ARTICLE
Auteur(s) : Farina Hashmi1,
James Malone-Lee1, Elizabeth Hounsell2
1Department of Medicine, University College London,
London, United Kingdom
2School of Biological and Chemical Sciences, Birkbeck
University of London, United Kingdom
accepté le 9 Septembre 2005
Despite the increasing awareness of the long-term complications
associated with the feet in diabetes, the statistics for diabetic
foot ulceration, in particular, are still distressing [1, 2]. The
multifarious pathologies that predispose the feet to ulceration,
and consequent lower limb amputation make the clinical management
of the diabetic foot a health issue of grave importance [3].In Type
2 diabetes mellitus (T2DM) plantar callus is an early sign of
potential ulceration [4] particularly as neuropathy, elevated
plantar pressures [5] (possibly secondary to limited sub-talar
joint and 1st metatarsophalangeal joint range of motion
[6, 7]), vascular insufficiency and trophic changes add to the
risk. This is supported by the observation that neuropathic ulcers
commonly form on the plantar aspect of the forefoot [8]. The
evidence connecting elevated plantar pressures with callus
development and ulceration in the insensate foot is contradictory
[8-11]. This renders uncertain the use of direct pressure
measurements as a predictive tool for plantar ulceration. The
damage to plantar skin as a result of callus is described as being
akin to that of a foreign body and considered a preferable
indicator of plantar ulceration than pressure measurements [4, 12].
This school of thought proposes that the pressure required to
produce callus may vary between individuals, but once formed, the
local response of the skin to focal pressures ultimately leads to
ulceration. The differences may be due to the biochemical and
physical nature of pedal skin in the diabetic state, which could
also influence the propagation of callus tissue. This is pertinent
as diabetes is accompanied by widespread biochemical and functional
abnormalities in connective tissues such as dermal and tendon
collagen [13, 14], some of which are analogous to changes observed
in aging, like the reduction in tissue elasticity [13]. These data
and empirical observations suggest that there is a variation in the
texture of callus in the diabetic foot from dry to macerated, but
invariably it is thick and bulky, particularly around the periphery
of a neuropathic ulcer.Cheiroarthropathy and its association with
thick, tight and waxy skin on the dorsum of the hands is an
established clinical presentation of Type 1 and 2 diabetes mellitus
[15-17]. This collection of pathologies has been attributed to the
non-enzymatic glycation (NEG) of tendon and skin collagen proteins
[18, 19]. Non-enzymatic glycation is a Schiff base reaction
involving the attachment of the aldehyde group of a sugar (e.g.
glucose) to the primary amine groups of the protein, forming an
intermediate Amadori product (AP) – a marker for early glycation
(e.g. fructose-lysine) [20]. The acid hydrolysed fructose-lysine
can be assayed using high performance liquid chromatography (HPLC).
With time the AP undergoes a series of sequential reactions
resulting in the formation of advanced glycation end-products
(AGEs) [20]. The AGEs form covalent cross-links with adjacent
protein strands causing a reduction in the tissue mobility [13,
21].The yellow hue observed in palmar skin and nails of patients
with diabetes is attributed to the accumulation of AGEs [22].
Yellow nails on the feet are also seen but no colouration is seen
on plantar skin, with the exception of yellow callus. Studies
measuring the amount of frustose-lysine in hair [23], nails [24]
and plantar SC [25, 26] hypothesised that the glycation of these
keratin rich proteins is subject to the same laws applicable to
proteins of other tissues, therefore suggesting a possible
relationship between the NEG of plantar epidermal proteins and
glycaemic control. The methods used in these studies were
semiquantative and further explorations into the extent of
consequent AGE formation were not made. In addition, no
investigations into plantar skin mechanics were attempted. The NEG
of epidermal proteins formed during the differentiation process
should, like all other proteins studied so far, be related to
glucose levels over that period.The observation of thickened skin
in diabetes may range from clinically inapparent, but measurable,
increases in skin thickness unassociated with symptoms, to
clinically apparent thickening of skin involving the fingers, hands
and upper back region [27]. Some methods used for determining skin
thickness, such as pinching the skin, give a purely qualitative
estimation [28]. Studies that have used high frequency ultrasound
imaging have reported thicker skin in diabetics compared to
non-diabetics, the greatest differences seen in the thigh and not
the foot (dorsal region), suggesting that the increased skin
thickness is not necessarily uniform in diabetics [29]. In
addition, the full skin thickness measurements made from the
ultrasound images are not accurate as the dermal/hypodermal
boundary is not clear. The boundary between the epidermis and the
dermis is well defined, so that epidermal thickness measurements
can be made with greater accuracy. No studies of this kind have
been carried out on plantar skin of people with diabetes where the
majority of neuropathic ulcers form. Only one published study has
investigated plantar skin, in a small number of non-diabetic
volunteers, using ultrasound imaging [30]. It has only been
assumed, therefore, that the skin of the lower extremities in
diabetes is thicker.This study aimed to measure the accumulation of
the early glycation adduct, furosine, and the AGE, pentosidine in
plantar epidermal proteins, along with the in vivo measurement of
plantar epidermal thickness and flexibility. The glycation products
were quantified by the adaptation of existing HPLC techniques [31]
that proved to be specific and sensitive. The viscoelastic
properties of the skin were determined by the application of
vertical, negative pressure onto the surface of the plantar skin
using the Cutometer® 580 MPA (Courage & Khazaka Ltd,
Denmark). Plantar epidermal thickness was measured from images
captured using the Dermascan® C Ver, 3, ultrasound
scanner (Dermascan C®, Cortex Technology, Denmark).
