
Additionally, Figures S1, S2, S3 and S4 in the supplementary material show the within-group results of the coordinate-based DLPFC-FC, and Figures S7 and S8 show the between-group results. The DLPFC-FC patterns are similar to those using BA 46 as DLPFC seeds. Figures S5 and S6 show the within-group results of PCC-FC, and Figure S9 show the group differences. As contrast to healthy controls, the connectivities between PCC and most of DMN regions, including PCC, angular gurus, medial frontal gyrus and superior/middle frontal gyrus, were significantly reduced, which are consistent with previous studies , .
Correlations between DLPFC-FC and Neuropsychological Measures
As shown in Figure 6 (A) and (B), correlation analysis revealed that two clusters showed significant correlation between DLPFC-FC and neuropsychological performances in MCI patients. The connectivity of the left DLPFC with a cluster in the left thalamus showed significant or near significant correlation with CDT (r = 0.55, P = 0.02) and MMSE (r = 0.43, P = 0.06). The connectivity between the right DLPFC and a cluster in the right IPL was significantly correlated with CVLT (Immediate Recall) (r = 0.61, P = 0.01) and CVLT (Short Delayed Recall) (r = 0.49, P = 0.04). All the other correlations were not significant (P>0.1). The post hoc multiple comparisons of the correlations between DLPFC-FC and neuropsychological measures were also done to correct for the number of neuropsychological measures (See Supplementary Materials: 2 for more details).
DLPFC-FC: Controlling for Regional GM Atrophy
A VBM analysis was performed to reveal the GM volume differences between MCI patients and healthy controls. The results showed most significant atrophies in the frontal, temporal, parietal, occipital and sub-cortical regions (See Figure S10 (A) and (B) in Supplementary Materials), and the global maxima atrophy was located in the precuneus. This is consistent with numerous previous studies of MCI ,  and AD patients , .
By taking into account the GM atrophy as a covariate, we found that statistical significance was reduced in the regions showed decreased DLPFC-FC in MCI patients as compared to healthy controls (as shown in Table 4 and Table 5). For example, there was a 26 voxels' cluster (T = 3.40) in the right superior frontal gyrus (BA 6) showing significantly group differences (Table 2) but the cluster decreased to 22 voxels (T = 3.38) after the GM correction (Table 4). However, the region in the right middle frontal gyrus (Table 2, 4) which showed increased DLPFC-FC in MCI patients as compared to healthy controls were not affected (The maximal T value changed from 3.13 to 3.27, although the voxel number changed from 25 to 24.).
When controlling for the GM atrophy, the correlations between DLPFC-FC and neuropsychological performances were also changed. While the correlation trend were still reserved, the significant correlations between the connectivity of the left DLPFC with the left thalamus and CDT (r = 0.38, P = 0.1), MMSE (r = 0.20, P = 0.26) were reduced to be not significant. The significant correlations between the connectivity of the right DLPFC with the right IPL and CVLT (Immediate Recall) (r = 0.64, P = 0.01), CVLT (Short Delayed Recall) (r = 0.48, P = 0.05) were kept.
The present study utilized bilateral DLPFC as seeds to compare the DLPFC-FC differences between MCI patients and healthy controls. The DLPFC-FC patterns of healthy controls in the current study covered the two pathways: fronto-parietal circuits and fronto-striatial-thalamus circuits. This is consistent with a previous study . The fronto-parietal and fronto-striatial-thalamus circuits in the current study appear to share the majority elements of the proposed fronto-parietal network and cingulo-opercular network in Dosenbach et al. . Using DLPFC (BA 46) as seeds enable us to follow the two pathways simultaneously, while only parts of them were traced using other regions as seeds . The DLPFC-FC pattern of healthy controls acting as a baseline, the effects of functional disconnection and compensation together with the altered asymmetry were observed in MCI patients.
