Inflammation

, Volume 35, Issue 2, pp 723–729

Timing of Insulin Therapy Affects the Inflammatory Response in Endotoxemic Rats

Authors

  • Bo Zou
    • Medical School of Nanjing University
  • Qiyi Chen
    • Medical School of Nanjing University
  • Shaoqiu Tang
    • Medical School of Nanjing University
  • Tao Gao
    • Research Institute of General SurgeryNanjing General Hospital of Nanjing Military Command
  • Juanjuan Zhang
    • Research Institute of General SurgeryNanjing General Hospital of Nanjing Military Command
  • Fengchan Xi
    • Research Institute of General SurgeryNanjing General Hospital of Nanjing Military Command
    • Research Institute of General SurgeryNanjing General Hospital of Nanjing Military Command
Article

DOI: 10.1007/s10753-011-9367-8

Cite this article as:
Zou, B., Chen, Q., Tang, S. et al. Inflammation (2012) 35: 723. doi:10.1007/s10753-011-9367-8

Abstract

The aim of the present study was to determine whether timing of insulin administration influences the hepatic and serum proinflammatory and anti-inflammatory cytokines during endotoxemia stimulated by lipopolysaccharide (LPS). Eighty-one male Sprague–Dawley rats were divided into different time groups and insulin was given 30 min pre-LPS administration or hour 0, 1, 3, 6, 12, 24 after the induction of endotoxemia, respectively. Hepatic and serum proinflammatory cytokines IL-1β, IL-6, and TNF-α, and anti-inflammatory cytokine IL-10 were detected 24 and 48 h after the induction of endotoxemia. Compared with sham control rats, serum concentrations of proinflammatory cytokines IL-1β, IL-6, and TNF-α and anti-inflammatory cytokine IL-10 significantly increased on 24 and 48 h after induction of endotoxemia. Similarly, LPS administration also significantly increased the hepatic IL-1β, TNF-α, IL-6, and IL-10 protein concentration 48 h after LPS injection. Compared with levels in positive LPS controls animals receiving saline, on 24 and 48 h after LPS injection, insulin administrated ahead of 6 h after LPS injection significantly decreased the serum IL-1β, IL-6, and TNF-a concentration (P < 0.05), and significantly increased anti-inflammatory cytokine IL-10 concentration (P < 0.05); hepatic IL-1β and IL-6 expression were (P < 0.05) significantly decreased compared with levels in positive LPS controls. But, the significant decrease of hepatic TNF-a expression and significant increase of hepatic IL-10 were only seen in the animals in which insulin was administrated at 30 min pre-LPS or coadministrated with LPS. Insulin administrated 6 h after LPS injection lost the ability to significantly reduce serum or hepatic IL-1β, TNF-α, and IL-6 concentrations. Insulin has a protective role in systemic inflammatory response syndrome related to sepsis, such as downregulation of proinflammatory cytokines and upregulation of anti-inflammatory cytokine production. However, timing of insulin administrated may change its effect of inflammatory response in endotoxemic rats. Insulin administrated 6 h after LPS injection weaken the ability to protect inflammatory response related to sepsis.

KEY WORDS

insulinendotoxinproinflammatory cytokineanti-inflammatory cytokinetiming of therapy

INTRODUCTION

Sepsis and septic shock, associated with substantial morality and health care resources, is a significant public health problem. The systemic inflammatory response syndrome (SIRS) triggered by lipopolysaccharide (LPS) from Gram-negative bacteria affects many organs and may lead to death. Since its discovery in 1921, insulin has been considered a key metabolic hormone. In recent years, our understanding of insulin’s effect on pathophysiological reaction has greatly improved. The role of insulin is one of the controversial topics in inflammatory response of patients with sepsis. Intensive insulin therapy (IIT) was shown to decrease mortality in critically ill patients with or without sepsis [1, 2]. Insulin given at doses to maintain blood glucose less than 110 mg/dl prevented the incidence of multi-organ failure and thus improved clinical outcome and rehabilitation of patients mainly undergoing thoracic surgery. Insulin administration furthermore decreased the incidence of sepsis and septic events in these patients. One theme that has received a modest degree of attention in the extensive literature has surrounded this topic is the debate about whether it is glycemic control per se, or the role of insulin (including both glycemic and its many nonglycemic effects) that modulates the potential therapeutic benefit of this intervention.

Recently, subsequently large randomized controlled trials failed to replicate the mortality benefit of IIT in medical and mixed medical–surgical ICU [3]. Moreover, a NICE-SUGAR study [4] showed that IIT increased mortality among adults in the ICU: a blood glucose target of less than or equal to 180 mg/dl resulted in lower mortality than did a target of 81 to 108 mg/dl. The increased mortality of IIT therapy was presumed to partly contribute to more insulin infusion. However, results from animal studies indicate that insulin has a protective role in systemic inflammatory response related to sepsis, such as inhibition of transcription factor NF-κB with subsequent downregulation of proinflammatory events as well as a reduced production of reactive oxygen species, and influence on T-cell differentiation towards a TH2-specific response [5].