Materials and methods
Unless otherwise stated, all chemicals were purchased from
Sigma-Aldrich Company Ltd, Dorset UK).
Patients
The study group of 103 people above the age of 18 with T2DM (58
males and 45 females) was recruited from the Diabetes Unit and the
Diabetic Eye Screening Unit at the Whittington hospital, London UK.
All the patients attending the hospital were from one health care
district in North London. The median age was 66 years with an
interquartile range (IQR) of 12 years. The mean diabetes duration
was 6.4 ± 7.0 years. The control group consisted of 87 adults
without diabetes (38 males and 49 females) who were recruited by a
poster and leaflet campaign in a number of the out-patient units at
the Whittington hospital. The median age of the controls was 57
years (IQR: 16 years).
All volunteers were excluded from the study on presentation of
the following: the presence of diffuse peripheral neuropathy other
than diabetic neuropathy (e.g. malignancy, alcohol abuse, drug
abuse, anaemia, vitamin B12 deficiency or untreated
hypothyroidism); significant neurological disorders other than
diabetic polyneuropathy (e.g. stroke with significant neurological
deficit, transient ischaemic attacks, multiple sclerosis, epilepsy,
dementia and paraplegia); previous or present treatment with
cytotoxic drugs and/;or radiotherapy; drug medications that may
affect the autonomic nervous system and epidermal keratin
metabolism; individuals with systemic disorders likely to affect
epidermal cell kinetics (e.g. eczema, psoriasis and
scleroderma).
All subjects gave written and verbal informed consent. The Local
Research Ethics Committee approved the study. One investigator
carried out all the physical examinations. Examinations of the
lower limb extremity were carried out on both legs. All
experimental data was collected from the left foot of each
participant.
SC collection
The SC specimens were taken from the plantar metatarsal area using
a scalpel. Areas involved with sepsis and specimens contaminated
with blood were avoided. Each specimen was weighed and stored at
– 70 °C until assayed.
Protein extraction from SC specimens
The protein extraction used in this study was adapted from previous
work carried out in this field [32]. In brief, callus samples were
incubated in urea, sodium dodecyl sulfate (SDS) and
β-mercaptoethanol for 12 h at 50 °C. The mixture was
dialysed against degassed water until a clear, colourless solution
remained. Immunoblots of the sodium dodecyl sulfate polyacrylamide
gel electrophoresis (SDS-PAGE) separated proteins were probed with
anti-keratin monoclonal antibodies AE1 and AE3 (Chemicon
International, Temecula, CA). The majority of the proteins were
identified as being epidermal keratins with trace amounts of other
proteins.
Preparation of protein extracts for HPLC analysis
The protein extracts (0.2 mg) were hydrolysed in HCl (200 μl, 6 M),
at 110 °C under nitrogen for 6 h for the furosine assay and
18 h for the pentosidine assay. The hydrolysates were
lyophilised over night to dryness and redissolved in trifluoracetic
acid (TFA, 0.1% aqu). The protein concentration of each hydrolysate
was adjusted to a 1 mg/ml solution, using the Lowry protein assay
(BioRad Laboratories Ltd, Hertfordshire, UK), in preparation for
HPLC analysis.
Preparation of blood serum for HPLC analysis
Venous blood (5 ml) was collected from each volunteer at the same
time as callus was sampled. Serum proteins were precipitated out
using cool trichloracetic acid (2.25 ml, 6% w/v). The protein
pellet was hydrolysed in HCl (200 μl, 6 M) under nitrogen at
110 °C for 24 h. The sample was lyophilised over night
and then reconstituted in 0.1% TFA (aq) to a concentration of 1
mg/ml protein content in preparation for injection into the HPLC
system for the detection of furosine and pentosidine.
Synthesis of standards for HPLC assays
Furosine standard was purchased from Neosystem Groupe SNPE
(Strasbourg, France). Pentosidine standard was synthesised from the
method described by Henle et al. [33]. In brief,
Nα-acetyl-L-arginine (3.243 g, 15 mmol) and
Nα-acetyl-L-lysine (941 mg, 5 mmol) were dissolved
in phosphate buffer at 65 °C for 48 h under continuous
stirring. During this time 10 portions (75 mg each) of D-ribose
were periodically added to the reaction mixture. The mixture was
hydrolysed in vacuum-sealed glass vials using HCl (30 ml, 12 M) at
110 °C for 2 h. The resulting solution was reduced under
low pressure at 60 °C and dissolved in distilled water (20 ml). The
pentosidine was purified using cation exchange chromatography and
preparative HPLC.
