Diabetes Volume 64, January 2015

279

Chris Moran,1,2 Gerald Mnch,3,4 Josephine M. Forbes,5,6 Richard Beare,1,2,7 Leigh Blizzard,8
Alison J. Venn,8 Thanh G. Phan,1,2 Jian Chen,1,2,7 and Velandai Srikanth1,2,8

Type 2 Diabetes, Skin

Autouorescence, and
Brain Atrophy
Diabetes 2015;64:279283 | DOI: 10.2337/db14-0506

associated with the excessive accumulation of advanced

glycation end products (AGEs) in tissues (3). AGEs are
products of nonenzymatic reactions between reactive carbonyl groups of compounds (such as glucose) with proteins, lipids, or nucleic acids (4). There is a large body of
evidence from in vitro model research supporting a role
for AGEs in neurodegeneration and Alzheimer disease
(AD) (4). Greater serum levels of AGEs are associated
with cognitive decline (5) and lower gray matter volume
(GMV) in older people (4). Given these observations, it
is possible that AGEs play a mechanistic role in T2DMrelated brain atrophy. However, there have been no previous studies examining the role of AGEs in T2DM-related
brain atrophy.
Tissue AGEs can be measured reproducibly and noninvasively in the skin by means of a specialized light emitter
and detector. Skin autouorescence (SAF) measured in this
manner has been shown to be highly correlated with
biopsy-derived skin AGE concentrations (6). We hypothesized that SAF levels would either mediate or modify the
association between T2DM and brain atrophy.

Type 2 diabetes mellitus (T2DM) is associated with an

increased risk of incident cognitive impairment and
dementia (1). Brain atrophy may be a key driver of
T2DM-related cognitive dysfunction (2). T2DM is also

RESEARCH DESIGN AND METHODS

1Stroke and Ageing Research Group, Department of Medicine, Southern Clinical

School, Monash University, Melbourne, Victoria, Australia
2 Neurosciences, Monash Medical Centre, Monash Health, Melbourne,
Australia
3 Department of Pharmacology, School of Medicine, University of Western
Sydney, New South Wales, Australia
4 Molecular Medicine Research Group, University of Western Sydney, New
South Wales, Australia
5 Translational Research Institute, Mater University of Queensland, Brisbane,
Queensland, Australia
6 Mater Clinical School, University of Queensland, Brisbane, Queensland,
Australia

Sampling

We used a cross-sectional study design, and sampling

methods have been described previously (2). Participants

7 Developmental

Imaging, Murdoch Childrens Research Institute, Melbourne,

Australia
8 Menzies Research Institute Tasmania, Hobart, Tasmania, Australia
Corresponding author: Velandai Srikanth, [email protected].
Received 28 March 2014 and accepted 15 July 2014.
This article contains Supplementary Data online at http://diabetes
.diabetesjournals.org/lookup/suppl/doi:10.2337/db14-0506/-/DC1.
2015 by the American Diabetes Association. Readers may use this article as
long as the work is properly cited, the use is educational and not for prot, and
the work is not altered.

COMPLICATIONS

Type 2 diabetes mellitus (T2DM) is associated with brain

atrophy, but the mechanisms underlying this link are
unknown. Advanced glycation end products (AGEs)
accumulate in T2DM, resulting in inammation, oxidative stress, and protein cross-linking, which are known
contributors to neurodegeneration. We aimed to study
whether tissue AGE accumulation is associated with
T2DM-related brain atrophy. We performed brain magnetic resonance imaging, cognitive tests, and noninvasive skin autouorescence (SAF; a measure of tissue
AGE levels) on people aged >55 years with and without
T2DM. Multivariable linear regression was used to study
the relationships among T2DM, SAF, and gray matter
volume (GMV). There were 486 people included in the
study. T2DM was associated with greater SAF. Greater
SAF, T2DM, and cognitive impairment were each associated with lower GMV independently of age, sex, and
total intracranial volume. SAF partially mediated the association between T2DM and GMV. Longitudinal studies
may help conrm whether tissue AGE accumulation is
associated with brain atrophy in T2DM.

