Abstract
Objective Advanced glycation end products (AGEs) and the interaction with their receptors (RAGE) play an important role in the pathogenesis of diabetic foot (DF) associated with diabetic neuropathy. Our study examined the association between asymmetric dimethyl arginine (ADMA), fructosamine, nitric oxide (NO), and soluble (s) RAGE levels in serum of diabetic patients with and without neuropathy.
Methods Circulating levels of ADMA, fructosamine, NO, and sRAGE, estimated either chemically or by enzyme-linked immunosorbent assay, were examined in 60 type 2 diabetes mellitus (T2DM) overweight/obese (body mass index, 30.5 ± 1.5 kg/m2) male patients and 20 age-matched (55 ± 3 years) obese healthy subjects as control group. The T2DM subjects were categorized as patients without DF (n = 30), and the remaining were patients with DF associated with neuropathy.
Results First sRAGE levels were significantly increased in T2DM patients without DF in comparison to healthy controls (1656.6 [1198.8–2065.4] vs 1111.7 [909–1605.3] pg/mL, respectively; P < 0.05). However, in the DF group (1049.6 [783.7–1221.8] pg/mL), its level decreased significantly in comparison to both groups (P < 0.05). However, ADMA and fructosamine were significantly higher in diabetic patients with DF than both T2DM without DF and healthy controls. Moreover, NO was significantly lower in DF than in diabetic patients without DF and controls (5 ± 0.4 and 8 ± 0.4 vs 42 ± 2.5 μmol/L, respectively; P < 0.05). Finally, sRAGE levels were significantly correlated with ADMA, fructosamine, and NO.
Conclusions Soluble forms of the receptor for advanced glycation end product could be an endogenous protection factor against occurrence of DF, hence may be of therapeutic value in the treatment of DF.
One of the “diabetes-related causes” is neuropathy, a common complication of diabetes, in which nerves are damaged as a result of hyperglycemia.1Diabetic foot (DF) is defined as ulceration, infection, and gangrene infection of the foot associated with diabetic neuropathy (DN) and is the result of the complex interaction of different factors.2Approximately 50% of diabetics may eventually develop nerve damage.3Moreover, depending on affected nerves, symptoms of DN can range from pain and numbness in extremities to digestive system, urinary tract, blood vessels, and heart problems.4Pathogenic mechanisms linking hyperglycemia with DN include increased polyol pathway flux, protein kinase C activation, increased hexosamine pathway flux, enhanced formation of reactive oxygen species, increased production of vascular endothelial growth factor, and altered generation of endothelial nitric oxide (NO).5
The neurological examination currently used is designed to identify loss of protective sensation rather than early neuropathy.6Hence, this study aims to explore the mechanisms and consequences of DF complications related to this disorder in Egypt, to address the best diagnostic and/or prognostic test(s)/marker(s).
As we stated earlier, the hypothesis receiving considerable interest in diabetes mellitus (DM) is the role of protein glycation.7The result of nonenzymatic glycoxidation of macromolecules is the formation of complex, heterogeneous molecules of advanced glycation end products (AGEs).8Soluble forms of the receptor for advanced glycation end products (sRAGEs) were previously shown to appear to be associated with vascular risk factors in DM and the metabolic syndrome.9Soluble forms of the receptor for advanced glycation end products consists of an endogenous splice variant of RAGE (esRAGE) lacking the transmembrane domain of the receptor as well as proteolytically cleaved forms shed into the bloodstream by action of extracellular metalloproteinases.10Both sRAGE and esRAGE were shown to act as decoys binding inflammatory RAGE ligands like AGEs. It is speculated that sRAGE might counteract inflammatory reflexes triggered by AGEs.9However, it seems questionable that the circulating forms of RAGE exert a biological effect because the sRAGE concentrations found in plasma are approximately 1000 times lower than needed for the binding of AGEs.10Nevertheless, associations of sRAGE and esRAGE with different aspects of metabolic, vascular, and autoimmune disease might make them valuable risk markers,11as well as being vulnerable targets for therapy.
Moreover, NO and asymmetric dimethylarginine (ADMA), an endogenous inhibitor of NO synthase (NOS),12need to be explored if both may take part in DF development. The current study will test if sRAGE is related to clinical factors leading to deterioration of DF associated with neuropathy and if its soluble form may act as a naturally occurring inhibitor of the signaling induced by the interaction of AGEs with its cellular receptor. To our knowledge, the interrelation between fructosamine and sRAGE as well as between ADMA and NO in DF has not been examined to date. Since then, recent trials of therapeutic agents and approaches to preventing type 2 diabetes mellitus (T2DM) neuropathic complications have been disappointing. Therefore, highlighting the AGE-RAGE axis role may direct pharmacological research to this axis as a therapeutic target.