HPLC assay for furosine and pentosidine in epidermal and serum
protein hydrolysates
Hydrolysates were analysed by reversed phase HPLC (equipment:
Waters Ltd, Hertfordshire, UK), with gradient elution using aqueous
TFA (0.1%) as the primary mobile phase and TFA (0.1%) in
acetonitrile as the second mobile phase. Furosine was eluted at ~
7 min on a C18 column, 100 mm × 4.6 mm, particle size
7 μm (Hypersil®, Thermo Life Sciences Ltd, UK) with
a UV absorbance of 280 nm. Pentosidine was eluted at ~ 10 min
on a reverse phase C18 BDS column, 250 mm × 4.6 mm, particle
size 5 μm (Hypersil®, Thermo Life Sciences Ltd, UK)
at ≈ 10 min at fluorescence excitation wavelength of 335 nm
and emission wavelength of 385 nm.
Measurement of the mechanical properties of plantar skin, in
vivo, using vertically applied negative pressure
The Cutometer® 580 MPA consists of a hand held probe
that is placed on the surface of the skin. A constant vacuum
pressure (500 mbar) is generated encouraging the skin to rise into
the aperture (2 mm in diameter) within the probe. The diameter
of the probe ensures minimal dermal involvement, therefore the
measurements made mainly reflect the epidermal mechanics. An
optical sensor, housed in the probe, measures the elevation of the
skin, which is displayed by the software as a displacement – time
curve (60 s with the application of pressure followed by
60 s without). It is from this curve that the five indices of
elasticity were recorded (( figure 1 )): 1) series
elastic element on stretching – Ue, 2) series elastic element on
retraction – Ur, 3) time constant on stretching – Uv1,
4) time constants on retraction – Uv2, and 5) plasticity
– Ud/Uf. Plasticity was defined as the ratio of the final
displacement reading (at 120 s) to the maximum displacement
reading (at 60 s).
The displacement curves during and after the removal of pressure
were fitted to the Kelvin model of viscoelasticity, which
represented a step plus an exponential curve fit. The exponential
curves fits for the rising and falling portions of the curves are
described by equations 1 and 2, respectively.
Where: Uv1 = distance between the end of the rising
elastic stretch and the displacement at 60 s, Uv2 =
distance between the end of the falling elastic stretch and the
displacement at 120 s t = time and τ = time constant (( figure 1 )).
All volunteers were seen in the morning between 9 am and 12
noon. Before the readings were taken the volunteer lay in a supine
position on the bed for approximately 15 min. The data were
collected from the plantar aspect of the 3rd
metatarsophalangeal joint area of each volunteer. The probe was
applied perpendicular to the skin for every reading and care was
taken not to apply any external pressure. The room temperature and
humidity were noted for every set of readings taken.
Measurement of the thickness of plantar epidermis, in vivo,
using high frequency ultrasound imaging
The images were captured before the elasticity data were collected,
on the same site and with the patient in the same position. Care
was taken to maintain the probe perpendicular to the skin surface
during scanning and to minimise the pressure of the transducer on
the skin surface to ensure a satisfactory scan. The same swept gain
was used for every image captured. The images comprised of
different tissues with different echogenicity. The intensity of the
reflected echoes (which is different for different tissue types)
were evaluated by the microprocessor, and visualised as coloured
two-dimensional images (B-mode). Quantitative data were collected
from A-mode images. The Dermascan C® scanner, used in
this study, presented an average skin thickness measurement based
on 224 A-mode scans, therefore minimising any variations in the
tissues on the site being investigated. Images typical of the skin
on three sites on the foot are described in ( figure 2 ).
Statistical analysis
Application of mathematical functions, graphical representation of
data and curve fitting were all performed using Kaleidagraph™, for
Windows, version 3.5 (Synergy Software (PCS Inc.)) and
Microsoft® Excel 2000, version 5.0 (Microsoft
Corporation). All statistical analyses were carried out by SPSS for
Windows (SPSS Inc. 1989-2001). Values in the text were quoted as
the median, minimum and maximum values and interquartile range
(IQR), where the range is between the 25th and
75th percentiles. Comparisons of the skewed distributed
variables between T2DM and control groups were analysed by the
Mann-Whitney test. The null hypothesis was rejected when p ≤ 0.05.
Associations between data sets were determined by Spearman
correlation analysis.
Results
HPLC assay for furosine and pentosidine in epidermal and serum
protein hydrolysates (table 1)
( Table 1 )There was no significant
difference in the concentration of furosine between the two groups
(p = 0.503, 95% CI 102.5 μg/mg protein, 182.6 μg/mg protein). There
was a greater concentration of furosine than pentosidine in callus
collected from the test and control groups. There was a
significantly greater concentration of pentosidine in diabetic
plantar callus than controls (p = 0.001, 95% CI 29.95 μg/mg
protein, 71.04 μg/mg protein) (( figure 3 )).
There was no correlation between furosine in plantar callus and
furosine in blood serum (T2DM r = – 0.115, controls r =
0.084). A similar observation was recorded for pentosidine
concentrations (T2DM r = – 0.023, controls r = 0.136). There
was no association between furosine and pentosidine concentrations
and age in both groups (T2DM furosine r = 0.088; T2DM pentosidine r
= 0.095; controls furosine r = – 0.193; controls pentosidine r
= – 0.143). There was no correlation between the HbA1c values
and the accumulation of furosine and pentosidine in diabetic
plantar keratin (furosine r = 0.107, pentosidine r = 0.158).