280

Diabetes, SAF, and Brain Atrophy

Diabetes Volume 64, January 2015

were included from two studies: the Cognition and Diabetes in Older Tasmanians study (CDOT) and the Tasmanian Study of Cognition and Gait. Participants with
T2DM aged $55 years were recruited into CDOT between
January 2008 and January 2010 using the National Diabetes Service Scheme database as a sampling frame. The
Tasmanian Study of Cognition and Gait sample was
recruited by mailing approach letters to eligible registrants aged $55 years, living in the same Southern Tasmanian postcodes as those in the CDOT study, and has
been described previously (4). The phenotype of T2DM
was based on self-report and conrmed using a single
plasma glucose level according to standard criteria (fasting plasma glucose $7.0 mmol/L, random plasma glucose
$11.1 mmol/L, and HbA1c .6.5% [48 mmol/mol]). People living in a nursing home, those with insufcient English for cognitive testing, or contraindication to magnetic
resonance imaging (MRI) were excluded. The Southern
Tasmanian Health and Medical Human Research Ethics
Committee and the Monash University Human Research
Ethics Committee approved the study, and we obtained
written, informed consent.

maps of gray and white matter. Tissue maps were

smoothed using an isotropic 8-mm Gaussian kernel. A
single expert manually segmented both hippocampi using
established methods known to have high test-retest reliability in our laboratory (intraclass correlation coefcient 0.97) (8). Tissue volumes of the segmented areas
(total gray, white matter, and hippocampal) were calculated using standard voxel-counting algorithms.

Measurements

Data Analysis

SAF

The analyses were conducted on a complete dataset

consisting of those in whom both measures of SAF and
brain imaging were available.
Logistic regression was used to describe the associations of T2DM and brain atrophy with cognitive impairment. Linear regression was used to estimate the
associations of SAF and T2DM with measures of brain
atrophy. Covariates for age, sex, total intracranial volume
(TICV), and other variables were added to the regression
models for brain atrophy if their inclusion produced
a statistically signicant increase in model t or changed
the coefcient of the covariate for T2DM by .10%. Putative factors considered were hypertension (dened as
mean BP .140/90 mmHg or previous diagnosis), ever
smoked tobacco, creatinine, mean steps per day, history
of ischemic heart disease, stroke, hyperlipidemia, BMI and
waist-to-hip ratio, and the use of specic medications that
have been shown to inuence AGE levels (pravastatin,
irbesartan, and metformin) (911). We then examined
whether SAF mediated the associations estimated between T2DM and brain atrophy. For this, we entered
SAF into the model relating T2DM to brain volume outcome measures adjusting for age, sex, smoking, serum
creatinine, and TICV. If the introduction of SAF substantially attenuated the regression coefcient of the binary
covariate for T2DM, and the coefcient of SAF remained
largely unchanged from its value without T2DM in the
model, it was considered a potential mediator. We also
investigated any modifying effect (interaction) of SAF using a test of signicance of the coefcient of a covariate
formed as the product of the covariates for T2DM and
SAF. Statistical analyses were carried out using STATA
version 11.1 (StataCorp, College Station, TX).

We used the AGE reader (DiagnOptics BV, Groningen, the

Netherlands) to measure SAF. The spectrometer reader uses
a light source to illuminate ;4 cm2 of skin on the volar
surface of the right arm 10 cm below the elbow fold. SAF is
calculated as the ratio between the emission light and
reected excitation light, multiplied by 100 and expressed
in arbitrary units. In our laboratory, the test-retest reliability for SAF was high (intraclass correlation coefcient 0.93;
n = 11) when individuals were measured 5 days apart.
MRI Scans

MRI scans were obtained using a single 1.5T General

Electric scanner with the following sequences: highresolution T1-weighted spoiled gradient echo (GRE; TR
35 ms; TE 7 ms; ip angle 35; eld of view 24 cm; 120
contiguous slices; and isotropic voxel size 1 mm3); T2weighted fast spin echo (repetition time [TR], 4,300 ms;
echo time [TE], 120 ms; number of excitations, 1; turbo
factor, 48; and voxel size, 0.90 3 0.90 3 3 mm); uid
attenuated inversion recovery (TR, 8,802 ms; TE, 130 ms;
TI, 2,200 ms; and voxel size, 0.50 3 0.50 3 3 mm); and
GRE (TR, 0.8 ms; TE, 0.015 ms; ip angle, 30; and voxel
size, 0.9 3 0.9 3 7 mm).
Brain Volumes