Functional Disconnection in MCI Patients
Comparing to healthy elderly, MCI patients showed the significant reduced parietal connectivity with the DLPFC. This disconnection effect was still strong even after controlling for GM atrophy. These results demonstrated a functional disconnection within a distributed frontal-parietal network, and confirmed that disconnection between the anterior and the posterior regions can also be detected in MCI, the early stage of AD.
Moreover, significant correlations between the right DLPFC-FC with a cluster in the right IPL and two CVLT scores (i.e., Immediate Recall and Short Delayed Recall) were observed.
When controlling for GM atrophy, the two correlations were still significant. This may suggest that this disconnection contributes to the impairment of episodic memory retrieval in MCI patients, which was measured using CVLT. Previous findings demonstrated that the right DLPFC and parietal cortex were recruited in episodic retrieval, while the left prefrontal cortex was associated with episodic encoding , which may give explanations to the effect of the right lateralization.
We found a region in the medial/superior frontal cortex and some sub-cortical regions including thalamus and putamen showed decreased connectivity with the DLPFC even controlling for GM atrophy. This finding extends the previous studies and implies that the fronto-striatial-thalamus connection may also be disrupted in MCI patients.
Without GM atrophy correction, the correlation between the connectivity of the left DLPFC with a cluster in the left thalamus and CDT and MMSE was significant. After controlling for GM atrophy, the correlation is still positive while the statistic power is reduced. This finding implies the existed fronto-striatial-thalamus disconnection can partially explain the executive function decline (measured by CDT) and disease severity (measured by MMSE) in MCI patients, while GM atrophy may also in part contribute to these effects.
Taking together, based on the facts that the fronto-parietal and fronto-striatial-thalamus circuits are highly correlated with human cognition, our findings suggested that cognitive impairments in MCI may be related to a reduction in the integrated activity of these circuits.
Within-frontal Functional Compensation in MCI Patients
An increased right frontal connectivity with the left DLPFC was also observed in this study.
Previous studies have reported increased activity in the right DLPFC , , and increased functional connectivity between the right prefrontal cortex and other regions  in MCI patients using resting state or various cognitive tasks. This study is also consistent with a recent task-fMRI study of our group , in which number series completion task was used to examine the inductive reasoning ability of MCI patients. MCI patients and healthy controls showed similar accuracies, reaction times and the activation pattern of the bilateral DLPFC. However, the functional connectivity between the left and right DLPFC in MCI patients was significantly stronger than that in healthy controls (as shown in Figure S11 in Supplementary Materials). Consistent with previous studies, the current finding may suggest a compensation mechanism of right prefrontal cortex in MCI patients. It was thus postulated that MCI patients may use additional neural resources in the right prefrontal regions to compensate for losses of cognitive functions.
Altered DLPFC-FC asymmetry in MCI patients
It was found that, as compared to the healthy controls, the DLPFC-FC asymmetries in MCI patients were altered (see Figure 5). The current study then extends the previous findings of brain regional asymmetry (either structural or functional) in MCI patients ,  to the domain of resting state functional connectivity. Brain functional asymmetry is a fundamental characteristic of the intrinsic functional architecture of the brain , and the asymmetry phenomena and its developmental trajectory in the whole life-span have been observed using resting-state fMRI . Thus, the altered functional asymmetry in MCI may underlie its physiological mechanism, and complement for the above findings of DLPFC-FC functional disconnection and compensation.
Comparison with a Previous Study
DLPFC-FC has also been used in first-episode schizophrenia (FES) patients. Zhou et al.  also used bilateral BA46 as seeds and examined the functional dysconnectivity of the DLPFC in FES based on resting-state fMRI. They found that the bilateral DLPFC showed reduced functional connectivity to the parietal lobe, posterior cingulate cortex (PCC), thalamus and striatum and enhanced functional connectivity between the left DLPFC and the left mid-posterior temporal lobe and the paralimbic regions in FES patients. The similar functional dysconnectivity associated with the DLPFC exists both in FES and MCI patients, however, the increased connectivites are different. Considering the potential confounding factors involved in the two group patients (e.g., age, gender, scanning protocols, experimenter, etc.), whether the functional disconnection and functional compensation in this study are specific for MCI is still unclear, and further comparative studies should be performed in the future.