Table 1 Concentrations of furosine and pentosidine
determined by HPLC assays of epidermal and serum proteins (values
depicted as zero represent trace peaks on the chromatograms that
could not be resolved)
|
Furosine [ng mg-1 of protein]
|
Pentosidine [ng mg-1 of protein]
|
|
(n = 103)
|
(n = 87)
|
|
SC protein
|
Serum protein
|
SC protein
|
Serum protein
|
|
T2DM
|
Median
|
507.3
|
99.2
|
72.8
|
119.6
|
|
Max.
|
1690.5
|
2526.1
|
848.3
|
1780.2
|
|
Min.
|
0.0
|
0.0
|
0.0
|
0.0
|
|
IQR
|
589.1
|
268.6
|
171.3
|
260.2
|
|
Controls
|
Median
|
471.2
|
122.7
|
10.5
|
25.6
|
|
Max.
|
6331.6
|
580.5
|
485.3
|
134.5
|
|
Min.
|
0.0
|
0.0
|
0.0
|
0.0
|
|
IQR
|
930.1
|
110.2
|
54.1
|
1569.4
|
Measurement of the mechanical properties of plantar skin, in
vivo, using vertically applied negative pressure (table 2)
( Table 2 )The series elastic element on
retraction (i.e. after the negative pressure was released) of the
epidermis was significantly greater in the T2DM group than the
controls (p = 0.04, 95% CI 0.287 μm, 0.173 μm) (( figure 4 )). The
difference remained significant once the data were corrected
according to the epidermal thickness (p = 0.001).
There was no significant difference in the time constants on
stretching (p = 0.800, 95% CI 10.6 s, 6.7 s) and
retraction (p = 0.547, 95% CI 10.1 s, 6.3 s) between the
T2DM and control groups (( figure 5 )). The epidermal
plasticity of non-diabetic plantar epidermis was significantly
greater than that of diabetic plantar epidermis (p = 0.007, 95% CI
13.23, 3.43, ( figure
6 )).
Table 2 In vivo viscoelasticity data of plantar skin in
response to the application and release of vertically applied
negative pressure (values depicted as zero, for the times constants
and series elastic elements, represent curves with negligible
gradients that could not be manipulated).
|
|
T2DM (n = 103)
|
Controls (n = 87)
|
|
Series elastic element on stretching [μm]
|
Median
|
34
|
26
|
|
Max
|
280
|
607
|
|
Min
|
0
|
0
|
|
IQR
|
59
|
89
|
|
Time constant on stretching [s]
|
Median
|
98.3
|
5.4
|
|
Max
|
61.3
|
45.4
|
|
Min
|
0.0
|
0.0
|
|
IQR
|
10.4
|
10.9
|
|
Series elastic element on retraction [μm]
|
Median
|
21
|
17
|
|
Max
|
250
|
319
|
|
Min
|
0
|
0
|
|
IQR
|
48
|
24
|
|
Time constant on retraction [s]
|
Median
|
7.3
|
7.9
|
|
Max
|
41.3
|
32.2
|
|
Min
|
0.0
|
0.0
|
|
IQR
|
12.0
|
13.1
|
|
Plasticity
|
Median
|
0.02
|
0.03
|
|
Max
|
1.00
|
1.00
|
|
Min
|
0.00
|
0.00
|
|
IQR
|
0.07
|
0.05
|
Measurement of the thickness of plantar epidermis, in vivo,
using high frequency (20 MHz) ultrasound imaging
The plantar epidermal images were significantly thicker (i.e. they
had greater echogenicity) in the T2DM group in comparison to
controls (p = 0.017, 95% CI 0.19 mm, 0.21 mm, ( figure 7 )).
There was no association between epidermal echogenicity on the
feet and the duration of diabetes (r = – 0.133) or glycaemic
control (r = – 0.017). The plantar epidermal thickness of the
healthy volunteers was significantly greater than the T2DM patients
without neuropathy (p = 0.024).
There was no association between age and epidermal flexibility
and echogenicity in both groups (p ≥ 0.05). When comparing the
images between the sexes, female non-diabetic images had a thicker
appearance than those taken from the male controls.
Discussion
The primary aim of these investigations was to test for an
association between epidermal mechanics and the accumulation of
glycation products, principally furosine and pentosidine within the
superficial layers of the plantar SC. The accurate quantification
of these glycation products was successful. The results have
implications for the physical behaviour of plantar skin in
diabetes.