Three-dimensional T1 and axial GRE sequences were

registered into standard Montreal Neurological Institute
space using Functional Magnetic Resonance Imaging of the
Brains Linear Image Registration Tool. A multispectral
segmentation process was applied using three-dimensional
T1 and GRE sequences using Statistical Parametric Mapping software version 5 (7) to produce tissue probability

Other Measurements

Standardized questionnaires were used to record demographic and clinical information. Weight, height, waist
and hip circumferences, habitual physical activity using
a pedometer worn over 1 week, and blood pressure (BP) in
a sitting position as an average of three recordings from
the right arm were measured and BMI calculated. A
standardized cognitive battery was used to test domains
of memory, speed, and executive and visuospatial
function (Supplementary Table 1) as described previously (2). Diagnosis of cognitive impairment was
assigned, blinded to T2DM status, if function in any of
the domains was ,1.5 SDs from age-, sex-, and educationadjusted norms.

diabetes.diabetesjournals.org

Moran and Associates

RESULTS

There were 285 people with T2DM (mean age 67.5 years,
SD 6.9) and 201 in the non-T2DM comparison group
(mean age 73.4 years, SD 6.9) with SAF measures. A total
of seven participants had inaccurate measures of SAF and
were excluded from the analysis. Summary measures of
the characteristics of each group are presented in Table 1.
Comparisons of the characteristics of people with and
those without T2DM are presented in Supplementary
Table 2.
Associations of SAF With Study Factors

Greater SAF was associated with greater age (b = 0.014;

P , 0.001), but not with sex (b = 0.07; P = 0.26). After
adjustment for age and sex, T2DM was associated with
greater SAF (b = 0.14; P = 0.007). In the whole sample
(T2DM and non-T2DM), greater BMI (P , 0.001), less
habitual physical activity (P = 0.009), and greater serum
creatinine (P = 0.006) were individually associated with
greater SAF. Among those with T2DM, greater SAF was
associated with greater HbA1c (P , 0.001) and longer
duration of T2DM (P = 0.016). Among those without
T2DM, there was no association between SAF and HbA1c.
T2DM was associated with lower GMV (standardized
b = 20.020; P = 0.05), but not with total hippocampal
volume (standardized b = 0.03; P = 0.51) or white matter
volume (standardized b = 0.002; P = 0.90). Lower GMV
was signicantly associated with the risk of any cognitive

Table 1Sample characteristics

Mean (SD) or N (%)
total (n = 486)
Age (years)

69.9 (7.5)

Female sex

208 (43)

Diabetes

285 (59)

Formal education (years)

11.3 (3.7)

Self-reported history of hypertension

or mean systolic BP .140 or mean
diastolic BP .90 mmHg

374 (77)

Use of BP-lowering medications

284 (58)

Statin use

223 (46)

Ischemic heart disease

81 (17)

TIA or stroke

32 (7)

Hyperlipidemia

153 (31)

Ever smoked

253 (52)

BMI (kg/m2)
Normal (BMI 2025)
Overweight (BMI 2530)
Obese (BMI .30)

29.1 (5.0)
86 (18)
216 (44)
177 (36)

Mean steps per day

Serum creatinine (mmol/L)

6,337 (3,372)
78.5 (24.7)

Any cognitive impairment

146 (30%)

SAF (AUs)

2.05 (0.53)

AU, arbitrary unit; TIA, transient ischemic attack.

281

impairment (b = 20.03; CI 20.04 to 20.01; P = 0.005).

In the whole sample, greater levels of SAF were signicantly associated with the risk of any cognitive impairment (b = 0.41; CI 0.010.82; P = 0.05) and with lower
GMV (standardized b = 20.036; P , 0.001), but not
with hippocampal volume (standardized b = 20.046;
P = 0.258) or white matter volume (standardized b =
0.019; P = 0.096) independent of age, sex, smoking, serum creatinine, and TICV. Addition of SAF attenuated the
association between T2DM and GMV by 20%, rendering
it nonsignicant (standardized b = 20.016; P = 0.12),
whereas SAF remained independently associated with
GMV in the model (standardized b = 20.034; P ,
0.001). Additional adjustments for BMI, HbA1c, or duration of T2DM did not change these relationships (data
not shown). There was no interaction between T2DM and
SAF in explaining GMV. Fig. 1 shows the scatter plots of
the association between SAF and GMV stratied by T2DM
status.
DISCUSSION

