Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jan 31.
Published in final edited form as: Acta Diabetol. 2014 Jul 4;52(1):133–142. doi: 10.1007/s00592-014-0618-7

Hyperactivation of working memory related brain circuits in newly-diagnosed middle-aged type 2 diabetics

Xiao-Song He 1,5,§, Zhao-Xin Wang 1,6,§, You-Zhi Zhu 2, Nan Wang 3, Xiaoping Hu 4, Da-Ren Zhang 1, De-Fa Zhu 3,*, Jiang-Ning Zhou 1,*
PMCID: PMC4416650  NIHMSID: NIHMS611134  PMID: 24993663

Abstract

Type 2 diabetes mellitus (T2DM) is well known for its adverse impacts on brain and cognition, which lead to multidimensional cognitive deficits and wildly-spread cerebral structure abnormalities. However, existing literatures are mainly focused on patients with advanced age or extended T2DM duration. Therefore, it remains unclear whether and how brain function would be affected at the initial onset stage of T2DM in relatively younger population. In current study, twelve newly-diagnosed middle-aged T2DM patients with no previous diabetic treatment history and twelve matched controls were recruited. Brain activations during a working memory task, the digit n-back paradigm (0-, 1- and 2-back), were obtained with functional magnetic resonance imaging (fMRI) and tested by repeated measures ANOVA. Whereas patients performed the n-back task comparably well as controls, significant load-by-group interactions of brain activation were found in the right dorsolateral prefrontal cortex (DLPFC), left middle/inferior frontal gyrus, and left parietal cortex, where patients exhibited hyperactivation in the 2-back but not the 0-back or 1-back condition compared to controls. Furthermore, the severity of chronic hyperglycemia, estimated by glycosylated hemoglobin (HbA1c) level, was entered into partial correlational analyses with task-related brain activations, while controlling for the real-time influence of glucose, estimated by instant plasma glucose level measured before scanning. Significant positive correlations were found between HbA1c and brain activations in the anterior cingulate cortex and bilateral DLPFC only in patients. Taken together, these findings suggest there might be a compensatory mechanism due to brain inefficiency related to chronic hyperglycemia at the initial onset stage of T2DM.

Keywords: type 2 Diabetes, chronic hyperglycemia, working memory, functional magnetic resonance imaging (fMRI)

Introduction

Type 2 diabetes mellitus (T2DM) is a common metabolic disease characterized by hyperglycemia due to reduced insulin sensitivity and relative insulin deficiency [1], which is also well known for its adverse impacts on the brain [1, 2]. For example, insulin resistance is associated with reduced insulin transportation across the blood-brain barrier in humans, leading to alterations in body weight, food intake, and glucose homeostasis [3]. So far, T2DM is found to be associated with impairments of cerebral signal conduction, neurotransmission, synaptic plasticity, cerebral structure and finally cognitive function, which are in all referred to as “diabetic encephalopathies” [4, 5] and now accepted as complications of diabetes [5]. A large and growing body of epidemiological literatures has provided evidence for T2DM related dysfunction in multiple cognitive domains, such as psychomotor speed [6-10], executive function [9, 11], and memory [8-17]. In line with these cognitive impairments, cerebral atrophy [10, 11, 18], white matter lesions [10, 19], and altered resting-state functional connectivity [19-21] are frequently reported in T2DM patients as well. However, it should be noted that most of these studies included patients with advanced age which could be influencing both behavioral performances and imaging assessments. Even in studies with middle-aged patients [6, 12, 16, 19], the results could also be confounded with recurrent hypoglycemic attacks during years of medical treatment [1], not to mention impacts from duration [9] and medication [22] themselves. Therefore, whether T2DM mainly by itself can influence the brain at the initial onset stage remains unclear.

Accordingly in current study, a blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) experiment was conducted to explore the brain function in newly-diagnosed middle-aged T2DM patients during a working memory task, the digit n-back task, which requires participants to monitor a series of digits and to respond to the one presented N trials back [23, 24]. The focus on working memory related brain circuits was inspired by previous structure imaging studies, whereas T2DM related grey matter loss is mainly found in an extended brain network including the prefrontal [10, 25], medial frontal [10], precentral [18], and parietal [26] regions, which is also consistent with the brain network involved in working memory process [23, 27, 28]. Especially, prefrontal atrophy, shown as enlarged lateral ventricles, was found in a group of recent onset middle-aged T2DM patients who were diagnosed and treated within a year [25]. However despite the wildly-spread structure abnormalities within the working memory related brain circuits, neuropsychological results regarding to working memory dysfunction in T2DM patients were divergent. Some studies reported preserved working memory function in T2DM patients [14, 15, 17], while significant working memory dysfunction was observed in T2DM patients during acute hyperglycemia [8, 13]. Considering fMRI is sensitive to modest functional reorganization even in patients who can achieve comparable behavioral performance [29-31], we hypothesized that at least altered brain activation would be found in these patients, whether with or without declined working memory task performance.

The second goal of this study was to explore the role of chronic hyperglycemia in the disruption of brain functioning at the initial onset stage of T2DM. Glycosylated hemoglobin (HbA1c) was used to reflect the severity of chronic hyperglycemia in these patients, since its concentration is positively associated with the mean blood sugar concentration over previous weeks to months [32]. Then the potential association between HbA1c and brain function was tested with correlational analysis. Besides, in order to minimize the confounding influences of aging, recurrent hypoglycemic attacks, diabetic duration, and medication, all patients were middle-aged, and had been diagnosed with diabetes for the first time, hence were not previously treated with any kind of diabetic treatment. We also excluded patients with hyperlipidemia or hypertension to avoid confounding factors from other diabetes-associated comorbidities.