Glycation
The published data in the field of diabetes and NEG of proteins,
over the past two decades has made a valid contribution towards a
clearer understanding of the origins of the secondary complications
of diabetes. As glycation potentially involves any protein, many
tissues have been investigated. Blood tissue was used for
pioneering studies in this field, for its ease of access and for
the dynamic nature of its constituent proteins in relation to
glucose during the hyperglycaemic and normoglycaemic states. The
pursuit for therapeutic interventions, to inhibit the glycation of
human proteins, still exists today and is complicated by the
multifarious nature of glycation products. The evolution of this
area of biochemistry, aided by the acceptance of the heterogeneous
nature of these modified proteins, has entered the 21st
century with a less blinkered view of the singular influence that
glycation adducts have on the progression of tissue pathology in
diabetes. During the latter part of the 1990s a shift towards the
investigation of reactive oxygen species (ROS) by the mitochondrial
electron transfer chain provided a connection between ROS and the
four hypotheses that explain cell pathology in diabetes, i.e.
biochemical pathways triggered by hyperglycaemia: increased polyol
pathway flux; activation of protein kinase C isoforms; increased
hexoseamine pathway and increased accumulation of AGEs [34]. Free
radicals also accelerate the formation of AGEs, which in turn
supply more free radicals [35]. Other proteins that are modified by
glycation include low-density lipoproteins, collagen [36], lens
crystallins [37], and peripheral nerve proteins [38]. This mosaic
of interrelated processes explains pathologies not only seen in
diabetes, but in rheumatoid arthritis, Alzheimer’s disease and
cardiovascular disease.
For a short period of time different investigators explored the
measurement of furosine, in plantar callus tissue, hair and nails.
These groups concentrated their efforts on measuring the amount of
furosine in these keratin rich tissues, with a vision of using the
concentration of furosine as an index for hyperglycaemia. It seems
surprising that further scrutiny for the presence of AGEs in these
tissues was not attempted, as the confirmation of the presence of
an early glycation product suggests a potential for AGE formation.
The local effects of protein glycation on the physical properties
of the skin were not considered. However, in these studies the sole
aim was to develop alternative ways of determining glycaemic
control. Different investigators recognised the implication that
protein glycation may have on the mechanical properties of tissues
such as tendon and dermal collagen, but a detailed investigation of
the skin on the feet was not pursued, hence the origins of this
series of studies.
The points of NEG, studied in these experiments, were the
formation of furosine, which exclusively involves the terminal
amino group of a lysine residue; and pentosidine involving both
lysine and arginine residues. From the genesis of the keratin
polypeptide chain through to the complex filament network
formation, there are arginine and lysine residues present at all
points along the protein filament, which are potentially
susceptible to NEG. This study is unique in identifying the AGE
pentosidine in plantar SC. The concentration of pentosidine was
significantly greater in the samples taken from patients with T2DM
than healthy controls and even though there was no significant
difference in furosine concentration between the two groups a
greater concentration of furosine was measured in comparison to the
pentosidine in both groups. The range of concentrations of furosine
and pentosidine in the T2DM group was greater than those recorded
in the non-diabetic group. Several hypotheses as to why a greater
amount of furosine is present in this tissue can be proposed. If
the furosine is exclusively converted to pentosidine then the
corneocytes desquamate before the process of conversion from the
early to late glycation products is complete. This is likely in
plantar skin in diabetes where an increase in the epidermal transit
time, as a result of mechanical stimulation, may produce
corneocytes in a hyperglycaemic environment. This could explain the
high levels of furosine. Glucose utilisation is increased with an
increase in the epidermal cell proliferation. Plantar epidermal
tissue, as a result of mechanical stimulation has a high cell
turnover rate and the utilisation of glucose by keratinocytes is
increased commensurately. In a hyperglycaemic environment, a
surplus of glucose may be a substrate for glycation of all the
proteins expressed by the keratinocyte cells.
Furosine could be the intermediate compound of the conversion to
other AGEs in addition to pentosidine. This conforms to the
existing knowledge that the modifications of proteins into
different AGEs originate from furosine, it is therefore reasonable
to suggest that a fraction of furosine generated in the epidermis
is converted to pentosidine and the remaining furosine is modified
to form other AGEs, an example of one being CML. The presence of
the AGE carboxymethyllysine was detected (but not quantified) in a
subgroup (n = 12) from this study (unpublished data).
It is possible that the sugar attachment could be occurring at
different times and different points along the keratin protein
chain, therefore at the point when the tissue is removed from the
foot there could be furosine and pentosidine formations along the
length of the same keratin chain due to the sugar modification
initiated at different points in time.
All these hypotheses are feasible, however without knowing the
full complement of AGEs in the same tissue, it is impossible to
surmise the kinetics of the protein glycation.
The measurement of a significantly greater amount of pentosidine
in the SC of T2DM subjects than controls introduces the importance
of the clinical relevance of this data. It is well established that
the accumulation of AGEs correlates with the age of the tissue
being investigated. The accumulation of AGEs in human tissue
proteins is related to the turnover rate of those tissues: the
longer the lifespan of the individual, the greater the potential
for the increased accumulation of AGEs. It could be that this
specialised area of skin, with its unique epidermal kinetics, needs
the focus of glycation redirected to the age of the tissue as
opposed to the age of the person. In this case the epidermal cells
(keratinocytes) have an epidermal transit time of approximately 28
days. This value applies to human volar skin and is subject to
variation on different sites of the body. In the case of plantar
skin the epidermal kinetics are disturbed so as to adapt to the
pressures of gait. The perturbation encourages an increased rate of
cell renewal and decreased desquamation, hence the thickening of
the stratum corneum. The accumulation of pentosidine may be
governed by the imbalance of cell renewal and clearance on the
surface of the skin.