Limitations and Further Considerations
There are still technical and biological limitations in the current study. The first limitation of our present study lies in the effects of physiological noises such as cardiac and respiratory rhythms. We used a relatively low sampling rate (TR = 2 s) for multi-slice acquisitions. According to the Nyquist sampling theorem, high frequency noises such as cardiac rhythm can not be completely removed just by band-pass filtering of 0.01–0.08 Hz. In the future, these physiological effects may be estimated and removed by simultaneously recording the respiratory and cardiac cycle during the data acquisition. The second limitation is that we cannot exclude the possibility that the group differences of DLPFC-FC, to some extent, resulted from different spontaneous thoughts between the two groups, given that the resting state is associated with spontaneous thoughts and internal cognitive processing. In the future, on the one hand, a large number of samples will be collected. By considering spontaneous thoughts as events of no interest, “grand average” on big samples may be statistically helpful. On the other hand, we will ask subjects to report their spontaneous thoughts after scanning, which may contribute to exploring the behavioral group differences in such a resting state study. Additionally, this study was cross-sectional. It has been suggested that about 1/3 of MCI patients in this study develop AD. It would be important in the future to explore changes in DLPFC-FC between those who do and do not progress to AD, and to investigate longitudinal changes in MCI/AD patients.
In summary, abnormal functional connectivity of the DLPFC in MCI patients was for the first time examined in the present study. The findings demonstrated the coexistence of the fronto-parietal and fronto-striatial-thalamus disconnections and within-prefrontal-lobe compensation in MCI patients.
Within-group maps of the left DLPFC-FC in healthy control groups. The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Within-group maps of the left DLPFC-FC in MCI groups. The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Within-group maps of the right DLPFC-FC in healthy control groups. The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Within-group maps of the right DLPFC-FC in MCI groups. The voxels with hot color represent DLPFC positive functional connectivity, and the voxels with cold color represent DLPFC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Within-group maps of the PCC-FC in NC groups. The voxels with hot color represent PCC positive functional connectivity, and the voxels with cold color represent PCC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Within-group maps of the PCC-FC in MCI groups. The voxels with hot color represent PCC positive functional connectivity, and the voxels with cold color represent PCC negative functional connectivity. The statistical threshold was set at a corrected P<0.05. Left is the left.
Regions showing significant group differences in functional connectivity to the left DLPFC. Areas in hot color indicate brain regions with significantly decreased functional connectivity with the left DLPFC in MCI patients compared to healthy controls. Areas in cold color indicate brain regions with significantly increased functional connectivity with the left DLPFC in MCI patients compared to healthy controls. The threshold was set at a corrected threshold of p<0.05. Left is the left.
Regions showing significant group differences in functional connectivity to the right DLPFC. Areas in hot color indicate brain regions with significantly decreased functional connectivity with the right DLPFC in MCI patients compared to healthy controls. No regions showed significantly increased functional connectivity with the right DLPFC in MCI patients compared to healthy controls. The threshold was set at a corrected threshold of p<0.05. Left is the left.
Regions showing significant group differences in PCC functional connectivity. Areas in hot color indicate brain regions with significantly decreased PCC functional connectivity in MCI patients compared to healthy controls. No regions showed significantly increased PCC functional connectivity in MCI patients compared to healthy controls. The threshold was set at a corrected threshold of p<0.05. Left is the left.
Axial view (A) and sagittal view (B) of the T-statistical maps of GM volume differences between MCI and healthy controlling using a voxel-based-mophometry method. There were significant differences in the frontal, temporal, parietal, occipital lobe and sub-cortical regions. The statistical threshold was set at P<0.001 and cluster size >324 mm3, which corresponded to a corrected P<0.05. Left is the left.