Material and Methods

Participants

Twelve right-handed middle-aged T2DM patients and twelve demographically matched right-handed euglycemic controls (see respective demographic information in Table 1) participated in the present study. All patients were outpatients of the First Affiliated Hospital of Anhui Medical University with recent onset of diabetes related symptoms, who were diagnosed with T2DM according to the World Health Organization 1999 criteria [33] for the first time and had no previous diabetic treatment history. Both patients and controls in this study had normal or correct-to-normal vision and normal liver and renal functions. Magnetic resonance images were visually inspected by an experienced radiologist blind to the diabetic diagnosis, who ruled out the presence of moderate atrophy or other structural changes in each participant. Furthermore, we verified the visual inspection by implementing a supplementary voxel-based morphometry (VBM) [34] analysis on the anatomical imaging data of participants, which revealed no significant global or regional difference of grey matter volume between patients and controls (procedure described in Supplementary Document 1). Participants were excluded if they had a history of cerebral hemorrhage or infarction, neurological or psychiatric disorders, or had systolic blood pressure ≥ 140mmHg and (or) diastolic blood pressure ≥ 90mmHg, plasma triglycerides ≥ 1.7 mmol/L, or plasma total cholesterol ≥ 5.2mmol /L. No presence of ketoacidosis was recorded in any participant during the entire experiment.

Table 1. Comparison of type 2 diabetic (n = 12) and control (n = 12) participants on demographic and clinical characteristics.

Type 2 Diabetics Healthy controls P
Age (years) 41.1 ± 6.1 (33-52) 40.5 ± 6.2 (31-53) 0.8
Gender (M/F) 11/1 8/4 0.1
Education (years) 9.8 ± 4.4 (1-15) 9.3 ± 3.7 (2-15) 0.8
Triglyceride 1.22 ± 0.25 (0.72-1.50) 1.22 ± 0.22 (0.82-1.51) 0.9
Cholesterol 3.72 ± 0.57 (2.59-4.70) 3.86 ± 0.64 (2.95-4.81) 0.6
Blood pressure (mm Hg):
Systolic 115.3 ± 7.0 (105-126) 118.3 ± 7.0 (109-129) 0.3
Diastolic 77.3 ± 4.2 (72-83) 78.2 ± 3.5 (72-83) 0.6
HOMA2-%β 12.1 ± 4.8 (5.3-19.3) 105.8 ± 21.1 (74.9-141.9) <0.001
Instant plasma glucose (mmol/L) 23.5 ± 3.1 (17.7-27.7) 6.8 ± 1.0 (5.4-8.7) <0.001
HbA1c (%) 10.0 ± 1.8 (8.1-13.2) 5.1 ± 0.5 (4.8-6.0) <0.001
Open in a new tab

Notes: Abbreviations: HOMA2-%β: homeostasis model assessment for b-cell function. HbA1c: glycosylated hemoglobin. Data are means ± S.D. (range)

All patients underwent the fMRI scanning the day immediately after their diagnosis*. To avoid the influence of medication, they were asked to withhold their medical treatment until the scanning was over. The total withholding time (from diagnosis to the end of the scanning) was less than 24 hours in every patient. Written informed consent was obtained from all participants in this study. The experiments were performed with the approval of Anhui Medical University Human Subjects Review Committee, and were conducted according to the principles expressed in the Declaration of Helsinki.

Glucose and insulin levels

Plasma glucose (PG) and insulin levels were measured during Oral Glucose Tolerance Test (OGTT). The HbA1c was assayed as an estimate for chronic hyperglycemia. In order to control for the real-time influence of glucose on brain activation, instant PG level, which was measured immediately before scanning as an estimate for the averaged real-time PG level during the whole fMRI experiment, was used as a controlled variable in the following correlational analyses.

Homeostasis model assessment version 2 (HOMA2) was used to assess pancreatic β-cell function (HOMA2-%β) from fasting PG and fasting insulin levels, which was completed by using the HOMA Calculator version 2.2, available from the Diabetes Trials Unit website (http://www.dtu.ox.ac.uk) [35, 36].

Paradigm of fMRI experiment

A blocked digit n-back design [24] was used in the present study with three gradually increased task-loads (0-, 1-, and 2-back). Each task block started with a 2-s instruction (0-back: “report the current digit”; 1-back: “report the first digit prior to the current one”; 2-back: “report the second digit prior to the current one”. Displayed in Chinese). After a 1-s delay, 10 digits were presented serially with 2-s duration and followed by a 1-s fixation. During this 3-s period, the participants gestured the recalled digit with right hand, which was recorded by one author standing beside the scanner. Another 2-s instruction (“rest” in Chinese) followed by 31-s fixation was used as a baseline control block, during which the participants were asked to rest with their eyes open. The fixation, instruction and digits extended about 0.8° × 0.8°, 8° × 2°, and 2° × 1°, respectively in visual angle.

All participants underwent 2 fMRI runs with 4 epochs in each run. Each epoch consisted of a 0-back, a 1-back and a 2-back task block, and followed by a baseline control block. The sequence of the task blocks was 0-, 1-, 2-back in one run, and 2-, 1-, 0-back in the other. The two task sequences were counterbalanced across all the participants in each group.

MRI data acquisition

Imaging data were collected on a 1.5 Tesla Philips Infinion MR System (Royal Philips Electronics, Netherlands). During each functional scan, 179 volumes consisting of 16 axial slices were obtained with a T2*-weighted echo-planar imaging (EPI) sequence (TR = 3 s, TE = 40 ms, FOV = 24 cm × 24 cm, matrix = 64 × 64, flip angle = 90°, slice thickness = 4 mm, gap between two adjacent slices = 1.2 mm). Corresponding high-resolution T1-weighted imaging data were also obtained with a spin echo (SE) sequence (2D anatomical data) and a spoiled gradient recalled echo (SPGR) sequence (3D anatomical data).

Analysis of fMRI data

MRI data were analyzed using AFNI (Analysis of Functional NeuroImages, http://afni.nimh.nih.gov/afni/) [37]. The first three volumes of each scan were excluded from the analysis in order to account for the approach to steady state of the longitudinal magnetization. Then data was processed to remove any linear drift and to correct motion and subsequently smoothed with a Gaussian filter (FWHM = 6 mm) and normalized. Any scan in which the head motion was larger than 2 mm was excluded from further analysis.