This is the rst study examining the relationship among

tissue AGE accumulation, T2DM, and GMV. T2DM was
associated with greater accumulation of tissue AGEs (as
measured by SAF) and with lower GMV. Consistent with
our previous study of circulating AGEs (4), we found that
greater SAF was independently and modestly associated
with lower GMV, but additionally demonstrate that SAF
may partially mediate the association between T2DM and
lower GMV. The associations we describe are novel and
provide a solid basis for further studying the relationship
between tissue AGE accumulation and brain atrophy.
There is strong evidence from basic science research
that tissue AGE accumulation plays a role in the pathogenesis of dementia (12). In the case of AD (the most
common type of dementia), autopsy studies have shown
the process of atrophy is due to the accumulation of extracellular amyloid plaque and intracellular tau neurobrillary
tangles (13). Greater levels of AGEs have been found to be
colocated with amyloid plaques (14,15) and paired helical
lament tau in sporadic AD (16) and may act by stabilizing
plaques and promoting brillation of tau through protein
cross-linking (17,18). We speculate that SAF may reect
AGE-mediated cross-linking of other cellular proteins,
such as in neurons.
AGEs may also be directly cytotoxic to neurons in
culture (19) and able to directly induce inammation and
oxidation (12) by binding with receptor for AGE (RAGE)
in mitochondria, generating free radicals, and reducing
clearance of pre-existing reactive oxygen species (17). Furthermore, RAGE interacts with serum b-amyloid, increasing the transport of b-amyloid across the bloodbrain
barrier, activating proinammatory cytokines, and reducing cerebral blood ow (20).
Strengths of our study include the large sample size,
reproducible and sensitive measures of tissue AGEs and

282

Diabetes, SAF, and Brain Atrophy

Diabetes Volume 64, January 2015

Figure 1The association of SAF with GMV. The raw data points of the association of SAF and GMV stratied by T2DM. The tted lines
show the stratied associations between SAF and GMV adjusted for age, sex, smoking, serum creatinine, and TICV. In those without
T2DM: SAF b = 24.78; standardized b = 20.41; P = 0.003; and adjusted R2 = 0.97. In those with T2DM: SAF b = 23.60; standardized b =
20.03; P = 0.02; and adjusted R2 = 0.96 .The high R2 values reect the adjustment for TICV (head size), which is collinear with GMV. The
tted lines for the two groups also overlap considerably, demonstrating a lack of interaction between T2DM and SAF in explaining GMV.

brain structure, clear denition of T2DM, and careful

statistical modeling. We carefully adjusted as required for
smoking, renal function, BMI, hypertension, and hyperlipidemia that may be related to both AGEs and brain or
vascular health. Although medications used to treat these
conditions (pravastatin and irbesartan) (9,10) and specic
antidiabetes drugs such as metformin (11) have been postulated to have a protective effect against the effects of
AGEs, adjusting for the use of these medications did not
change our ndings (data not shown). Our study has
some limitations. Our study is cross-sectional, limiting
inferences of causality, and needs to be conrmed in longitudinal analyses. The AGE reader does not measure
AGEs that do not exhibit autouorescence (nonuorophores) and may also measure non-AGE uorophores
(6). However, the results of a number of earlier studies
support the use of SAF as a surrogate marker of uorescent and nonuorescent AGE content in the skin (6,21).
Given the modest strength of b coefcients for SAF
with GMV, it is likely that AGE accumulation is only one
of a large number of pathways that contribute to the development of dementia in T2DM, explaining why SAF
only partially mediated the T2DMGMV relationship.
The clinical relevance of these results is uncertain, but
they support further research to understand the role of
AGEs in the pathogenesis of dementia in relation to
T2DM and overall. Prospective studies are needed to assess if tissue AGE accumulation is causally related to brain
atrophy in T2DM and, subsequently, to study whether
limiting AGE accumulation may slow neurodegeneration.

