Type 1 diabetes (T1D) is a chronic autoimmune disease of multifactorial etiology that results in destruction of pancreatic β cells [1]. Since the 1980s, the incidence of T1D has increased globally [2]. Through the DiaMond Study (Diabetes Mondiale Project Group) published in 2000, Chile was classified as a country with a low incidence of T1D with 2.36 new cases per year per 100,000 inhabitants of Santiago, Chile's capital city, under 15 years of age (data obtained between 1990 and 1992) [3]. Between 1994 and 2000, however, the incidence of T1D in Chile increased to 4.5 per 100,000 inhabitants (<15 years) of Santiago [4]. Together, the increase in worldwide T1D incidence and the reduced number of genotypes that confer an increased T1D risk suggest that the environment is a predisposing factor to T1D. Potential mechanisms may include modifications in gene expression due to epigenetic changes [5].

Over the last 3 decades, there has been an increasing percentage of obesity in both developed and developing countries [6]. In Chile, it has been reported that 50% of the child population is overweight or obese [7]. The global increase in prevalence of overweight and obesity may result in epigenetic changes due to excessive accumulation of adipose tissue and subsequent infiltration of macrophages in production of proinflammatory cytokines [8]. Several cytokines that are released by adipose tissue, such as tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-6), have been proposed along with other mediators of inflammation to being related to generation of both T1D and type 2 diabetes [9].

There is increasing evidence linking miRNAs to autoimmune pathologies [10]. Other studies have shown that certain miRNA activity is related to adipose tissue development and function [11]. In this context, miR-378 has been reported to be upregulated in human adipocytes treated with IL-6, TNF-α, and leptin [12]. This miRNA is coded by an intron of the peroxisome proliferatoractivated receptor gamma coactivator 1β gene (PPARGC1β), counterbalancing the metabolic actions of peroxisome proliferator-activated receptor gamma coactivator 1β (PGC1β). Micro RNA-378 (miRNA-378) has also been identified as integral to the human regulatory system that controls systemic energy homeostasis and total oxidative capacity of insulin target tissues [13]. Furthermore, it is now apparent that miR-378 plays a role in several biological processes, such as cancer metastasis, osteoblast differentiation, and adipogenesis [14,15].

Given that Chile has been facing concurrent increases in both T1D and childhood obesity, Chile provides a suitable environment in which to further study the mechanisms related to the association between excess body fat and pathogenesis of T1D. Therefore, we sought to evaluate the expression profile of miRNA-378, levels of IL-6 and TNF-α, and their possible association with body fat in T1D Chilean patients.

Table 1 shows the clinical, immunological, and inflammatory characteristics of both the T1D and control groups. Nutritional status, body fat (%), IL-6 (pg/mL), and TNF-α (pg/mL) was similar between the groups (Table 1). Categorizing subjects by sex and pubertal stage resulted in a higher body fat percentage in pubertal girls even though there was no significant difference (Table 2). No significant differences were observed in BMI and z-score BMI between the groups. Waist circumference was statistically significantly higher in the pubertal groups compared to the prepubertal groups (Table 2).

The correlation analysis between inflammatory cytokines and anthropometric parameters is shown in Table 3. Significant correlations were observed for TNF-α and IL-6 with BMI and body fat in the total study sample but not in the stratified analysis by group (T1D vs. controls).

Fig. 1 shows the relative expression of miR-378 in T1D patients and controls. A higher expression of miR-378 was observed in those with T1D compared to the controls (P<0.05).

Fig. 2 shows the relative expression of miR-378 in those with T1D and the controls by pubertal stage. A lower expression of miR-378 was observed in prepubertal controls in comparison to pubertal controls, prepubertal T1D, and pubertal T1D (P<0.05).

Fig. 3 shows the correlations between miRNA-378 expression level and autoantibody concentration in those with T1D. Both AntiGAD65 and AntiZnT8 antibodies were statistically significantly positively correlated with miRNA-378. No statistically significant correlation was observed between the AntiIA2 antibody and miRNA-378.

Finally, Table 4 summarizes the correlations between cytokines and body composition related to miR-378 expression. However, no statistically significant correlations were observed among these parameters.

To the best of our knowledge, this is the first study of miR378 expression in T1D juvenile patients. Despite the previously described association between miR-378 and adipose tissue, we did not find associations between miRNA-378 with proinflammatory states or body composition even though we found that miRNA-378 was greater in those with T1D compared to the controls as well as among all subjects undergoing puberty. Furthermore, while the overall study sample had a high prevalence of overweight, this study also found no statistically significant differences in nutritional status (BMI z-score and percent body fat) of children with T1D and the controls.

In recent years, an increase in overweight T1D patients has been described. One study performed by Flechtner-Mors et al. [17] with 18,382 T1D children and teenagers from Germany and Austria showed that, between 2000 and 2006, there was an increase in obesity and overweight prevalence in both girls and boys with T1D. In addition, Dubose et al. [18] developed an investigation including 32,936 children from Germany, Austria, and the United States showing that, according to WHO classification criteria, 24% of patients are overweight and 12% are obese. In contrast, a study by Mosso et al. [19] of 30 T1D Chilean juvenile patients found that only 17% were overweight. The study of Gregory et al. [20] showed a higher body fat percentage in T1D pubertal girls compared to T1D prepubertal subjects and T1D pubertal boys.