The preprocessed data were analyzed in two ways: group and individual analysis [24]. In the group analysis, all the participants' functional data sets of each group were averaged after being stereotaxically transformed into Talairach space with 4 mm resolution. Correlation analyses based on the direct contrast between the task (including the 0-, 1- and 2-back block) and baseline were then carried out to generate the activation map for each group (P ≤ 10−10, cluster size ≥ 4 voxels), which were related to the n-back task irrespective of the task-load (0-, 1-, and 2-back). For individual analysis, a mixed regression analysis with General Linear Model was employed, comprising three task-related square-wave block regressors, each for 0-, 1-, and 2-back condition, respectively, and 6 regressors corresponding to the head motion covariates [38]. As a result, three activation maps (P ≤ 0.05, cluster size ≥ 4 voxels) were generated, and a combined activation map was obtained by the logical ‘OR’ of these maps. The regions of interests (ROIs) for individual participants were then identified on this combined map. For each ROI, mean parameter estimates at the three n-back task-loads were calculated for each participant.

Statistical analysis

Group level comparisons of demographic and clinical characteristics were conducted with independent two-sample t-test except for gender, which was analyzed by a Chi-square test. In addition, Pearson and Spearman correlation analyses were carried out between HbA1c and instant PG level to verify whether they were associated with each other.

The performance accuracies and BOLD responses in the n-back task were evaluated with a 2 (patients, controls) × 3 (0-, 1-, 2-back) repeated measures ANOVA. In order to explore the relationship between chronic hyperglycemia and brain activation while controlling for the real-time influence of glucose, partial correlational analyses were conducted between HbA1c and task-related BOLD responses (averaged across all task-loads for better signal-to-noise ratio), with instant PG level as the controlled variable. All statistical analyses were performed using SPSS for windows (version 17, SPSS Inc, Chicago, IL, USA), with the level of significance set to 0.05 (two-tailed).

Results

Clinical data

During the OGTT, the patients showed significantly higher PG levels (P < 0.001, all time points) and lower plasma insulin levels (P < 0.01, within 60 minutes) than controls (See plotting in Supplementary Fig. 1). Besides, the patients also exhibited significantly elevated HbA1c and instant PG level but lower HOMA2-%β index, in contrast with controls (P < 0.001, Table 1). There were no significant differences in the plasma level of triglyceride, cholesterol, and systolic/diastolic pressures between groups (P > 0.05, Table 1).

It is noteworthy that neither Pearson nor Spearman correlation analysis could reveal a significant correlation between HbA1c and instant PG level (P > 0.05), which meant patients with severer chronic hyperglycemia did not correspondingly exhibit higher averaged real-time PG level during the experiment.

Performance accuracy of the n-back task

The overall accuracies of all 3 task-loads of the n-back task were very high in both groups (Patients: 0-back, 100.0 ± 0 %; 1-back, 98.5 ± 3.1%; 2-back, 92.6 ± 5.3%; Controls: 0-back, 99.7 ± 0.6 %; 1-back, 98.5 ± 2.2%; 2-back, 95.8 ± 5.4%). The accuracies were significantly decreased with difficulty level across the three n-back task-loads in either groups (F(2, 22) > 4.1, P < 0.05). No group difference (F(1, 22) = 2.8, P = 0.11) or load-by-group interaction (F(2, 44) = 2.0, P = 0.14) was found.

Functional MRI data

Regions of interest

During the n-back task, a common brain network, including bilateral dorsolateral prefrontal cortices (DLPFC), bilateral middle/inferior frontal gyri (M/IFG), bilateral premotor areas (PM), the anterior cingulate cortex (ACC), and bilateral parietal cortices (PA) (Fig. 1), was activated in both groups, which is consistent with previous reports [23, 24]. These regions were then identified from the combined activation map described above as ROIs. The bilateral visual cortices were used as the control ROIs to rule out the potential global effect of T2DM since the perceptual load of visual stimuli is sustained in all task conditions.

Fig. 1.

Fig. 1

Open in a new tab

Brain areas involved in the digit n-back task were presented. The activation map for the controls was depicted (P < 10−10) upside while the activation map for the type 2 diabetic patients was depicted (P < 10−10) downside. The patients presented a similar but more wildly distributed activation pattern comparing to controls. Abbreviations: T2DM: Type 2 diabetes mellitus. DLPFC: dorsolateral prefrontal cortex. M/IFG: middle/inferior frontal gyrus. ACC: anterior cingulate cortex. PA: parietal cortex. PM: premotor area. R: right. L: Left

BOLD responses as a function of n-back load

Consistent with early studies [23], the load effect of BOLD response, which referred to the increase of BOLD response corresponding to the increase of task difficulty (load), appeared in the DLPFC, M/IFG, PM, ACC, and PA of both hemispheres in both groups (F(2, 22) > 5, P < 0.01). No load effect was found in visual cortices for either groups (F(2, 22) < 3.2, P > 0.05) except in the left visual cortex of controls (F(2, 22) = 4, P = 0.03), which was also identified in a previous study [39] and may result from top-down modulation.

Notably, significant load-by-group interaction was found in right DLPFC, left M/IFG, and left PA (F(2, 44) > 3.8, P < 0.05). In these regions, patients showed elevated BOLD responses than controls during the 2-back (P < 0.05) but not 0-back or 1-back condition (Fig. 2). No significant group difference or load-by-group interaction was detected in the bilateral visual cortices, the control ROIs.

Fig. 2.