The BOLD response in the bilateral DLPFC region during number series completion task. The connectivity between the bilateral DLPFC in MCI patients was stronger than that of normal controls. This figure was amended from Yang et al. (2009).
Conceived and designed the experiments: KL. Performed the experiments: ZW YY PL. Analyzed the data: PL XJ. Contributed reagents/materials/analysis tools: ZW YY PL. Wrote the paper: PL.
- 1. De Lacoste MC, White CL III (1993) The role of cortical connectivity in Alzheimer's disease pathogenesis: A review and model system. Neurobiol Aging 14: 1–16.
- 2. Delbeuck X, Van der Linden M, Collete F (2003) Alzheimer's disease as a disconnection syndrome? Neuropsychol Rev 13: 79–92.
- 3. Gomez-Isla T, Hyman BT (1997) Connections and cognitive impairment in Alzheimer's disease. In: Hyman BT, Duyckaerts C, Christen Y, editors. Connections, Cognition, and Alzheimer's Disease. Berlin: Springer. pp. 149–166.
- 4. Wang L, Zang YF, He Y, Liang M, Zhang XQ, et al. (2006) Changes in hippocampal connectivity in the early stages of Alzheimer's disease: Evidence from resting state fMRI. NeuroImage 31: 496–504.
- 5. Wang K, Liang M, Wang L, Tian LX, Zhang XQ, et al. (2007) Altered functional connectivity in early Alzheimer's disease: A resting-state fMRI study. Hum Brain Mapp 28: 967–978.
- 6. Grady CL, Furey ML, Pietrini P, Horwitz B, Rapoport SI (2001) Altered brain functional connectivity and impaired short-term memory in Alzheimer's disease. Brain 124: 739–756.
- 7. Grady CL, McIntosh AR, Beig S, Keightley ML, Burian H, et al. (2003) Evidence from functional neuroimaging of a compensatory prefrontal network in Alzheimer's disease. J Neurosci 23: 986–993.
- 8. Petersen RC (2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256: 183–194.
- 9. Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, et al. (1999) Mild cognitive impairment, clinical characterization and outcome. Arch Neurol 56: 303–308.
- 10. Biswal B, Yetkin FZ, Haughton VM, Hyde JS (1995) Functional connectivity in the motor cortex of resting human brain using echo-planar MRI. Magn Reson Med 34: 537–541.
- 11. Lowe MJ, Mock BJ, Sorenson JA (1998) Functional connectivity in single and multislice echoplanar imaging using resting-state fluctuations. NeuroImage 7: 119–132.
- 12. Cordes D, Haughton VM, Arfanakis K, Wendt GJ, Turski PA, et al. (2000) Mapping functionally related regions of brain with functional connectivity MR imaging. Am J Neuroradiol 21: 1636–1644.
- 13. Hampson M, Peterson BS, Skudlarski P, Gatenby JC, Gore JC (2002) Detection of functional connectivity using temporal correlations in MR images. Hum Brain Mapp 15: 247–262.
- 14. De Luca M, Beckmann CF, De Stefano N, Matthews PM, Smith SM (2006) fMRI resting state networks define distinct modes of long-distance interactions in the human brain. NeuroImage 29: 1359–1367.
- 15. Greicius MD, Krasnow B, Reiss AL, Menon V (2003) Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. Proc Natl Acad Sci USA 100: 253–258.
- 16. Raichle ME, Macleod AM, Snyder AZ, Powers WJ, Gusnard DA, et al. (2001) A default mode of brain function. Proc Natl Acad Sci USA 98: 676–682.
- 17. Bai F, Watson DR, Yu H, Shi Y, Yuan Y, et al. (2009) Abnormal resting-state functional connectivity of posterior cingulate cortex in amnestic type mild cognitive impairment. Brain Res 1302: 167–174.
