The role of zonulin in the pathogenesis of diabetic retinopathy

Introduction

Diabetic retinopathy (DR) is one of the severe microvascular complications of diabetes mellitus (DM) and is the most prevalent cause of preventable visual loss in adulthood. Mainly chronic exposure to high blood glucose levels and other factors initiates activation of some biochemical pathways that finally give rise to retinal microvascular damage and dysfunction. These pathways include sorbitol accumulation, advanced glycation end-product formation, increased generation of reactive oxygen species, renin-angiotensin system upregulation, increased expression of vascular endothelial growth factor and inflammation, etc. [1].

Low-grade chronic inflammation is shown to have a prominent role in DR pathogenesis. There is an increase in the adherence of leukocytes from the initial stages of DR. Chemokines that regulate the activation and attraction of leukocytes such as monocyte chemotactic protein 1 (MCP-1), macrophage inflammatory protein 1-α (MIP-1α) and MIP-1β were in DM patients. Also, inflammatory cytokines, such as interleukin-6 (IL-6), IL-8, IL-1β, and tumor necrotizing factor-alpha (TNF-α) are increased in patients with DM and DR [2], [3], [4], [5]. In addition, the roles of adipokines such as omentin-1 [6] and neuregulin-4 [7] in the development of DM and microvascular complications [8] were reported in several studies.

Recently increased attention has been paid to the relationship between chronic inflammation and gut microbiota. Altered gut microbiota composition or dysbiosis indicates an imbalance between commensal symbionts and pathobionts, significant elevation of pathobionts and decrease of beneficial butyrate-producing bacteria [9]. Many studies examined the link between gut microbiota and DM and revealed evidence of gut microbiota dysbiosis [10]. On the other hand, several human and animal studies implicated gut microbiota dysbiosis in pathogenesis of diabetic retinopathy [9] and also showed improvements of gut barrier function resulted in decreased retinopathy by the way of the gut-eye axis [11].

For monitoring intestinal permeability, a biomarker called zonulin, a 47 kDa protein that was revealed as prehaptoglobulin-2, was shown to regulate tight junctions of enterocytes and in this way modulates intestinal permeability physiologically [12]. It was shown that the promoter region of the haptoglobulin gene that encodes zonulin is modulated by IL-6 via STAT3 activation and miR-18a induction [13]. Upregulation of zonulin leads to disassembly of tight junctions, uncontrolled antigen trafficking from the intestinal lumen to submucosa and consequently altered intestinal permeability, which gives rise to chronic inflammatory diseases [14], [15]. Nowadays, various studies indicated that circulatory levels of zonulin are elevated in several metabolic diseases such as obesity, insulin resistance, metabolic syndrome, polycystic ovary syndrome, type 1 DM and T2DM [16], [17], [18], [19], [20], [21].

Although IL-6 is shown to be one of the main regulators behind the pathological process in inflammatory ophthalmic diseases such as diabetic retinopathy [22], there is no report about IL-6 levels during the progression of DR simultaneously with zonulin. To date, knowledge about the interrelationship between intestinal permeability and clinical variables for patients with a relatively long duration of T2DM is sparse. In this respect, we aimed to evaluate serum zonulin levels along with systemic levels of IL-6 in T2DM patients with retinopathy complications and to investigate the correlation with metabolic and biochemical parameters.

Materials and methods

Results

Eighty-nine T2DM patients (30 patients with NDR, 29 patients with NPDR, 30 patients with PDR) and 30 HC were included in this study. Characteristics of the study population are summarized in Table 1. There were no significant differences between groups for sex distribution, BMI, ALT, uric acid, TG, HDL-C and TC levels (p>0.05). Significant differences between groups were present for age (p=0.001), duration of diabetes (p<0.001), fasting blood glucose (p<0.001), creatinine (p<0.001), LDL-C (p=0.009), HbA1c (p<0.001), zonulin (p=0.049) and IL-6 levels (p<0.001).

Zonulin and IL-6 levels were significantly different between groups with assessment of the Kruskall Wallis test followed by Mann Whitney U test. Pairwise comparisons of groups for zonulin levels showed no statistical difference, but IL-6 levels were significantly increased in NPDR and PDR groups compared to the HC group. Additionally, IL-6 levels were significantly increased in NPDR and PDR groups compared to the NDR group (Table 1, Figures 1 and 2).

Figure 1:

Serum zonulin levels in study groups (HC, healthy control; NDR, no diabetic retinopathy; NPDR, non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy).

Figure 2:

Serum IL-6 levels in study groups (HC, healthy control; NDR, no diabetic retinopathy; NPDR, non-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy).