Fig. 2

Open in a new tab

Graphs representing brain activation in selected regions of interest at each task load of the N-back task. The right DLPFC, left M/IFG, and left parietal cortex showed significant load-by-group interaction, where type 2 diabetic patients elicited higher BOLD responses in the 2-back task relative to controls. No significant group difference or load-by-group interaction was found in the bilateral visual cortices (left visual cortex was shown as an example). Abbreviations: DLPFC: dorsolateral prefrontal cortex. M/IFG: middle/inferior frontal gyrus. Hollow diamond, type 2 diabetic patients (n = 12); hollow square, normal controls (n = 12). Error bar is 1 SE. * P < 0.05, two tailed

To further validate our observation, ANCOVA analyses were also used to retest above load-by-group interactions. Potential confounding factors, such as age, cholesterol, and blood pressure, were included one at a time as the covariate, to rule out the possible residual effects. Consistent with above results, significant load-by-group interaction still could be found in right DLPFC, left M/IFG, and left PA (F(2, 42) > 3.4, P < 0.05).

Correlational analysis

After controlling for the instant PG level, positive correlations between HbA1c and the BOLD responses in the ACC (r = 0.71, P = 0.015) and bilateral DLPFC (Left: r = 0.61, P = 0.046; Right: r = 0.65, P = 0.032) were revealed in patients (see residual plotting in Supplementary Fig. 2). However, no significant correlation between HbA1c and BOLD responses was found in any of the ROIs in the control group (P > 0.05, Table 2).

Table 2. Partial correlations (controlled for plasma glucose level before scanning) between averaged BOLD responses and glycosylated hemoglobin (HbA1c) concentration.
ROI Type 2 diabetics (n = 12) Controls (n = 12)
r P r P
DLPFC Left .611 * .046 −.012 .973
Right .646 * .032 −.352 .289
M/IFG Left .179 .598 −.179 .598
Right .005 .989 −.161 .636
ACC .707 * .015 −.189 .578
Premotor Left .370 .263 .099 .773
Right .559 .074 .074 .828
PA Left .376 .254 .025 .941
Right .383 .245 −.240 .477
VC Left .300 .371 −.257 .446
Right .169 .620 −.428 .189
Open in a new tab

Notes: Abbreviations: ROI: regions of interests. DLPFC: dorsolateral prefrontal cortex. M/IFG: middle/inferior frontal gyri. ACC: anterior cingulate cortex. Premotor: premotor area. PA: parietal cortex. VC: visual cortex.

*

P < 0.05, two tailed

Discussion

Our present study provided neurobiological evidence that newly-diagnosed middle-aged T2DM patients exhibited abnormal brain activation during a working memory task. Specifically, compared to controls, greater BOLD responses during the 2-back condition were found in multiple brain regions of patients, including the right DLPFC, left M/IFG, and left PA. Furthermore, T2DM patients but not controls with higher HbA1c level tended to have larger BOLD responses in the ACC and bilateral DLPFC, after controlling for the instant PG level. Since other factors which might also affect brain function, such as aging, diabetic duration, recurrent hypoglycemia attacks, medication, or other diabetic comorbidities were mainly controlled in this study, the above findings may suggest a major role of chronic hyperglycemia in the alteration of brain function at the initial onset stage of T2DM.

A notable fact is that our patients showed abnormally elevated BOLD responses within the right DLPFC, left M/IFG, and left PA only during the 2-back but not the 0-back and 1-back conditions, whereas their task performances were comparably high as controls'. Such “extra” recruitment of brain activation or so called hyperactivation has been suggested as a compensational strategy of the brain during the early stage of progressive inefficiency, which was more commonly reported in Alzheimer's Disease (AD) related studies [40, 41]. In these studies, compensating hyperactivation is considered as an important imaging maker for the prognosis of AD since greater hippocampal hyperactivation is associated with severer hippocampus atrophy [42, 43] and predicts more subsequent cognitive decline [44, 45]. In analogy to these findings, for recent onset middle-aged T2DM patients, the present results showed hyperactivation in their prefrontal brain regions, whereas previous study revealed their prefrontal atrophy represented by enlarged lateral ventricles [25]. In line with recently accumulating literatures, these evidence may further support the idea that T2DM and AD may share a similar pathophysiological pathway [46].

The T2DM related hyperactivation observed in this study may not be interpreted as a global effect. Firstly, it is task-load specific since the group level difference was only found in the 2-back condition. Secondly, it is also regional specific since no group level difference was found in the bilateral visual cortices during either task condition. The preservation of visual cortical activation in these T2DM patients who were scanned during instant hyperglycemic state is also in line with a previous study, which showed the lack of effect of acute hyperglycemia on visual cortical activation [47]. Therefore our findings may not reflect the consequence of a global elevation of glucose metabolism or oxygen consumption due to the progress of T2DM.

As the second goal of this study, the association between chronic hyperglycemia and brain activation was also explored. After controlling for the instant PG level, we found that T2DM patients with higher HbA1c level had larger task-related BOLD responses in the ACC and bilateral DLPFC. Previous epidemiological studies showed that higher HbA1c was associated with lower cognitive function [48, 49] in T2DM patients. Particularly in middle-aged T2DM patients, HbA1c was negatively correlated with psychomotor efficiency [6] and inhibitory control ability [50], which are dominantly related to the function of DLPFC and ACC [51, 52]. Other than decline in behavioral performance, elevated HbA1c was also associated with severer prefrontal atrophy in middle-aged [16] and adolescent [53] T2DM patients. In line with above evidence, our results indicated that T2DM patients with severer chronic hyperglycemia tend to have more compensating activation in the anterior part of the working memory related brain network even at the initial onset stage of the disease, which may provide a novel insight of T2DM related regional vulnerability.

The partial correlation analysis was used to rule out the confounding from the real-time influence of glucose on brain function, since patients with higher HbA1c could also happen to exhibit higher real-time PG level during the experiment, especially given the fact that our patients were treatment-free and HbA1c did reflect the daily accumulation of instant PG levels. Therefore we firstly tested the association between HbA1c and instant PG level, and found non-significant correlation both with parametric and non-parametric methods. Then by implementing the instant PG level as a controlled variable, the ACC and bilateral DLPFC survived in the correlation between HbA1c level and brain activation, leaving the right PM marginal significant. Taking all these evidence into consideration, it seemed that the association between HbA1c level and brain function in our results might not be confounded by the real-time influence of glucose but represent an impact from chronic hyperglycemia.