Several publications have shown a progressive increase of body fat during puberty compared to the prepuberty stage [20,21]. Comparing those with T1D to those without T1D, weight gain during puberty tends to be higher in those with T1D in both sexes [22,23] and only in girls without T1D [17,19]. The greater weight gain in those with T1D may be explained as undesirable consequence of subcutaneous insulin treatment [23,24] that is known to have a double inhibitor effect of lipolysis and as a lipogenesis stimulator, promoting fat accumulation. This situation is increased during puberty due to higher insulin requirements caused by both higher anabolism and lower insulin sensibility [25].

In our study, no significantly higher BMI or body fat % was observed in T1D patients compared to controls when the analysis was stratified by age and pubertal stage. While we observed a higher waist circumference in pubertal T1D and controls, this was expected considering that weight, and possibly height, are likely to increase during puberty.

A limitation of our study is that we measured total body fat and not the distribution of body fat. Research conducted in El Cairo [26] also found that those with T1D had higher total body and abdominal fat compared to the controls. Taken together, our results are consistent with those by Paulino et al. [27] who also found no significant differences in BMI z-score or body fat % by group (with or without T1D) even though they found a higher waist circumference in T1D patients compared to controls. This fat distribution has been related to glucose control in teenagers, and the excess of adipose tissue has been linked to cardiovascular risk factors in T1D [28].

Our study showed similar inflammatory cytokine levels in diabetic patients compared to controls, which is in agreement with the literature since most T1D patients considered in this study had adequate metabolic control (glycosylated hemoglobin<7.5%) and were diagnosed less than 1 year prior. Recently, Alnek et al. [29], in concordance with our study, found elevated level of IL-1β but not IL-6 or TNF-α in T1D children.

In our study, miR-378 overexpression was observed in T1D patients compared to control patients. However, mi-378 expression was not statistically significantly correlated with percent body fat or inflammatory cytokine level. This may be explained by the similar inflammatory cytokines levels, nutritional status, and body fat of the T1D patients and the controls recruited for this study.

Carrer et al. [13] demonstrated that mice genetically lacking miR-378 are resistant to obesity induced by a high fat diet. Therefore, a reduction in fat mass deposits was observed as well as a diminution of adipocyte size, raising the possibility that this miRNA is necessary for efficient hypertrophy and lipid absorption in adipocytes. miR-378 counteracts the metabolic actions of PGC-1β [13]. PGC-1β acts in conjunction with transcription factors such as estrogen-related receptors (ERRα, ERRβy, and ERRγ), activating mitochondrial biogenesis, fatty acids oxidation, glucose uptake, and gluconeogenesis in several tissues [30].

Studies have shown that diabetic subjects with insulin resistance or obesity have decreased mitochondrial capacity in skeletal muscle that is linked to low PGC-α and PGC-1β expression and, therefore, to deregulation of both mitochondrial genes and their subsequent enzymatic activity [31,32]. miRNA-378 is known to be highly induced during adipogenesis. A study performed by Xu et al. [12], where human mature adipocytes were treated with TNF-α, IL-6, leptin, and resistin, showed miR-378 overexpression in adipocytes treated with TNF-α, IL-6, and leptin but not with resistin. Ishida et al. [33] found that miRNA-378 represses the adiponectin gene in human adipocytes that in turn likely reduces adiponectin expression. In addition, Gerin et al. [34] demonstrated that miRNA378/378* overexpression during adipogenesis increases triacylglycerol accumulation in adipocytes due to the increase of de novo lipogenesis. Therefore, we hypothesize that T1D, as described for other miRNAs [35], may contribute to miRNA-378 overexpression with consequent negative regulation of PGC-1β/ERRs and its target genes. Recent studies have reported that epigenetic factors such as miRNAs play a key role in pathogenesis of T1D, primarily affecting functioning of the immune system and apoptosis of pancreatic β cells [35].

When T1D patients and controls were categorized by pubertal stage, we observed lower miRNA-378 expression in all prepubertal subjects compared to pubertal subjects, suggesting a puberty effect on miRNA-378 expression. Given that miRNA-378 is necessary for adipose tissue hypertrophy [36], it can be explained by the lower insulin sensitivity and increase in fat mass that occurs during puberty [20,21,23]. The higher miR378 expression we found in T1D patients may be related to the greater gain in fat mass during puberty that has been previously observed in diabetic patients [17,19,22,23]. It can also be linked to Wilkin’s “Accelerator Hypothesis,” which indicates that adiposity responsible for insulin resistance could increase betacell demand and induce apoptosis with autoantigen release, accelerating autoimmune damage in genetically predisposed subjects [37].

miR-378 is related to PGC-1β, a coactivator of ERRs. ERR expression and its role in pubertal development have yet to be fully defined. A recent study [38] demonstrated that ERRγ is expressed in mouse Leydig cells, increases synthesis of testicular testosterone, and regulates steroidogenic genes expression, which could link it to pubertal development.

Our study showed positive correlations between autoantibodies and miR-378 relative expression that was only statistically significant for AntiGAD65 and AntiZnT8. Several studies have compared the expression profiles of miRNAs with those of autoantibodies in human subjects with T1D. Salas-Pérez et al. [39] observed decreases in miR-21a and miR-93 expression in T1D children compared to healthy controls but found no correlation between expression of these miRNA and autoantibody profiles. In contrast, the study by Sebastiani et al. [40] is consistent with our findings in which greater expression of miR-326 in T1D subjects was positively correlated with AntiGAD65 and AntiIA2 autoantibodies.

Taken together, the finding indicate that microRNA may have a modulating role in the autoimmune response of T1D patients. Further studies are necessary to determine if the differential miR-378 expression in T1D patients compared to that of healthy controls is due to an event after or before T1D onset–the latter of which could position miR-378 as a potential biomarker for T1D.