- 18. Qi Z, Wu X, Wang Z, Zhang N, Dong H, et al. (2010) Impairment and compensation coexist in amnestic MCI default mode network. NeuroImage 50: 48–55.
- 19. Sorg C, Riedl V, Muhlau M, Calhoun VD, Eichele T, et al. (2007) Selective changes of resting-state networks in individuals at risk for Alzheimer's disease. Proc Natl Acad Sci USA 104: 18760–18765.
- 20. Curtis CE, D'Esposito M (2003) Persistent activity in th prefrontal cortex during working memory. Trends Cogn Sci 7: 415–423.
- 21. Murray LJ, Ranganath C (2007) The dorsolateral prefrontal cortex contributes to successful relational memory encoding. J Neurosci 27: 5515–5522.
- 22. Fan J, McCandliss BD, Fossella J, Flombaum JI, Posner MI (2005) The activation of attentional networks. NeuroImage 26: 471–479.
- 23. Heekeren H, Marrett S, Ruff DA, Bandettini PA, Ungerleider LG (2006) Involvement of human left dorsolateral prefrontal cortex in perceptual decision making in independent of response modality. Proc Natl Acad Sci USA 103: 10023–10028.
- 24. Goel V, Dolan RJ (2004) Different involvement of left prefrontal cortex in inductive and deductive reasoning. Cognition 93: B109–B121.
- 25. Barbey AK, Barsalou LW (2009) Reasoning and problem solving: models. Encyclopedia of Neuroscience 8: 35–43.
- 26. Tekin S, Cummings JL (2002) Frontal-subcortical neuronal circuits and clinical neuropsychiatry: An update. J Psychosom Res 53: 647–654.
- 27. Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, et al. (2007) Distinct brain networks for adaptive and stable task control in humans. Proc Natl Acad Sci USA 104: 11073–11078.
- 28. Elgh E, Larsson A, Eriksson S, Nyberg L (2003) Altered prefrontal brain activity in persons at risk for Alzheimer's disease: An fMRI study. International Psychogeriatrics 15: 121–133.
- 29. Kaufmann L, Ischebeck A, Weiss E, Koppelstaetter F, Siedentopf C, et al. (2008) An fMRI study of the numerical Stroop task in individuals with and without minimal cognitive impairment. Cortex 44: 1248–1255.
- 30. Saykin AJ, Wishart HA, Rabin LA, Flashman LA, McHugh TL, et al. (2004) Cholinergic enhancement of frontal lobe activity in mild cognitive impairment. Brain 127: 1574–1583.
- 31. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, et al. (2005) The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Nat Acad Sci USA 102: 9673–9678.
- 32. Toro R, Fox PT, Paus T (2008) Functional coactivation map of the human brain. Cereb Cortex 18: 2553–2559.
- 33. Casey BJ (In press) Frontostriatal and Frontocerebellar Circuitry underlying Cognitive Control. Developing individuality in the Human Brain. American Psychological Association: Washington, DC.
- 34. Kehagia AA, Murray GK, Robbins TW (2010) Learning and cognitive flexibility: frontostriatal function and monoaminergic modulation. Curr Opin Neurobiol 20: 199–204.
- 35. van den Heuvel OA, Groenewegen HJ, Barkhof F, Lazeron RH, van Dyck R, et al. (2003) Frontostriatal system in planning complexity: a parametric functional magnetic resonance version of Tower of London task. NeuroImage 18: 367–374.
- 36. McCarthy G, Puce A, Constable RT, Krystal JH, Gore JC, et al. (1996) Activation of human prefrontal cortex during spatial and nonspatial working memory tasks measured by functional MRI. Cereb Cortex 6: 600–611.
- 37. Rossi S, Cappa SF, Babiloni C, Pasqualetti P, Miniussi C, et al. (2001) Prefrontal cortex in long-term memory: an “interference” approach using magnetic stimulation. Nat Neurosci 4(9): 948–952.