The mechanisms underlying changes in the brain activation during hyperglycemia are not clear [54]. One possible explanation is the metabolic changes caused by hyperglycemia, such as suppressed blood-brain glucose transfer [55], decreased regional cerebral blood flow [56] and reduced regional glucose metabolism [57]. Another possibility is the “toxic” effect of hyperglycemia which can lead to progressive structural and functional abnormalities in the brain [58]. Such glucose toxicity is mediated by increased flux of glucose through the polyol and hexosamine pathway [59], increased productions of oxidative stress [60], and accumulations of advanced glycation end-products [61]. Nevertheless, the present study may provide some clues for the role that hyperglycemia could have played in the pathogenesis of cerebral dysfunction at the early stage of T2DM.

Note that the T2DM related brain hyperactivation may also be associated with underlying brain atrophy [10] or cerebrovascular lesions that are frequently reported in former T2DM studies, such as white matter hyperintensities [9, 10], cerebral infarcts [9, 10], or microbleeds [10] (for reviews, see [62, 63]). In the present study, moderate to severe atrophy or structural changes were ruled out by visual inspection in the first instance since they could also be caused by other etiologies and confound with our results. Then a supplementary VBM analysis was implemented (see Supplementary Document 1 for details) but no group difference of grey matter volume was found at neither global nor regional level. Therefore the functional change we observed in these T2DM patients may not be attributed to potential gray matter atrophy in their brain. However, still it should be admitted that cerebrovascular lesions like white matter hyperintensities, cerebral infarcts, or microbleeds are better assessed with more specific MRI sequences [64] which hence are beyond the scope of current study. Accordingly, the potential existence of such cerebrovscular lesions at the initial onset stage of T2DM remains to be explored in the future.

In summary, our study showed that newly-diagnosed middle-aged T2DM patients exhibited brain hyperactivation within the right DLPFC, left M/IFG, and left PA during a high working memory load task condition (2-back). Furthermore, patients with severer chronic hyperglycemia, represented by higher HbA1c, tended to have greater brain activation in the ACC and bilateral DLPFC. These findings provided exploratory evidence for the impact of chronic hyperglycemia on patients' brain network and suggested a compensational mechanism at the initial onset stage of T2DM.

Supplementary Material

Supplmental document 1
Supplemental Figure 1
Supplementla Figure 2

Acknowledgments

We thank Mr. Zhong-Lin Pan for assisting with MRI data collection. We also thank Dr. Ning Ma and Dr. Nan Li for their advices. Thanks for two anonymous reviewers for their very helpful comments on our previous manuscript. This research was supported by the Natural Science Foundation of China (Nos. 30700235, 31070986, 30870764, 91132304, and 81272152), National Institutes of Health of the United States (RO1EB002009), and China Postdoctoral Science Foundation (2012M520424).

Footnotes

*

In order to better understand how the brain is affected by daily hyperglycemic events, we intentionally scanned our patients about one hour after their regular lunches, to mimic the “worst scenario” of their circadian glycemic fluctuations. All controls were also scanned with the same arrangement.

Conflict of Interest: Xiao-Song He, Zhao-Xin Wang, You-Zhi Zhu, Nan Wang, Xiaoping Hu, Da-Ren Zhang, De-Fa Zhu, Jiang-Ning Zhou declare that they have no conflict of interest.

Statement of Human and Animal Rights: All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008.

Statement of Informed Consent: Informed consent was obtained from all participants for being included in the study.