Associations between the concentration of furosine and
pentosidine in plantar epidermal proteins and circulating blood
serum proteins and glycaemic control
This study demonstrated no correlation between the glycated adducts
in the blood and those in plantar proteins. There was also no
correlation between glycated epidermal proteins and HbA1c levels.
This renders uncertain the hypothesis that glycaemic control may
affect plantar SC thickening (that is callus formation) in
diabetes. The observation that there was no association between
glycation of the keratin proteins and HbA1c levels, could be due to
hyperglycaemic “memory”. This phenomenon has been seen in
microvasculature, where there is a persistent progression of
hyperglycaemia induced alterations during the subsequent periods of
normal glucose homeostasis [34]. Animal and human studies have
demonstrated the development of retinopathy long after the glucose
levels were normalised [39]. This hyperglycaemic “memory” is
explained by the production of mitochondrial superoxides, at
physiological concentrations of glucose, with the resulting
continued activation of the pathways in the absence of
hyperglycaemia. The half-life of haemoglobin tissue (~ 120
days) is approximately three to four times longer than the
epidermal turnover time, therefore one would expect a relatively
greater potential for the accumulation of AGEs in the blood serum
protein than plantar epidermal keratins. This was reflected in the
pentosidine data, where the ratio of pentosidine concentration in
the callus to that measured in the blood serum was 1:3 in T2DM and
1:2 in the healthy subjects. The reverse was recorded with the
furosine data where its concentration in callus was greater than in
blood serum (diabetics 4:1; non-diabetics 7:1), again possibly
reflecting the hyperproliferative state of the callus tissue and
the different rates of AGE formation in different tissue types. It
would be valuable to investigate the superoxide activity of
keratinocytes as well as exploring how post-prandial hyperglycaemia
and glucose excursions might activate the mechanisms that, cause
carbonyl and oxidative stress secondary to PKC activation [34]. It
would be pertinent to probe for these markers in pathological
callus in diabetes and callus associated with neuropathic
ulceration. It would also be of interest to identify whether
specific keratins in pathological callus are glycated.
Mechanical properties
The physical changes in SC tissue on removal from the foot are
profound, notably dehydration and shrinkage, thus making the in
vitro exploration of the tissue impossible. The evaluation of the
viscoelasticity of plantar skin in vivo in relation to vertical,
negative pressure has been successfully conducted in this study
using the Cutometer® MPA 580.
The mechanical properties of the epidermis result from its
geometric form and the intrinsic properties of its constituent
materials, mainly elastic, high modulus keratin fibres embedded in
a viscoelastic matrix of a lower modulus. Keratins in the epidermis
provide the skin with the resilience and flexibility that is
essential during natural joint and muscle movement. This is of
particular importance on the foot, where the skin lies closely
associated with muscle and bone tissue. The plantar epidermis is
the first point of contact with the ground (or shoe) and is
designed to withstand shear, compression and torsion stresses [40].
The function of keratin filaments in epidermal cells is to impart
mechanical integrity to the cells [41].
The plantar skin of the T2DM patients, although more elastic, it
was also less plastic than the skin of healthy controls. This means
that the epidermis has the ability to retract in diabetes more
readily than in the non-diabetic state. It is possible that the
realignment of the tissues, after the deforming force is released
and the skin has been retracted, takes longer in diabetes.
Andreassen et al. (1981) [42] demonstrated profound alterations in
the collagen of diabetic rat skin, leading to an increase in the
stiffness and strength. The stiffness is characterised by
additional glycated cross-links formed between the collagen
molecules along the whole extent of the triple helix, thus reducing
the ability to change the fibre orientation in this way, i.e.
reducing the ability to stretch under stress. This experimental
model could be transferable to epidermal tissue in vivo. Therefore,
if there is a greater number of cross-links between keratin
filaments, the ability to lengthen when under stress decreases.
Whether this is caused by cross-linking via AGEs is not clear. The
contribution of lipids and water to the viscoelastic properties of
plantar skin cannot be ignored.
A key contributor to adjustments of epidermal mechanics is
dehydration causing a reduction in the interaction between keratin,
non-helical regions and water extractable materials (e.g. free
amino acids) between the keratin fibres, reducing the skin’s
elasticity [43]. In diabetes, autonomic neuropathy results in
clinically anhydrotic skin on the foot (72% of diabetics in this
study). With less water in the stratum corneum, the elastic and
viscoelastic properties of the epidermis will be affected. The
protruding amino terminal domains may also be subject to glycation,
further disturbing the dynamics between the keratin filaments.
Ultrasonography
The low acoustic density of skin often creates problems for skin
thickness measurements, as it is difficult to identify the
interface between the dermis and deeper structures [44]. The images
taken of the plantar skin in this study did not show the dermal
structures clearly. This is advantageous as it makes the
epidermal-dermal boundary definition clearer. The plantar epidermal
images presented with a greater echogenicity in the T2DM patients
in comparison to controls, implying a denser plantar epidermis in
diabetes. The equipment provides images of skin that are the
nearest possible to the in vivo representation of the skin on
different body sites. The images provide a more accurate indication
of tissue thickness than histological techniques that ultimately
result in post-fixation deformities in the structures.
The plantar epidermal thickness of the healthy volunteers was
significantly greater than the T2DM patients without neuropathy,
suggesting that the disease process with regards to thickening of
the tissues has occurred before the clinical manifestations of
neuropathy are evident.