SUBJECTS AND METHODS
Subjects
We recruited 80 male participants (age range, 50–72 years). Participants were divided into 3 groups. Their descriptive data are shown in Tables 1 and 2 and Figure 1. Group I is composed of healthy control volunteers (n = 20) (age, 55 ± 3.1 years). Group II is composed of T2DM patients (n = 30) with DM intermediate duration between 4 and 6 years, who did not develop neuropathy as well as not having DF (age, 56 ± 3 years). Group III consisted of T2DM patients (n = 30) with DM duration of 17 years, who developed DF-associated neuropathy (age 55 ± 3.9 years).
All patients were diagnosed by an orthopedic surgeon based on Wagner’s classification together with vibratory perception threshold/temperature discrimination, tests of pinprick sensation, ankle reflex assessment, and monofilament examination, together with a general inspection of skin integrity and musculoskeletal deformities.6
For each participant, a data sheet was completed with the patient’s identification code and demographic data: age and duration of diabetes, as well as the antidiabetic agents used, if any. Exclusion criteria included cancer, other endocrine disorders, and ischemic cerebrovascular disorders, as well as peripheral and coronary ischemic disorders. Moreover, macrovascular or microvascular complications, other than neuropathy in groups II and III were excluded by clinical and physical examination. These diseases were excluded upon the endocrinologist examination and that of independent ophthalmologist and nephrologist unaware of the groups studied and the aim of the study.
The study protocol was approved by the Faculty of Medicine Ethics Committee, Ain Shams University. The study was carried out in accordance with the regulations and recommendations of the Declaration of Helsinki. Informed consent was obtained from all participants.
Demographic Data
A DF case was determined based on the “Wagner classification” according to the following criteria: grade 0: foot clinically normal but with varying degrees of neuropathy and presence of bony deformities of the foot, place it as risk, level 1: superficial ulcer existence, yet it does not affect the subcutaneous tissue (superficial cellulitis); grade 2: deep ulcer uncomplicated, involving the tendon, bone, or capsule, but with the absence of osteomyelitis; grade 3: deep ulcer, complicated, with demonstrations of infectious osteomyelitis, and abscess; grade 4: necrotizing gangrene limited (digital, forefoot, and heel); and grade 5: extensive gangrene.13For more conformation, group 1 (control) subjects also underwent the same sets of tests of incipient neuropathy.
Body mass index (BMI) was calculated as an index of the weight in kilograms divided by the square of the height in meters. Center for Disease Control and Prevention classifies that the reference range of BMI are between 18.5 and 24.9 kg/m2, overweight BMI between 25 and 29.9 kg/m2, and the obese BMI greater than30 kg/m2.14Both the diabetic groups and the control group were selected to have matching BMI (Table 1).
Materials
Venous blood samples, from overnight fasting participants, were collected and divided into 2 portions. The first portion was collected in a plain vacutainer, left to clot for 30 min, and centrifuged (1500×g / 20 min) for separation of the serum.
Biochemical Analysis
Aliquots of the separated serum were stored at −20°C for subsequent determination of fasting blood sugar (FBG)15and lipids (total cholesterol and triacylglycerol) by using standard enzymatic techniques.16High-density lipoprotein cholesterol was determined after precipitation of apolipoprotein B-containing lipoproteins.17However, low-density lipoprotein cholesterol was determined according to the method of Hatch and Lees.18Nitric oxide was determined by colorimetric determination of total nitrate/nitrite in serum according to the method of Green et al.19by the Greiss reaction.
To exclude liver and kidney dysfunction, complete liver function and kidney function tests were done (data not shown). Fructosamine determination was based on the ability of ketoamines to reduce nitro blue tetrazolium to a formazan dye under alkaline conditions.20
Soluble forms of the receptor for advanced glycation end products and ADMA were quantitatively determined in serum by enzyme-linked immunosorbent assay kits provided by BioVendor R&D (Evropska, Brno, Czech Republic).
The second portion of the fasting venous blood (1 mL) was collected over EDTA, as an anticoagulant, (final concentration, 1 mg/mL) for glycated hemoglobin (HbA1c%) determination (Stanbio glycohemoglobin, Boerne, TX).21
Data Analysis
Data are expressed as mean ± SD for quantitative parametric measures in addition to median percentiles for quantitative nonparametric measures (sRAGE). When comparing normally distributed variables between patients and healthy controls, an independent t test was used for comparing means. Comparison between 2 independent parametric data groups was conducted using analysis of variance, followed by least significant difference as post hoc test. For comparison of skewedly distributed variables between the studied groups, median values were calculated, and Mann-Whitney U test was used. Statistical significance was done using the statistical package for social sciences program (version 19; SPSS Inc, Chicago, IL), at P < 0.05, whereas at P < 0.01 and P < 0.001 are highly significant. Pearson correlations coefficient test was applied to study the possible association between each 2 variables among each group for parametric data at P < 0.01 and P < 0.05 (2 tailed).