Funding. This study was funded by the National Health and Medical Research
Council (NHMRC; project grants 403000 and 436797), Australia. C.M. is a recipient of an Alzheimers Australia Dementia Research Foundation Scholarship

and a Monash Health Clinical Academic Fellowship. G.M., R.B., and T.G.P. are
recipients of NHMRC project grants. J.M.F. and A.J.V. are recipients of an NHMRC
fellowship and NHMRC project grants. L.B. is supported by an NHMRC Career
Development Fellowship and is a recipient of NHMRC project grants. V.S. is
a recipient of a National Heart Foundation/NHMRC Career Development Fellowship and NHMRC project grants.
Duality of Interest. No potential conicts of interest relevant to this article
were reported.
Author Contributions. C.M. drafted and revised the manuscript and
performed statistical analysis and data interpretation. G.M. and J.M.F. contributed to discussion and reviewed the manuscript. R.B. performed image analysis
and interpreted data. L.B. contributed to analysis and interpretation of the data
and reviewed the manuscript. A.J.V. contributed to interpretation of data and
reviewed the manuscript. T.G.P. supervised image analysis, contributed to discussion, and reviewed the manuscript. J.C. performed image analysis and interpreted data. V.S. developed the study concept and design, performed analysis
and interpretation of data, revised the manuscript, and obtained funding. V.S. is
the guarantor of this work and, as such, had full access to all the data in the
study and takes responsibility for the integrity of the data and the accuracy of the
data analysis.

References
1. Biessels GJ, Staekenborg S, Brunner E, Brayne C, Scheltens P. Risk of
dementia in diabetes mellitus: a systematic review. Lancet Neurol 2006;5:6474
2. Moran C, Phan TG, Chen J, et al. Brain atrophy in type 2 diabetes: regional
distribution and inuence on cognition. Diabetes Care 2013;36:40364042
3. Bucala R, Cerami A. Advanced glycosylation: chemistry, biology, and implications for diabetes and aging. Adv Pharmacol 1992;23:134
4. Srikanth V, Westcott B, Forbes J, et al. Methylglyoxal, cognitive function and
cerebral atrophy in older people. J Gerontol A Biol Sci Med Sci 2013;68:6873
5. Yaffe K, Lindquist K, Schwartz AV, et al. Advanced glycation end product
level, diabetes, and accelerated cognitive aging. Neurology 2011;77:1351
1356
6. Meerwaldt R, Graaff R, Oomen PH, et al. Simple non-invasive assessment
of advanced glycation endproduct accumulation. Diabetologia 2004;47:1324
1330
7. Ashburner J, Friston KJ. Unied segmentation. Neuroimage 2005;26:839
851

diabetes.diabetesjournals.org

8. Wrench JM, Wilson SJ, Bladin PF, Reutens DC. Hippocampal volume and
depression: insights from epilepsy surgery. J Neurol Neurosurg Psychiatry 2009;
80:539544
9. Ishibashi Y, Yamagishi S, Matsui T, et al. Pravastatin inhibits advanced
glycation end products (AGEs)-induced proximal tubular cell apoptosis and injury
by reducing receptor for AGEs (RAGE) level. Metabolism 2012;61:10671072
10. Matsui T, Yamagishi S, Takeuchi M, Ueda S, Fukami K, Okuda S. Irbesartan
inhibits advanced glycation end product (AGE)-induced proximal tubular cell injury in vitro by suppressing receptor for AGEs (RAGE) expression. Pharmacol Res
2010;61:3439
11. Ishibashi Y, Matsui T, Takeuchi M, Yamagishi S. Benecial effects of
metformin and irbesartan on advanced glycation end products (AGEs)-RAGEinduced proximal tubular cell injury. Pharmacol Res 2012;65:297302
12. Srikanth V, Maczurek A, Phan T, et al. Advanced glycation endproducts and
their receptor RAGE in Alzheimers disease. Neurobiol Aging 2011;32:763777
13. Braak H, Braak E. Neuropathological stageing of Alzheimer-related changes.
Acta Neuropathol 1991;82:239259
14. Vitek MP, Bhattacharya K, Glendening JM, et al. Advanced glycation end
products contribute to amyloidosis in Alzheimer disease. Proc Natl Acad Sci U S A
1994;91:47664770