These physical properties of the plantar epidermis in diabetes
would place the skin at risk of damage. The samples of callus taken
from the patients with diabetes were not different in clinical
appearance to callus sampled from non-diabetic volunteers, although
the biochemical analysis did show a difference.
Conclusion
In this study the plantar epidermis in T2DM was thicker, with a
decreased plasticity, than that of healthy controls. The SC had
greater pentosidine concentrations in people with T2DM compared
with controls.
Glycation of plantar epidermal proteins could play an important
role in the stiffening of the plantar epidermis. A direct
connection between the glycation of plantar epidermal proteins and
epidermal mechanics in diabetes has not been elucidated, though
this relationship is implied.
It would be of benefit to explore the formation of pentosidine
and other AGEs and reactive oxygen species in differentiating
keratinocytes, as well as assessing contributions by lipids, water
and other protein structures present in the epidermis. It would
also be useful to scrutinise the accumulation of glycation adducts
within tissues associated with plantar ulcers in T2DM. Work is
being conducted at the moment in this area. If keratin glycation
can be identified as being such a culprit for ulceration,
therapeutic options could be targeted in preventing this
progression.
Acknowledgements
The authors would like to thank all the people who volunteered for
the study.
The study was supported by the ‘Dr William M. Scholl Podiatric
Research and Development Fund – UK’.
The synthesis of pentosidine and CML was carried out with the
help of Professor Paul Thornalley and colleagues (Department of
Biochemistry, University of Essex, UK).
References
1 Pecoraro RE, Reiber GE, Burgess EM. Pathways of
diabetic limb amputation. Basis of prevention. Diabetes Care 1990;
13(5): 513-21.
2 Reiber GE, Vileikyte L, Boyko EJ, et al.
Causal pathways for incident lower – extremity ulcers in patients
with diabetes from two settings. Diabetes Care 1999; 22(1):
157-62.
3 Edmonds M, Boulton A, Buckenam T, et al.
Report of the Diabetic Foot and Amputation Group. Diabet Med 1996;
13: S27-S42.
4 Murray HJ, Young MJ, Hollis S,
Boulton AJM. The association between callus formation, high
pressures and neuropathy in diabetic foot ulceration. Diabet Med
1996; 13: 979-82.
5 Bennett PJ, Stocks AE, Whittam DJ. Analysis of
risk factors for neuropathic foot ulceration in diabetes mellitus.
Am Podiatr Med Assoc 1996; 86(3): 112-6.
6 Birke JA, Franks BD, Foto JG. First ray joint
limitation, pressure and ulceration of the first metatarsal head in
diabetes. Foot Ankle 1995; 16(5): 277-84.
7 Fernando DJS, Masson EA, Veves A,
Boulton AJM. Relationship of limited joint mobility to
abnormal foot pressures and diabetic foot ulceration. Diabetes Care
1998; 14(1): 8-11.
8 Sims DS, Cavanagh PR, Ulbrecht JS. Risk factors
in the diabetic foot. Recognition and management. Phys Ther 1988;
68(12): 1887-901.
9 Ahroni JH, Boyko EJ, Forsberg RC. Clinical
correlates of plantar pressure among diabetic veterans. Diabetes
Care 1999; 22(6): 965-72.
10 Boulton AJM, Hardisty CA, Betts RP,
et al. Dynamic foot pressure and other studies as diagnostic
and management aids in diabetic neuropathy. Diabetes Care 1983; 6:
26-33.
11 Cavanagh PR, Hennig EM, Rodgers MM,
et al. The measurement of pressure distribution on the plantar
surface of the diabetic foot. In: Whittle M, Harris D,
eds. Biomechanical Measurement in Orthopaedic Practice. Oxford
University Press, 1985: 159-68.
12 Young MJ, Cavanagh PR, Thomas G, et al.
The effect of callus removal on the dynamic plantar foot pressures
in diabetic patients. Diabet Med 1992; 9: 55-7.
13 Andreassen TT, Seyer-Hansen, Oxlund H.
Biomechanical changes in connective tissues induced by experimental
diabetes. Acta Endocrinol (Copenh) 1981; 98: 432-6.
14 Seibold JR, Uitto J, Dorwart BB,
Prockop DJ. Collagen synthesis and collagenase activity in
dermal fibroblasts from patients with diabetes and digital
sclerosis. J Lab Clin Med 1985; 105(6): 664-7.
15 Collier A, Patrick AW, Bell D, et al.
Relationship of skin thickness to duration of diabetes, glycemic
control and diabetic complications in male IDDM patients. Diabetes
Care 1989; 12(5): 309-12.
16 Kennedy L, Archer DB, Campbell SL, et al.
Limited joint mobility in Type I Diabetes Mellitus. Postgrad Med J
1982; 58: 485-6.
17 Fitzcharles MA, Duby S, Waddell RW,
et al. Limitation of joint mobility (cheiroarthropathy) in
adult non-insulin dependent diabetic patients. Ann Rheum Dis 1984;
43(2): 251-4.