References

  • 1.McCrimmon RJ, Ryan CM, Frier BM. Diabetes and cognitive dysfunction. Lancet. 2012;379:2291–9. doi: 10.1016/S0140-6736(12)60360-2. [DOI] [PubMed] [Google Scholar]
  • 2.Biessels GJ, Deary IJ, Ryan CM. Cognition and diabetes: a lifespan perspective. Lancet Neurol. 2008;7:184–90. doi: 10.1016/S1474-4422(08)70021-8. [DOI] [PubMed] [Google Scholar]
  • 3.Heni M, Schopfer P, Peter A, Sartorius T, Fritsche A, Synofzik M, Haring HU, Maetzler W, Hennige AM. Evidence for altered transport of insulin across the blood-brain barrier in insulin-resistant humans. Acta Diabetol. 2013 doi: 10.1007/s00592-013-0546-y. Epub ahead of print. [DOI] [PubMed] [Google Scholar]
  • 4.Biessels G. Diabetic Encephalopathy. In: Veves A, Malik R, editors. Diabetic Neuropathy. Humana Press; 2007. pp. 187–205. [Google Scholar]
  • 5.Sima AA. Encephalopathies: the emerging diabetic complications. Acta Diabetol. 2010;47:279–93. doi: 10.1007/s00592-010-0218-0. [DOI] [PubMed] [Google Scholar]
  • 6.Ryan CM, Geckle MO. Circumscribed cognitive dysfunction in middle-aged adults with type 2 diabetes. Diabetes Care. 2000;23:1486–93. doi: 10.2337/diacare.23.10.1486. [DOI] [PubMed] [Google Scholar]
  • 7.Mogi N, Umegaki H, Hattori A, Maeda N, Miura H, Kuzuya M, Shimokata H, Ando F, Ito H, Iguchi A. Cognitive function in Japanese elderly with type 2 diabetes mellitus. J Diabetes Complications. 2004;18:42–6. doi: 10.1016/S1056-8727(03)00078-3. [DOI] [PubMed] [Google Scholar]
  • 8.Sommerfield AJ, Deary IJ, Frier BM. Acute hyperglycemia alters mood state and impairs cognitive performance in people with type 2 diabetes. Diabetes Care. 2004;27:2335–40. doi: 10.2337/diacare.27.10.2335. [DOI] [PubMed] [Google Scholar]
  • 9.Manschot SM, Brands AM, van der Grond J, Kessels RP, Algra A, Kappelle LJ, Biessels GJ Utrecht Diabetic Encephalopathy Study G. Brain magnetic resonance imaging correlates of impaired cognition in patients with type 2 diabetes. Diabetes. 2006;55:1106–13. doi: 10.2337/diabetes.55.04.06.db05-1323. [DOI] [PubMed] [Google Scholar]
  • 10.Moran C, Phan TG, Chen J, Blizzard L, Beare R, Venn A, Munch G, Wood AG, Forbes J, Greenaway TM, Pearson S, Srikanth V. Brain atrophy in type 2 diabetes: regional distribution and influence on cognition. Diabetes Care. 2013;36:4036–42. doi: 10.2337/dc13-0143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.van Elderen SG, de Roos A, de Craen AJ, Westendorp RG, Blauw GJ, Jukema JW, Bollen EL, Middelkoop HA, van Buchem MA, van der Grond J. Progression of brain atrophy and cognitive decline in diabetes mellitus: a 3-year follow-up. Neurology. 2010;75:997–1002. doi: 10.1212/WNL.0b013e3181f25f06. [DOI] [PubMed] [Google Scholar]
  • 12.Dey J, Misra A, Desai NG, Mahapatra AK, Padma MV. Cognitive function in younger type II diabetes. Diabetes Care. 1997;20:32–5. doi: 10.2337/diacare.20.1.32. [DOI] [PubMed] [Google Scholar]
  • 13.Cox DJ, Kovatchev BP, Gonder-Frederick LA, Summers KH, McCall A, Grimm KJ, Clarke WL. Relationships between hyperglycemia and cognitive performance among adults with type 1 and type 2 diabetes. Diabetes Care. 2005;28:71–7. doi: 10.2337/diacare.28.1.71. [DOI] [PubMed] [Google Scholar]
  • 14.Arvanitakis Z, Wilson RS, Li Y, Aggarwal NT, Bennett DA. Diabetes and function in different cognitive systems in older individuals without dementia. Diabetes Care. 2006;29:560–5. doi: 10.2337/diacare.29.03.06.dc05-1901. [DOI] [PubMed] [Google Scholar]
  • 15.Bruehl H, Rueger M, Dziobek I, Sweat V, Tirsi A, Javier E, Arentoft A, Wolf OT, Convit A. Hypothalamic-pituitary-adrenal axis dysregulation and memory impairments in type 2 diabetes. J Clin Endocrinol Metab. 2007;92:2439–45. doi: 10.1210/jc.2006-2540. [DOI] [PubMed] [Google Scholar]
  • 16.Bruehl H, Wolf OT, Sweat V, Tirsi A, Richardson S, Convit A. Modifiers of cognitive function and brain structure in middle-aged and elderly individuals with type 2 diabetes mellitus. Brain Res. 2009;1280:186–94. doi: 10.1016/j.brainres.2009.05.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Arvanitakis Z, Bennett DA, Wilson RS, Barnes LL. Diabetes and cognitive systems in older black and white persons. Alzheimer Dis Assoc Disord. 2010;24:37–42. doi: 10.1097/WAD.0b013e3181a6bed5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chen Z, Li L, Sun J, Ma L. Mapping the brain in type II diabetes: Voxel-based morphometry using DARTEL. Eur J Radiol. 2012;81:1870–6. doi: 10.1016/j.ejrad.2011.04.025. [DOI] [PubMed] [Google Scholar]
  • 19.Hoogenboom WS, Marder TJ, Flores VL, Huisman S, Eaton HP, Schneiderman JS, Bolo NR, Simonson DC, Jacobson AM, Kubicki M, Shenton ME, Musen G. Cerebral White Matter Integrity and Resting-State Functional Connectivity in Middle-aged Patients With Type 2 Diabetes. Diabetes. 2014;63:728–38. doi: 10.2337/db13-1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Musen G, Jacobson AM, Bolo NR, Simonson DC, Shenton ME, McCartney RL, Flores VL, Hoogenboom WS. Resting-state brain functional connectivity is altered in type 2 diabetes. Diabetes. 2012;61:2375–9. doi: 10.2337/db11-1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Xia W, Wang S, Sun Z, Bai F, Zhou Y, Yang Y, Wang P, Huang Y, Yuan Y. Altered baseline brain activity in type 2 diabetes: a resting-state fMRI study. Psychoneuroendocrinology. 2013;38:2493–501. doi: 10.1016/j.psyneuen.2013.05.012. [DOI] [PubMed] [Google Scholar]
