Introduction

The true pathogenetic mechanisms of diabetic nephropathy are not known. However, low-grade inflammation and endothelial dysfunction may be a common denominator of microvascular and macrovascular complications [1–3], and it has been suggested that the activation of the complement system contributes to this cascade of inflammation [4].

The complement system can be activated by three different pathways, of which the lectin pathway probably antedates the classical and alternative pathways [5]. Mannan-binding lectin (MBL) is synthesised by the liver and secreted into the bloodstream [6]. MBL has binding sites for carbohydrates such as mannose and N-acetylglucosamine [6, 7]. If the carbohydrates are present in the correct pattern, e.g., as on the surface of micro-organisms, the binding of MBL will result in activation of the complement by MBL-associated proteases [6]. The serum concentration of MBL is highly variable in humans and the variability is largely due to structural polymorphisms leading to amino acid replacements in the collagen-like region, which decreases MBL assembly and stability but is also under the influence of polymorphisms in the promoter region [8].

Patients with type 1 diabetes and normal AER show increased levels of MBL when compared with healthy subjects; further, the MBL concentration was associated with AER even within the normoalbuminuric range while no correlation between MBL and C-reactive protein (CRP) was observed [4].

There are no studies focusing on MBL in incipient and overt nephropathy, where the levels of the low-grade inflammatory markers CRP and IL-6 are elevated when compared with healthy control subjects [3]. It is worth noting that diabetic nephropathy is associated with insulin resistance [9] as is chronic inflammation [10]. However, whether MBL is linked to insulin resistance in diabetic nephropathy is not known.

Therefore, the aim of the present study was to evaluate whether MBL is associated with diabetic nephropathy in type 1 diabetes, and whether there is an association between MBL, low-grade inflammatory markers and insulin resistance.

Subjects and methods

Subjects

A total of 191 type 1 diabetic patients were selected from the ongoing nationwide, multicentre Finnish Diabetic Nephropathy Study (FinnDiane) and divided into three groups according to their AER from three consecutive overnight or 24-h urine collections as described earlier [3].

Type 1 diabetes was defined as an onset of diabetes before the age of 35 years and permanent insulin treatment initiated within 1 year of diagnosis. Patients with normal AER were not allowed to be taking antihypertensive medication and were not allowed to have signs of cardiovascular disease. Patients with microalbuminuria and macroalbuminuria were required to be having treatment with ACE inhibitors, which may attenuate low-grade inflammation. Of the initial study population of 1,616 type 1 diabetic patients, 401 met these criteria, and from this cohort we randomly selected 191 patients to be representative of a wide range of AER. The duration of diabetes had to be at least 10 years. The ethical committees of all participating centres approved the study protocol. Written informed consent was obtained from each patient and the study was performed in accordance with the guidelines of the Declaration of Helsinki as revised in 2000.

Methods

Data on medication, cardiovascular status, diabetic complications, hypertension and cardiovascular disease were obtained using a standardised questionnaire, which was completed by the patient’s attending physician based upon medical files. Blood pressure was measured twice in the sitting position using a mercury sphygmomanometer after a rest of at least 10 min. Height and weight were recorded, and blood was drawn for the measurement of HbA1c, lipids, creatinine, inflammatory markers and MBL.

HbA1c and lipids were measured by enzymatic methods at the local hospitals. Serum creatinine was determined using a modified Jaffé reaction. As a measure of insulin sensitivity we calculated the estimated glucose disposal rate (eGDR) using an equation developed by Williams et al. [11], modified for HbA1c. To define the severity of renal disease in addition to AER we estimated GFR using the Cockroft–Gault formula [12]. CRP was measured by radioimmunoassay and IL-6 by high-sensitivity enzyme immunoassay as previously described [3]. Serum MBL concentrations were measured by an in-house time-resolved immunofluorometric assay [13]. Intra-assay and interassay coefficients of variation were below 5% and 10% respectively.

Statistical analysis

Data are expressed as means±SD for normally distributed values, and median (range) or median (interquartile range) for non-normally distributed values. Frequencies are given as percentages. Differences between groups for normally distributed variables were tested using ANOVA, and nonparametric data with the Kruskal–Wallis test. Correlations were calculated using the Pearson correlation coefficient. Multiple linear regression analysis with MBL as the dependent variable was used to assess independent relationships. The distribution of triglycerides, creatinine, AER, CRP, IL-6 and MBL were all skewed and therefore logarithmically transformed before being entered into regression analyses. All calculations were performed using a BMDP statistical package (BMDP Statistical Software, Los Angeles, CA, USA). A p value below 0.05 was considered statistically significant.