- 38. Ashburner J, Friston KJ (2005) Unified segmentation. NeuroImage 26: 839–851.
- 39. Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH (2003) An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. NeuroImage 19: 1233–1239.
- 40. Schon K, Atri A, Hasselmo ME, Tricarico MD, LoPresti ML, et al. (2005) Scopolamine reduces persistent activity related to long-term encoding in the parahippocampal gyrus during delayed matching in humans. J Neurosci 25: 9112–9123.
- 41. Zhou Y, Liang M, Jiang TZ, Tian LX, Liu Y, et al. (2007) Functional Dysconnectivity of the Dorsolateral Prefrontal Cortex in First-episode Schizophrenia Using Resting-state fMRI. Neurosci Lett 417: 297–302.
- 42. Bokde AL, Pietrini P, Ibanez V, Furey ML, Alexander GE, et al. (2001) The effect of brain atrophy on cerebral hypometabolism in the visual variant of Alzheimer disease. Arch Neurol 58: 480–486.
- 43. Oakes TR, Fox AS, Johnstone T, Chung MK, Kalin N, et al. (2007) Integrating VBM into the general linear model with voxelwise anatomical covariates. NeuroImage 34: 500–508.
- 44. He Y, Wang L, Zang Y, Tian L, Zhang X, et al. (2007) Regional coherence changes in the early stages of Alzheimer's disease: a combined structural and resting-state functional MRI study. NeuroImage 35: 488–500.
- 45. Baron JC, Chetelat G, Desgranges B, Perchey G, Landeau B, et al. (2001) In vivo mapping of gray matter loss with voxel-based morphometry in mild Alzheimer's disease. NeuroImage 14: 298–309.
- 46. Chételat G, Desgranges B, De La Sayette V, Viader F, Eustache F, et al. (2002) Mapping gray matter loss with voxel-based morphometry in mild cognitive impairment. Neuroreport 13: 1939–1943.
- 47. Buckner RL, Snyder AZ, Shannon BJ, LaRossa G, Sachs R, et al. (2005) Molecular, structural, and functional characterization of Alzheimer's disease: evidence for a relationship between default activity, amyloid, and memory. J Neurosci 25: 7709–7717.
- 48. Fletcher PC, Frith CD, Rugg MD (1997) The functional neuroanatomy of episodic memory. Trends Neurosci 20: 213–218.
- 49. Clément F, Belleville S (2010) Compensation and disease severity on the memory-related activations in mild cognitive impairment. Biol Psychiatry 68: 894–902.
- 50. Rosano C, Aizenstein HJ, Cochran JL, Saxton JA, De Kosky ST, et al. (2005) Event-related functional magnetic resonance imaging investigation of executive control in very old individuals with mild cognitive impairment. Biol Psychiatry 57: 761–767.
- 51. Yang YH, Liang PP, Lu SF, Li KC, Zhong N (2009) The role of the DLPFC in inductive reasoning of MCI patients and normal agings: An fMRI study. Sci China Ser C 52: 789–795.
- 52. Bahar-Fuchs A, Moss S, Rowe C, Savage G (2010) Olfactory performance in AD, aMCI, and healthy ageing: A unirhinal approach. Chem Senses 35: 855–862.
- 53. Shi F, Liu B, Zhou Y, Yu C, Jiang T (2009) Hippocampal volume and asymmetry in mild cognitive impairment and Alzheimer's disease: Meta-analyses of MRI studies. Hippocampus 19: 1055–1064.
- 54. Toga AW, Thompson PM (2003) Mapping brain asymmetry. Nat Rev Neurosci 4: 37–48.
- 55. Zuo XN, Kelly C, Di Martino A, Mennes M, Margulies DS, et al. (2010) Growing together and growing apart: regional and sex differences in the lifespan developmental trajectories of functional homotopy. J Neurosci 30: 15034–15043.