  • 22.Naor M, Steingruber HJ, Westhoff K, Schottenfeld-Naor Y, Gries AF. Cognitive function in elderly non-insulin-dependent diabetic patients before and after inpatient treatment for metabolic control. J Diabetes Complications. 1997;11:40–6. doi: 10.1016/1056-8727(95)00106-9. [DOI] [PubMed] [Google Scholar]
  • 23.Owen AM, McMillan KM, Laird AR, Bullmore E. N-back working memory paradigm: a meta-analysis of normative functional neuroimaging studies. Hum Brain Mapp. 2005;25:46–59. doi: 10.1002/hbm.20131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhu DF, Wang ZX, Zhang DR, Pan ZL, He S, Hu XP, Chen XC, Zhou JN. fMRI revealed neural substrate for reversible working memory dysfunction in subclinical hypothyroidism. Brain. 2006;129:2923–30. doi: 10.1093/brain/awl215. [DOI] [PubMed] [Google Scholar]
  • 25.Lee JH, Yoon S, Renshaw PF, Kim TS, Jung JJ, Choi Y, Kim BN, Jacobson AM, Lyoo IK. Morphometric changes in lateral ventricles of patients with recent-onset type 2 diabetes mellitus. PLoS One. 2013;8:e60515. doi: 10.1371/journal.pone.0060515. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Last D, Alsop DC, Abduljalil AM, Marquis RP, de Bazelaire C, Hu K, Cavallerano J, Novak V. Global and regional effects of type 2 diabetes on brain tissue volumes and cerebral vasoreactivity. Diabetes Care. 2007;30:1193–9. doi: 10.2337/dc06-2052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Smith EE, Jonides J, Marshuetz C, Koeppe RA. Components of verbal working memory: evidence from neuroimaging. Proc Natl Acad Sci U S A. 1998;95:876–82. doi: 10.1073/pnas.95.3.876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.D'Esposito M, Postle BR, Rypma B. Prefrontal cortical contributions to working memory: evidence from event-related fMRI studies. Exp Brain Res. 2000;133:3–11. doi: 10.1007/s002210000395. [DOI] [PubMed] [Google Scholar]
  • 29.Cabeza R, Daselaar SM, Dolcos F, Prince SE, Budde M, Nyberg L. Task-independent and task-specific age effects on brain activity during working memory, visual attention and episodic retrieval. Cereb Cortex. 2004;14:364–75. doi: 10.1093/cercor/bhg133. [DOI] [PubMed] [Google Scholar]
  • 30.Emery L, Heaven TJ, Paxton JL, Braver TS. Age-related changes in neural activity during performance matched working memory manipulation. Neuroimage. 2008;42:1577–86. doi: 10.1016/j.neuroimage.2008.06.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.He XS, Ma N, Pan ZL, Wang ZX, Li N, Zhang XC, Zhou JN, Zhu DF, Zhang DR. Functional magnetic resource imaging assessment of altered brain function in hypothyroidism during working memory processing. Eur J Endocrinol. 2011;164:951–9. doi: 10.1530/EJE-11-0046. [DOI] [PubMed] [Google Scholar]
  • 32.Koenig RJ, Peterson CM, Jones RL, Saudek C, Lehrman M, Cerami A. Correlation of glucose regulation and hemoglobin AIc in diabetes mellitus. N Engl J Med. 1976;295:417–20. doi: 10.1056/NEJM197608192950804. [DOI] [PubMed] [Google Scholar]
  • 33.Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med. 1998;15:539–53. doi: 10.1002/(SICI)1096-9136(199807)15:7<539::AID-DIA668>3.0.CO;2-S. [DOI] [PubMed] [Google Scholar]
  • 34.Ashburner J, Friston KJ. Voxel-based morphometry--the methods. Neuroimage. 2000;11:805–21. doi: 10.1006/nimg.2000.0582. [DOI] [PubMed] [Google Scholar]
  • 35.Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998;21:2191–2. doi: 10.2337/diacare.21.12.2191. [DOI] [PubMed] [Google Scholar]
  • 36.Wallace TM, Levy JC, Matthews DR. Use and abuse of HOMA modeling. Diabetes Care. 2004;27:1487–95. doi: 10.2337/diacare.27.6.1487. [DOI] [PubMed] [Google Scholar]
  • 37.Cox RW. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res. 1996;29:162–73. doi: 10.1006/cbmr.1996.0014. [DOI] [PubMed] [Google Scholar]
  • 38.Wang ZX, Zhang JX, Wu QL, Liu N, Hu XP, Chan RC, Xiao ZW. Alterations in the processing of non-drug-related affective stimuli in abstinent heroin addicts. Neuroimage. 2010;49:971–6. doi: 10.1016/j.neuroimage.2009.08.020. [DOI] [PubMed] [Google Scholar]
  • 39.Wessels AM, Rombouts SA, Simsek S, Kuijer JP, Kostense PJ, Barkhof F, Scheltens P, Snoek FJ, Heine RJ. Microvascular disease in type 1 diabetes alters brain activation: a functional magnetic resonance imaging study. Diabetes. 2006;55:334–40. doi: 10.2337/diabetes.55.02.06.db05-0680. [DOI] [PubMed] [Google Scholar]
  • 40.Dickerson BC, Salat DH, Greve DN, Chua EF, Rand-Giovannetti E, Rentz DM, Bertram L, Mullin K, Tanzi RE, Blacker D, Albert MS, Sperling RA. Increased hippocampal activation in mild cognitive impairment compared to normal aging and AD. Neurology. 2005;65:404–11. doi: 10.1212/01.wnl.0000171450.97464.49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Sperling R. Functional MRI studies of associative encoding in normal aging, mild cognitive impairment, and Alzheimer's disease. Ann N Y Acad Sci. 2007;1097:146–55. doi: 10.1196/annals.1379.009. [DOI] [PubMed] [Google Scholar]
  • 42.Dickerson BC, Salat DH, Bates JF, Atiya M, Killiany RJ, Greve DN, Dale AM, Stern CE, Blacker D, Albert MS, Sperling RA. Medial temporal lobe function and structure in mild cognitive impairment. Ann Neurol. 2004;56:27–35. doi: 10.1002/ana.20163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Hamalainen A, Pihlajamaki M, Tanila H, Hanninen T, Niskanen E, Tervo S, Karjalainen PA, Vanninen RL, Soininen H. Increased fMRI responses during encoding in mild cognitive impairment. Neurobiol Aging. 2007;28:1889–903. doi: 10.1016/j.neurobiolaging.2006.08.008. [DOI] [PubMed] [Google Scholar]