Results

The clinical characteristics of the patients are shown in Table 1. Since there were no differences in MBL or low-grade inflammatory markers between males and females, the data are pooled.

Table 1 Clinical characteristics of 191 type 1 diabetic patients
Full size table

MBL concentrations are depicted in Fig. 1. MBL increased in a strikingly similar fashion to CRP. Although there was a significant difference in MBL between normoalbuminuric and microalbuminuric patients (p=0.029) as well as between normoalbuminuric and macroalbuminuric patients (p=0.019), there was no difference between patients with microalbuminuria and those with macroalbuminuria.

Fig. 1
figure 1

MBL levels (median with interquartile range) in 191 type 1 diabetic patients without (1,154 μg/l [180–2,202]) and with diabetic nephropathy further divided into incipient (microalbuminuria: 1,713 μg/l [724–2,760]) and overt (macroalbuminuria: 1,648 μg/l [568–3,394]) nephropathy. # p=0.029 normoalbuminuria vs microalbuminuria; ## NS microalbuminuria vs macroalbuminuria; ### p=0.019 normoalbuminuria vs macroalbuminuria

Full size image

In univariate analysis, MBL correlated with eGDR (r=−0.17; p<0.05), HbA1c (r=0.21; p<0.01) and HDL cholesterol (r=−0.19; p<0.05). However, no significant correlations were observed between MBL and the inflammatory markers CRP (r=−0.01; p=0.92) or IL-6 (r=0.10; p=0.17), duration of diabetes, WHR, systolic or diastolic blood pressure, creatinine, GFR, total cholesterol, triglycerides or AER (data not shown). MBL did not correlate with AER in the combined group of patients with microalbuminuria or macroalbuminuria (data not shown). When HbA1c, eGDR and HDL cholesterol were entered into the multiple regression analysis model, HbA1c was the only variable independently associated with MBL (β±SEM: 0.26±0.08; p=0.003).

Discussion

We report increased circulating levels of MBL in type 1 diabetic patients with diabetic nephropathy. There was no relationship between MBL and inflammatory markers or between MBL and insulin resistance. In contrast, HbA1c, a known risk factor for diabetic nephropathy, was independently related to the MBL levels.

The finding that MBL correlated with glycaemic control but not with AER in patients with microalbuminuria and macroalbuminuria contrasts with that shown in patients with normal AER [4]. One possible explanation for these conflicting results may be the difference in HbA1c concentrations between our patient groups. The fact that MBL was not associated with AER is also in line with the finding that MBL was not associated with either IL-6 or CRP in this study. In this context it is worth noting that we have previously shown an association of AER with the inflammatory markers CRP and IL-6 in diabetic patients with incipient or overt nephropathy [3]. One explanation for the lack of correlation between MBL and inflammatory markers could also come from the simple fact that CRP may inhibit the production of MBL [14, 15]. On the other hand, it has been suggested that MBL and CRP are coordinated in the acute phase response, since CRP modulates the cascade in which the complement regulatory protein H regulates MBL-initiated cytolysis [15].

Even if we could not confirm a relationship between MBL and eGDR in the multivariate analysis, there was a negative correlation between MBL and eGDR as well as between MBL and HDL cholesterol in the simple regression analysis, suggesting that MBL may be a marker of insulin resistance. Such a notion is supported by the known positive association between insulin resistance and diabetic nephropathy in type 1 diabetes [9].

In diabetic nephropathy we do not yet know whether MBL plays an active role in the pathogenesis or the progression of already prevalent disease. On the other hand, it has been suggested that MBL plays an unfavourable role in other kidney diseases such as IgA nephropathy and Henoch–Schönlein purpura. MBL binding to IgA results in complement activation, and the deposition of MBL in association with IgA nephropathy is found in the mesangial area of the kidneys [16, 17]. IgA is a heavily glycosylated molecule with more mannose type N-linked glycan chains on the cell surface to be recognised and bound by MBL [18]. Whether MBL is also found in the diabetic kidney is still not known, but it can be speculated that MBL ligands are present in diabetic kidneys, leading to a deposition of MBL in the kidney or even other target organs with deleterious effects. The mechanism could be that hyperglycaemia stimulates the production of N-acetylglucosamine through the hexosamine pathway, and as a consequence an abundance of various secretory and cell membrane glycoproteins are modified by N-linked glycosylation, enabling these proteins to be targets for MBL.

In conclusion, MBL concentrations are increased in type 1 diabetic patients with diabetic nephropathy. Whether MBL also plays a pathogenetic role is still not known, but its independent association with glycaemic control raises the possibility that it may have deleterious effects. No independent direct associations between MBL and inflammatory markers or insulin resistance were observed.