Cytokines and type 1 diabetes complications: casual or causal association?

Type 1 diabetes (T1D) is associated with long-term microvascular and macrovascular complications, which are mediated by chronic hyperglycemia. The Diabetes Care and Complications Trial (DCCT) and the Epidemiology of Diabetes Interventions and Complications (EDIC) study established that tight glycemic control results in marked risk reduction for the development or progression of long-term complications 1, 2. The application of these results to practice is tempered by the realization that they were achieved in a carefully screened and motivated population and required intense professional investment. Despite the many advancements in diabetes care over the past 25 yr [intensive therapy, insulin pumps, human recombinant and insulin analogues, rapid self blood glucose monitoring and hemoglobin A1c (HbA1c) measurement], most children and adults with T1D fail to achieve target HbA1c 3. Because of the persistent difficulty achieving physiologic blood glucose control that would greatly reduce long-term complications of T1D, elucidating the specific mechanisms involved in hyperglycemia-induced tissue damage could provide another therapeutic target. The proinflammatory cytokines interleukin (IL)-1, tumor necrosis factor, interferon γ, IL-6, IL-10, IL-15, IL-17, and monocyte chemoattractant protein-1 have been implicated in the development of microvascular and macrovascular complications in both T1D and type 2 diabetes 4, 5. While several studies have demonstrated strong correlations between proinflammatory cytokines and endothelial damage, it remains to be demonstrated whether a cause-and-effect relationship exists between proinflammatory cytokines and the pathogenesis of diabetes complications. Whereas some investigators hypothesize that cytokine alterations play a nefarious role in disease pathogenesis, others contend that cytokine changes are a protective physiologic response to hyperglycemia-induced stress 6-8. In addition to the potentially important role cytokines play in the genesis or promulgation of microvascular and macrovascular complications, cytokines are widely implicated in the complex cascade of signals that control both innate (natural) and adaptive (acquired) immunity 9. Cytokines have been linked to the development of autoimmunity, including that to the beta cell 10. Cytokines play a key role in signaling dendritic cells, toll-like receptors, T cells and B cells, among others, and are controlled by both positive and negative feedback loops. Adding additional complexity to the picture, the magnitude and duration of the cytokine response are determined by the integration of signals mediated by both regulatory and effector T cells. Thus, it is important to consider that an imbalance between proinflammatory and anti-inflammatory cytokine activities may favor both the induction of autoimmunity and the chronic inflammation that leads to complications. In this issue of Pediatric Diabetes, Rosa et al. studied the kinetics of 14 proinflammatory and anti-inflammatory cytokines in pediatric T1D subjects who had hyperglycemia followed by euglycemia. They hypothesized that children with T1D and morning hyperglycemia would develop elevated proinflammatory cytokines and that such elevations would not immediately correct with normalization of blood glucose. The 22 subjects developed either mild postprandial hyperglycemia (n = 9, mean peak glucose 7.8 ± 0.7 mmol/L) or marked postprandial hyperglycemia (n = 13, mean peak glucose 15.6 ± 1 mmol/L) following a light breakfast. They were then given insulin infusions to achieve a stable glucose concentration (5–6.1 mmol/L) over the subsequent 30–150 min. Euglycemia was achieved and maintained in each child for at least 90 min such that glucose values for both the mild hyperglycemic and marked hyperglycemic groups were nearly identical over the final 120 min of the study. Glucose concentration was monitored every 10–15 min, and cytokine measurements were made on samples obtained at 30-min intervals. Although the authors performed broad exploratory analyses, they focused their study hypothesis on changes in IL-6 because the authors had previously reported a correlation between morning hyperglycemia and IL-6 levels in children with T1D 11. Initially thought to function solely as a proinflammatory cytokine, IL-6 plays a far more complex role and has also been implicated in the autoimmune genesis of T1D. Active in both the amplification of and protection from inflammation, IL-6 is produced by a wide variety of cells including immune cells, endothelial cells, skeletal and smooth muscle cells, adipocytes, and even beta cells and appears to play an essential role in host defense 8. Using unpaired t-tests to compare the average values of subjects in the two groups, IL-6 was higher in the hyperglycemic group when euglycemia was first achieved and remained significantly higher, even after both groups had maintained euglycemia for 120 min. In addition, they found that IL-1α, a proinflammatory cytokine, and IL-4, an anti-inflammatory cytokine, were similarly elevated. This interesting kinetic study is unique in children with T1D in that most publications of serum cytokine abnormalities of T1D patients have largely focused on single measures and cross-sectional comparison to control groups 5, 12-14. There are several limitations of the study, however. No significant differences were seen in the other 11 proinflammatory and anti-inflammatory cytokines that were measured. There have been no mechanistic studies indicating a cause-and-effect relationship between cytokine changes and vascular dysfunction. The relatively high variability of cytokine measurements in general and the small number of subjects in this study limit the ability to interpret the data. Furthermore, the lack of information about the subjects’ HbA1c levels, duration of diabetes, history of recent illnesses, autoimmunity status, pubertal stage, body mass index, insulin dosage, exercise (which the authors have previously studied), and glucose levels beyond the 2-h study period emphasizes that there are numerous additional variables that may affect cytokine measurement that need to be considered before making firm conclusions about the observed cytokine alterations. Age-matched type 2 diabetic patients would have provided additional insight and helped to separate the autoimmune from inflammatory effects. Finally, the lack of correction for multiple comparisons requires additional caution when interpreting potential differences between groups. Improved understanding of the mechanism for the relationship between hyperglycemia and neuropathy, nephropathy, retinopathy, and vasculopathy is vital for developing therapies to alleviate the devastating long-term complications of T1D. While the findings of Rosa et al. are provocative, the data must be interpreted with caution. Association may be casual rather the causal. Healthy skepticism awaits larger confirmatory studies as well as demonstration of the mechanism for cytokine-induced complications. Until such studies provide another means of reducing the burden of long-term complications of diabetes, we must continue to focus efforts on achieving optimal glycemic control in all our patients. We thank Dr Arlan Rosenbloom for his critical review.