  • 44.Miller SL, Fenstermacher E, Bates J, Blacker D, Sperling RA, Dickerson BC. Hippocampal activation in adults with mild cognitive impairment predicts subsequent cognitive decline. J Neurol Neurosurg Psychiatry. 2008;79:630–5. doi: 10.1136/jnnp.2007.124149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.O'Brien JL, O'Keefe KM, LaViolette PS, DeLuca AN, Blacker D, Dickerson BC, Sperling RA. Longitudinal fMRI in elderly reveals loss of hippocampal activation with clinical decline. Neurology. 2010;74:1969–76. doi: 10.1212/WNL.0b013e3181e3966e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Ryan JP, Fine DF, Rosano C. Type 2 diabetes and cognitive impairment: contributions from neuroimaging. J Geriatr Psychiatry Neurol. 2014;27:47–55. doi: 10.1177/0891988713516543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Gruetter R, Ugurbil K, Seaquist ER. Effect of acute hyperglycemia on visual cortical activation as measured by functional MRI. J Neurosci Res. 2000;62:279–85. doi: 10.1002/1097-4547(20001015)62:2<279::AID-JNR12>3.0.CO;2-3. [DOI] [PubMed] [Google Scholar]
  • 48.Ebady SA, Arami MA, Shafigh MH. Investigation on the relationship between diabetes mellitus type 2 and cognitive impairment. Diabetes Res Clin Pract. 2008;82:305–9. doi: 10.1016/j.diabres.2008.08.020. [DOI] [PubMed] [Google Scholar]
  • 49.Cukierman-Yaffe T, Gerstein HC, Williamson JD, Lazar RM, Lovato L, Miller ME, Coker LH, Murray A, Sullivan MD, Marcovina SM, Launer LJ. Relationship between baseline glycemic control and cognitive function in individuals with type 2 diabetes and other cardiovascular risk factors: the action to control cardiovascular risk in diabetes-memory in diabetes (ACCORD-MIND) trial. Diabetes Care. 2009;32:221–6. doi: 10.2337/dc08-1153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Ishizawa KT, Kumano H, Sato A, Sakura H, Iwamoto Y. Decreased response inhibition in middle-aged male patients with type 2 diabetes. Biopsychosoc Med. 2010;4:1. doi: 10.1186/1751-0759-4-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci. 2001;24:167–202. doi: 10.1146/annurev.neuro.24.1.167. [DOI] [PubMed] [Google Scholar]
  • 52.Smith EE, Jonides J. Storage and executive processes in the frontal lobes. Science. 1999;283:1657–61. doi: 10.1126/science.283.5408.1657. [DOI] [PubMed] [Google Scholar]
  • 53.Bruehl H, Sweat V, Tirsi A, Shah B, Convit A. Obese Adolescents with Type 2 Diabetes Mellitus Have Hippocampal and Frontal Lobe Volume Reductions. Neurosci Med. 2011;2:34–42. doi: 10.4236/nm.2011.21005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Strachan MW, Reynolds RM, Frier BM, Mitchell RJ, Price JF. The relationship between type 2 diabetes and dementia. Br Med Bull. 2008;88:131–46. doi: 10.1093/bmb/ldn042. [DOI] [PubMed] [Google Scholar]
  • 55.Gjedde A, Crone C. Blood-brain glucose transfer: repression in chronic hyperglycemia. Science. 1981;214:456–7. doi: 10.1126/science.7027439. [DOI] [PubMed] [Google Scholar]
  • 56.Duckrow RB, Beard DC, Brennan RW. Regional cerebral blood flow decreases during chronic and acute hyperglycemia. Stroke. 1987;18:52–8. doi: 10.1161/01.str.18.1.52. [DOI] [PubMed] [Google Scholar]
  • 57.Jakobsen J, Nedergaard M, Aarslew-Jensen M, Diemer NH. Regional brain glucose metabolism and blood flow in streptozocin-induced diabetic rats. Diabetes. 1990;39:437–40. doi: 10.2337/diab.39.4.437. [DOI] [PubMed] [Google Scholar]
  • 58.Gispen WH, Biessels GJ. Cognition and synaptic plasticity in diabetes mellitus. Trends Neurosci. 2000;23:542–9. doi: 10.1016/s0166-2236(00)01656-8. [DOI] [PubMed] [Google Scholar]
  • 59.Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature. 2001;414:813–20. doi: 10.1038/414813a. [DOI] [PubMed] [Google Scholar]
  • 60.Zhao B, Pan BS, Shen SW, Sun X, Hou ZZ, Yan R, Sun FY. Diabetes-induced central neuritic dystrophy and cognitive deficits are associated with the formation of oligomeric reticulon-3 via oxidative stress. J Biol Chem. 2013;288:15590–9. doi: 10.1074/jbc.M112.440784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Stranahan AM, Arumugam TV, Cutler RG, Lee K, Egan JM, Mattson MP. Diabetes impairs hippocampal function through glucocorticoid-mediated effects on new and mature neurons. Nat Neurosci. 2008;11:309–17. doi: 10.1038/nn2055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Risk of dementia in diabetes mellitus: a systematic review. Lancet Neurol. 2006;5:64–74. doi: 10.1016/S1474-4422(05)70284-2. [DOI] [PubMed] [Google Scholar]
  • 63.Brundel M, Kappelle LJ, Biessels GJ. Brain imaging in type 2 diabetes. Eur Neuropsychopharmacol. 2014 doi: 10.1016/j.euroneuro.2014.01.023. Epub ahead of print. [DOI] [PubMed] [Google Scholar]
  • 64.Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O'Brien JT, Barkhof F, Benavente OR, Black SE, Brayne C, Breteler M, Chabriat H, Decarli C, de Leeuw FE, Doubal F, Duering M, Fox NC, Greenberg S, Hachinski V, Kilimann I, Mok V, Oostenbrugge R, Pantoni L, Speck O, Stephan BC, Teipel S, Viswanathan A, Werring D, Chen C, Smith C, van Buchem M, Norrving B, Gorelick PB, Dichgans M nEuroimaging STfRVco. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 2013;12:822–38. doi: 10.1016/S1474-4422(13)70124-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplmental document 1
NIHMS611134-supplement-Supp_1.doc (37.5KB, doc)
Supplemental Figure 1
NIHMS611134-supplement-Supp_Fig_1.bmp (1.9MB, bmp)
Supplementla Figure 2
NIHMS611134-supplement-Supp_Fig_2.bmp (3.3MB, bmp)

RESOURCES