Significant positive correlations were found between zonulin with IL-6 (Figure 3) and BMI (r=0.201, p=0.028; r=0.336, p<0.001 respectively). Serum IL-6 level was also positively correlated with HbA1c, fasting blood glucose and duration of diabetes (r=0.299, p=0.001; r=0.364, p=0.000; r=0.422, p=0.000, respectively). HbA1c was positively correlated with fasting blood glucose and duration of diabetes (r=0.756, p=0.000; r=0.604, p=0.000 respectively). Additionally, fasting glucose was positively correlated with duration of diabetes (r=0.625, p=0.000).

Figure 3:

Spearman correlation analysis between serum zonulin and IL-6 levels (r=0.201, p=0.028).

Independent effects of BMI, HbA1c, duration of diabetes, zonulin and IL-6 on DR development were investigated by multivariate logistic regression and it was found that duration of diabetes and serum levels of IL-6 had an independent effect on DR development, but zonulin did not have an effect (Table 2).

Discussion

In the present study, we evaluated the levels of zonulin as a gut permeability marker and IL-6 as an inflammatory marker in non-complicated and complicated DR associated with T2DM. The initial assessments of zonulin levels were significantly different between groups, while the pairwise comparisons of groups showed no significance; however, the lowest level was found in the NPDR group and the highest level was found in the PDR group. Serum IL-6 levels were significantly increased in NPDR and PDR groups when compared to both HC and DM groups. These data show that the IL-6 levels were simultaneously increased with the progression of DM which may also represent the role of chronic inflammation in the diabetic retinopathy pathogenesis. To the best of our knowledge, this is the first study that assesses the serum zonulin and IL-6 levels in cases with T2DM and their possible association with the presence and severity of retinopathy.

In the literature, there are very few human studies that evaluate serum zonulin levels in T2DM patients [19], [20]. However, these studies did not include complicated diabetes with long duration, including newly diagnosed T2DM patients. Zhang et al. studied serum zonulin levels in subjects who were healthy, had impaired glucose tolerance and with newly diagnosed DM. The authors showed a progressive increase in serum zonulin level with the impairment of glucose tolerance and significantly highest levels in DM patients [19]. Jayashree et al. studied serum levels of lipopolysaccharide, IL-6, TNF-α and zonulin in healthy and newly diagnosed subjects. The authors found that the levels of all parameters were significantly increased in the T2DM group [20]. While significantly increased IL-6 levels are consistent with Jayashere et al. [20], zonulin levels are contradictory. There were no significant differences in the T2DM group compared to healthy controls; furthermore, the zonulin levels were decreased in the NPDR group but finally increased in the PDR group. The main reason for this contradiction is the totally different construction of our study groups as healthy controls, DM, DM with non-proliferative retinopathy and DM with proliferative retinopathy. Also, in our study, diabetic patients were diagnosed in the past and monitored by a clinician for a period of time (8–16 years duration).

Human gut microbiota varies between individuals [24]. In addition, antidiabetic and antibiotic therapies and diet alterations were shown to reshape the gut microbiota. Metformin (dimethyl diguanide), a very common treatment for T2DM, is shown to affect intestinal microbiota positively and causes a decrease in zonulin levels when used in T2DM patients [25], [26], [27]. Nowadays, metformin usage is suggested with the purpose of preventing and treating aging-dependent leaky gut and inflammation by regulating intestinal microbiome, goblet cell count and also mucin biology [27]. In the present study, there was no significant difference in metformin use between diabetic groups. High-fat diets are also related with decreased intestinal bacterial diversity and leaky gut [28]. The difficulty in interpreting these results is due to failure to evaluate gut microbiota in the present study and the fact that the participants did not consume a standard diet.

In the current study, the duration of DM, the level of fasting glucose, HbA1c and creatinine levels were steadily increased, and significantly higher levels were found in the PDR group compared to the HC and NDR groups (Table 1). This data is consistent with DR increasing proportionally to the disease duration in T2DM. Also, long-term exposure to high glucose levels increases the risk of vascular damage to the retina [1], [2].

Some limitations are present in this study. The cross-sectional nature of the study is the main one; serum zonulin and IL-6 levels of all participants were measured once. Other limiting factors are that a standard diet could not be applied to the participants included in the study and that gut microbiota was not evaluated.

Conclusion

IL-6 as a marker of chronic inflammation seems to be involved during the clinical progression of DR. The role of changes in serum zonulin level in the pathogenesis of DR could not be determined. However, further studies using a standardized diet and also evaluating gut microbiota may contribute to identifying the role of zonulin in DR pathogenesis.