RESULTS
Demographic data of the studied groups are indicated in Table 1. We detected a high significant elevation of FBG levels in diabetic patients without DF (group II) and diabetic patients with DF-associated neuropathy (group III) by 322.5 % and 404 %, respectively, when compared with healthy controls (group I) at P ⩽ 0.05 (Table 2; Fig. 1). These results were confirmed by the pattern of HbA1c % elevation in groups II and III, which showed a high significant increase by 190% and 213%, respectively, in comparison to group I.
Our study also revealed that diabetic patients without DF and diabetic patients with DF showed a significant decrease, compared with the healthy controls, for serum NO levels by 19% and 12%, respectively, at P ⩽ 0.05 (Table 2, Fig. 1). These latter two results are in accordance to the finding that a significant correlation coefficient was found between serum NO and HbA1c% at P = 0.03, r = −0.38, as shown in Figure 2. Moreover, ADMA was highly significantly elevated in groups II and III, compared with group I, by 139% and 160%, respectively, at P ⩽ 0.05 (Table 2, Fig. 1). Consequently, a negative significant relationship was found between serum ADMA and NO (r = −0.557, P = 0.001) as depicted in Figure 3A. Additionally, the correlation between sRAGE and NO showed a negative significant relationship (r = −0.446 at P = 0.014) as indicated in Figure 3B.
Soluble forms of the receptor for advanced glycation end product levels were significantly decreased in DF patients when compared with patients without DF by 59%, at P ⩽ 0.05 (Table 2; Fig. 1). However, there was a significant increase by 130% in its level in diabetic patients without DF compared with healthy controls. Moreover, there is a significant decrease in sRAGE level in T2DM with DF-associated neuropathy in comparison to healthy controls to 77%.
As depicted in Table 3, the effect of neuropathic grade according to “Wagner classification” on NO, fructosamine, ADMA, and sRAGE serum levels in T2DM with DF-associated neuropathy group (n = 30) indicated that stratification of the measured parameters by the “Wagner classification” showed a significant relation of Wagner grade with diabetes duration (P ⩽ 0.05). However, NO, fructosamine, and ADMA did not differ in the different neuropathic grades. Whereas in grades 3, 4, and 5, NO level was 4.65 ± 2, 5.26 ± 1.96, and 5.6 ± 2.6 μmol/L, respectively. Also, fructosamine level were comparable in these 3 neuropathic grades, with levels of 346.8 ± 35.5, 338.2 ± 43.8, and 344 ± 39 μmol/L, in grade advancing from 3 and 4 to 5, respectively. Moreover, this also was the case for ADMA levels with values of 0.98 ± 0.034 μmol/L in grade 3, 0.99 ± 0.042 μmol/L in grade 4, and 0.97 ± 0.04 μmol/L in grade 5 (P ≥ 0.05).
In contrast, sRAGE levels in the advanced grade 5 (750 [585.4–1249] pg/mL) was significantly lower than its level in either grade 3 (1064 [859–1245.6] pg/mL) or grade 4 (1048.6 [795–1251] pg/mL) (P ⩽ 0.05).
Finally and interestingly, the correlations between sRAGE and either ADMA or fructosamine illustrated a positive significant relationship (r = 0.384 at P = 0.036 and r = 0.475 at P = 0.008, respectively) as shown in Figure 4, A and B.
DISCUSSION
Diabetic foot is one of the most common complications of T2DM and is the most common cause of nontraumatic lower limb amputation resulting in significant disability.22On average, symptoms begin 10 to 20 years after first diabetes diagnosis.23Nerve injuries are caused by decreased blood flow as well as high FBG levels.24Long-term glycemic control is an important predictor of microvascular and macrovascular complications of diabetes.25An increase in intracellular glucose will lead to an increase in the flux of glucose to sorbitol via the polyol pathway, an increase in glucosamine-6- phosphate via the hexosamine pathway, the activation of protein kinase C via de novo synthesis of diacylglycerol, and the increase in AGE formation.26,27In addition, hyperglycemia may result in a decrease of NO production and/or inactivation of NOS by reactive oxygen and nitrogen species.28In diabetics, an endogenous deficiency in NOS enzyme leads to decreased wound NO production and a spectrum of related pathologies, such as impaired cutaneous vasodilatation, decreased neurogenic vascular response, and DN.29,30Our patients with DF showed significantly decreased NO serum level compared with the healthy control group. Nitric oxide is synthesized from its precursor L-arginine by endothelial NOS,31which is competitively inhibited by ADMA, an endogenous compound that is elevated in DM.32Consequently, our study showed that ADMA was highly significantly elevated in diabetic patients. The decreased production of NO during diabetic complications is supposed to be the consequence of its reduced production by NOS and/or inactivation of NOS by increased ADMA level together with hyperglycemia, obesity, hyperinsulinemia, oxidative stress,33produced either by glycosylated proteins or directly from vascular endothelium. High levels of HbA1c% observed in our groups were correlated with NO levels significantly, where fructosamine and glycosylated hemoglobin (HbA1c%) are two glycated proteins commonly used for monitoring diabetics. A single fructosamine measurement indicates the average glucose concentration over the