Moran and Associates

283

15. Smith MA, Taneda S, Richey PL, et al. Advanced Maillard reaction end
products are associated with Alzheimer disease pathology. Proc Natl Acad Sci
U S A 1994;91:57105714
16. Yan SD, Chen X, Schmidt AM, et al. Glycated tau protein in Alzheimer
disease: a mechanism for induction of oxidant stress. Proc Natl Acad Sci U S A
1994;91:77877791
17. Mnch G, Westcott B, Menini T, Gugliucci A. Advanced glycation endproducts and their pathogenic roles in neurological disorders. Amino Acids 2012;
42:12211236
18. Li XH, Lv BL, Xie JZ, Liu J, Zhou XW, Wang JZ. AGEs induce Alzheimer-like
tau pathology and memory decit via RAGE-mediated GSK-3 activation. Neurobiol
Aging 2012;33:14001410
19. Takeuchi M, Bucala R, Suzuki T, et al. Neurotoxicity of advanced glycation endproducts for cultured cortical neurons. J Neuropathol Exp Neurol 2000;59:10941105
20. Deane R, Du Yan S, Submamaryan RK, et al. RAGE mediates amyloid-beta
peptide transport across the blood-brain barrier and accumulation in brain. Nat
Med 2003;9:907913
21. Meerwaldt R, Hartog JW, Graaff R, et al. Skin autouorescence, a measure
of cumulative metabolic stress and advanced glycation end products, predicts
mortality in hemodialysis patients. J Am Soc Nephrol 2005;16:36873693

SUPPLEMENTARY DATA
Supplementary Table 1. Cognitive domains and tests.
Cognitive Domain
Memory factor
Visuospatial and planning factor
Speed factor
Executive function

Tests
Hopkins Immediate Verbal Recall
Hopkins Delayed Verbal Recall
Hopkins Recognition
Rey Complex Figure Copy Task
Rey Complex Figure Recall Task
Digit Symbol Coding
Digit Symbol Search
COWAT Word test
COWAT Category Test
Digit Span Recall
Stroop Dot Time
Stroop Word Time
Stroop Color time

2014 American Diabetes Association. Published online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db14-0506/-/DC1

SUPPLEMENTARY DATA
Supplementary Table 2. Sample characteristics by diabetes status.

Age (years)
Female sex
Formal education (years)
Self reported history of hypertension or
mean SBP >140 or mean DBP >90
mmHg
Use of blood pressure lowering
medications
Statin use
Ischemic Heart Disease
TIA or Stroke
Hyperlipidemia
Ever smoked
BMI (kg/m2)
Normal (BMI 20-25)
Overweight (BMI 25-30)
Obese (BMI>30)
Mean steps per day
Fasting blood glucose (mmol/l)
HbA1c (%)
(mmol/mol)
Serum creatinine (mol/L)
Age at diabetes diagnosis (years)
Median duration of T2DM (years)
Use of oral glucose lowering
medications
Use of insulin
Any cognitive impairment
Skin autofluorescence (AU)

T2DM
n=285
67.5 (6.9)
41% (117/285)
11.5 (3.4)
88% (251/285)

No T2DM
n=201
73.4 (6.9)
45% (91/201)
11.2 (3.9)
72% (144/201)

<0.001
0.35
0.40
<0.001

70% (199/285)

43% (86/201)

<0.001

62% (176/285)
19% (54/285)
8% (24/285)
48% (138/285)
53% (152/285)
30.5 (5.1)
11% (31/285)
41% (117/285)
47% (135/285)
6377 (3660)
7.7 (2.3)
7.2 (1.2)
55
78.7 (25)
57.5 (11.3)
7 (4-12)
62% (176/285)

23% (47/201)
13% (27/201)
4% (8/201)
7% (15/201)
51% (102/201)
27.2 (4.2)
27% (55/201)
50% (100/201)
21% (42/201)
6414 (3014)
5.3 (0.5)
5.6 (0.3)
38
78.4 (24)

<0.001
0.11
0.053
<0.001
0.60
<0.001
<0.001
0.048
<0.001
0.91
<0.001
<0.001

39 (19)
2.07 (0.51)

<0.001
0.49

4% (10/285)
107 (38)
2.03 (0.54)

P value

0.91

2014 American Diabetes Association. Published online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db14-0506/-/DC1