18 Buckingham B, Reiser KM. Relationship between the
content of lysyl oxidase-dependent cross links in skin collagen,
nonenzymatic glycosylation and long term complications in Type I
Diabetes Mellitus. J Clin Invest 1990; 86(4): 1046-54.
19 Hamlin RC, Kohn RR, Luschin JH. Apparent
accelerated aging of human collagen in Diabetes Mellitus. Diabetes
1983; 24(10): 902-4.
20 Monnier VM, Fogarty JF, Monnier CS,
Sell DR. Glycation, glycoxidation, and other Maillard reaction
products. In: Byung Pal Y, ed. Methods in Aging Research. CRC
Press, 1999: 657-81.
21 Kennedy L, Baynes JW. Non-enzymatic glycosylation
and the chronic complications of diabetes: an overview.
Dermatologia 1984; 26: 93-8.
22 Melling M, Pfeiler W, Karimian-Teherani D,
et al. Differential scanning calorimetry biochemical, and
biomechanical analysis of human skin from individuals with diabetes
mellitus. Anat Rec 2000; 259: 327-33.
23 Paisey RB, Clamp JR, Kent MJC, et al.
Glycosylation of hair: possible measure of chronic hyperglycaemia.
BMJ 1984; 288: 669-71.
24 Oimomi M, Hatanaka H, Ishikawa K, et al.
Increased fructose-lysine of nail protein in diabetic patients.
Klin Wochenschr 1984; 62: 477-8.
25 Delbridge L, Ellis CS, Robertson K,
Lequesne LP. Non-enzymatic glycosylation of keratin from the
stratum corneum of the diabetic foot. Br J Dermatol 1985; 112:
547-54.
26 Márová I, Záhejský J, Sehnolová H.
Non-enzymatic glycation of epidermal proteins of the stratum
corneum in diabetic patients. Acta Diabetol 1995; 32: 38-43.
27 Seibold JR. Digital sclerosis in children with
Insulin-Dependent Diabetes Mellitus. Arthritis Rheum 1982; 25(11):
1357-61.
28 Grgic A, Rosenbloom AL, Weber FT, et al.
Joint contracture - common manifestation of childhood diabetes
mellitus. J Pediatr 1976; 88(4): 584-8.
29 Huntley AC, Walter Jr. RM. Quantitative
determination of skin thickness in Diabetes Mellitus: relationship
between disease parameters. J Med 1990; 21(5): 257-64.
30 Ryan TJ, Thoolen M, Yang YH. The effect of
mechanical forces (vibration or external Compression) on the dermal
water content of the upper dermis and epidermis, assessed by high
frequency ultrasound. J Tissue Viability 2001; 11(3): 97-101.
31 Schleicher E, Wieland OH. Specific quantitation by
HPLC of protein (lysine) bound glucose in human serum albumin and
other glycosylated proteins. J Clin Chem Clin Biochem 1981; 19:
81-7.
32 Steinert PM. Structure, function, and dynamics of
keratin intermediate filaments. J Invest Dermatol 1993; 100(6):
729-34.
33 Henle T, Schwarzenbolz U, Klostermeyer H.
Detection and quantification of pentosidine in foods. Z Lebensm
Unters Forsch 1997; 204: 95-8.
34 Brownlee M. Biochemistry and molecular cell biology of
diabetic complications. Nature 2001; 414: 813-20.
35 Baynes JW. Role of oxidative stress in development of
complications in diabetes. Diabetes 1991; 40(4): 405-12.
36 Dyer DG, Dunn JA, Thorpe SR, et al.
Accumulation of Maillard reaction products in skin collagen in
diabetes and aging. J Clin Invest 1993; 91(6): 2463-9.
37 Lyons TJ, Silvestri G, Dunn JA, et al.
Role of glycation in modification of lens crystallins in diabetic
and nondiabetic senile cataracts. Diabetes 1991; 40(8): 1010-5.
38 Ryle C, Leow CK, Donaghy M. Nonenzymatic
glycation of peripheral and central nervous system proteins in
experimental diabetes mellitus. Muscle Nerve 1997; 20(5):
577-84.
39 Monnier VM, Bautista O, Kenny D, et al.
Skin collagen glycation, glycoxidation and crosslinknig are lower
in subjects with long-term intensive versus conventional therapy of
type I diabetes. Diabetes 1999; 48: 870-80.
40 Edwards C, Marks R. Evaluation of biomechanical
properties of human skin. Clin Dermatol 1995; 13: 375-80.
41 Fuchs E. Keratins and the skin. Annu Rev Cell Dev Biol
1995; 11(123): 153.
42 Andreassen TT, Seyer H, Oxlund H.
Biomechnaical changes in connective tissues induced by experimental
diabetes. Acta Endocrinol (Copenh) 1981; 98: 432-6.
43 Jokura Y, Ishikawa K, Tokuda H, et al.
Molecular analysis of elastic properties of the stratum corneum by
solid - state 13C - nuclear magnetic resonance
spectroscopy. J Invest Dermatol 1995; 104(5): 806-12.
44 Overgaard Olsen L, Takiwaki H, Serup J.
High-frequency ultrasound characterization of normal skin. Skin
thickness and echographic density of 22 anatomical sites. Skin Res
Technol 1995; 1: 74-80.
|
Printable version
Version PDF