previous 2 to 3 weeks.34Moreover, AGE formation is a result of the nonenzymatic reaction between sugars and free amino groups of proteins. Advanced glycation end products, through interacting with their specific receptor (RAGE), result in activation of proinflammatory states and are involved in numerous pathological situations.35Receptor for advanced glycation end product is a member of the Ig superfamily, encoded in the Class III region of the major histocompatibility complex.36Serum soluble RAGE levels have been found to be decreased in chronic inflammatory diseases, including atherosclerosis,37DM, renal failure,38and the aging process.39This is the condition in our study, where sRAGE expression levels were significantly decreased in diabetic patients with DF-associated neuropathy when compared with diabetic patients without DF. There seem to be three main mechanisms of nerve damage in diabetics. First, excess glucose causes endothelial injury. Endothelial damage is evidenced by the decreased NO. Second, there are changes in activation of various cellular pathways that alter cell function without immediately causing cell death. Third, AGE meets RAGE; time and the accumulation of altered molecules wreak additional havoc on patients’ tissues. It is assumed that microangiopathy caused by hyperglycemia-mediated decrease in NO will have sequelae similar to microangiopathy of various other causes: impaired blood flow to nerves, tissue hypoxia, and oxidative stress.40
Moreover, experimental diabetic neuropathy models recently provided evidence that engagement of RAGE and RAGE-dependent sustained activation of the proinflammatory transcription factor nuclear factor kappa B might significantly contribute to reduced nociception.41
Our results confirm the relation between sRAGE and DF and its deleterious effect on its severity. The duration of diabetes and severity of hyperglycemia are the major risk factors for developing microvascular complications. Once present, duration of diabetes seems to be a less important factor than hyperglycemia for progression from earlier to later stages.
Advanced glycation end product–RAGE axis has a substantial role in mechanisms leading to neuropathy but is unlikely to be the sole factor responsible for progressive neurological damage in diabetes.42
Moreover, an association between serum sRAGE levels and duration of diabetes in T2DM patients was found; hence, the severity of the disease was observed to be associated. Our results confirm the relation between sRAGE and DF and its deleterious effect on its severity. This could be explained as follows: the serum sRAGE does not have a relationship, at least directly, with the events that trigger the onset of diabetes. After the onset of diabetic condition, serum sRAGE level affects the course of development of neuropathy. When the DF patients were subdivided according to “Wagner classification,”13a decreased sRAGE level was noted as the neuropathic grade increased. The sRAGE levels significantly declined as the neuropathy complication advanced from simple (grade 3) and then to complex (grade 5) compared with the diabetic group with no neuropathy. This supports the hypothesis that sRAGE, by limiting the interaction of AGE with cell membrane, can protect vessels against AGE toxicity.43The more likely explanation is that AGEs upregulate RAGE expression, reflected as sRAGE in various tissues/cells9acting as a countermeasure to prevent tissue/cell damage by AGEs. Therefore, fructosamine and RAGE were correlated significantly.
In addition, results indicate that sRAGE can be a useful biomarker to indicate individual variations in susceptibility to DF. From the point of view that AGEs upregulate RAGE expression, therefore serum sRAGE levels were increased significantly in the diabetic group compared with the healthy group. However, in the DF condition, the accelerated rate of AGE protein accumulation makes significant consumption of sRAGE to neutralize and remove them (hence, serum sRAGE level was decreased significantly in the DF group in comparison with the other studied groups). Therefore, in the diabetic condition, the accelerated rate of AGE protein accumulation may be beyond the ability of the body to remove these products.44Thence, identifying RAGE-dependent inflammation as one pathomechanism underlying neuronal dysfunction might provide basis for new therapeutic approaches.
We would like to discuss some strengths of our study; reasonable differences were obtained despite that the study cohort is comparably small with lack of a well-controlled diabetic group, which could be a limitation; however, the current study was adequately powered to detect the differences seen.
In conclusion, measuring circulating sRAGEs may prove to be a valuable biomarker. As the biochemical process of advanced glycation, their receptors seems to be enhanced in the diabetic milieu resulting not only from hyperglycemia but also from decreased NO, obesity, and hyperlipidemia. Our observations highlight the opportunities for further research that would definitively establish AGE-RAGE axis as important investigations of DF.
ACKNOWLEDGMENT
The authors would like to thank Dr. Amr Zidan Nabeh, General Surgery, MSc., MD, Ain Shams University Hospitals, for his assistance in clinical samples collection.