The molecular cascade of SIRS characterized by release of proinflammatory cytokines, such as IL-1β, IL-6, TNF-α, has been well recognized in previous animal experiments. And insulin showed potent anti-inflammatory effect on modulating inflammatory processes in these experiments. However, in these experimental models, the insulin is used to administer before or around the time of the septic challenge. Unfortunately, the early promise offered by this agent has not been realized in the clinical situation. Still, data aiming directly at timing effect of insulin on SIRS is lacking. We hypothesized that early or later administration may change the effect of insulin therapy, which may be one of the explanations for various different conclusions in some clinical and experimental animal studies. The aim of the present study was to determine whether timing of insulin administration influences the systemic inflammatory response during endotoxemia.

MATERIALS AND METHODS

Animal Care

The study was approved by the animal welfare committee of the province of JiangSu (China). Male Sprague–Dawley rats (n = 72, 350–400 g) were placed in wire-bottom cages and housed in a temperature-controlled room with a 12-h light, 12-h dark cycle. Rats were acclimatized to their environment for 7 days before the study. All animals were anesthetized with thiopentone sodium (Intraval 120 mg/kg, intraperitoneally), and anesthesia was maintained by supplementary injections of thiopentone sodium (approximately 1–2 mg/kg/h intravenously) as required. All rats were given glucose (4.5 mg/kg/h intravenously) throughout the experiment.

Study Protocol

To test the change of inflammatory response in endotoxemia, rats were divided into two groups LPS group (Escherichia coli serotype O111:B4 3 mg/kg, n = 10) and sham control group (rats were treated with saline, n = 8). In both groups, rats did not receive insulin administration. LPS (E. coli serotype O111:B4; Sigma-Aldrich, St Louis, MO) at a dose of 3 mg/kg body weight was chosen, because this represents half of the 50% lethal dose for LPS, but induces a severe endotoxemic reaction as described by Jeschke [6].

To test whether inflammatory response after administration of LPS was influenced by timing insulin therapy, rats were divided into six groups (n =8~10/group), and insulin was given 30 min pre-LPS administrated or hour 0 (n = 8), 1 (n = 9), 3 (n = 8), 6 (n = 10), 12 (n = 9) and 24 (n = 10) after the induction of endotoxemia. The insulin used was protamine-insulin (Berlininsulin H, Berlin-Chemie AG, Berlin, Germany), a form of insulin that is released over a 24-h period. The dose of 5 IU/kg body weight of insulin was chosen based on previous studies of the effects of insulin in vivo [6, 13].

Sacrifice and Collection of Samples

Blood was collected to stabilize for 20 min, at t = 24 and 48 h after administration of LPS by puncture of the vena cava inferior, and centrifuged at 1,000 × g for 15 min. The serum was stored at −80°C until analysis. Animals were killed 48 h after LPS injection by an overdose of anesthesia. Samples of liver were harvested, fragmented, snap-frozen in liquid nitrogen, and stored at −73°C for analysis.

Hepatic and Serum Cytokine Protein Concentrations

Hepatic and serum proinflammatory cytokines IL-1β, IL-6, and TNF-α, and anti-inflammatory cytokine IL-10 were determined by ELISA (R&D Systems, Inc., Minneapolis, MN). Serum was used either pure or diluted, and measurements were performed according to kit guidelines. Liver was completely homogenized in a lysis buffer (HEPES, sucrose, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate, dithiothreitol, phenylmethylsulfonylfluoride, leupeptin, EDTA, and pepstatin) in a ratio of 1:6. The exact concentration was 100 mm HEPES, 10% sucrose, and 0.1% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate. On 10 ml of this solution, one tablet of Complete Mini was added and stored on ice. After homogenization, samples were centrifuged at 4°C at 14,000 rpm for 10 min. The clean supernatant was then used to determine cytokine protein concentration.

Statistics

Statistical analysis involved a standard software application (SPSS, Chicago, IL, USA).Groups were compared by one-way analysis of variance and a post-hoc Tukey test. Probability values <0.05 were considered statistically significant. Data were expressed as mean values ± standard error of the mean.

RESULTS

Serum Glucose and Electrolytes

Administration of LPS caused a significant decrease in blood glucose levels on 24 h when compared with sham control (Table 1). Insulin administrated in 24 h did not change the blood glucose concentration compared with that in positive LPS controls. Furthermore, there were no significant differences in serum sodium, potassium, calcium, and phosphate levels between endotoxemia and insulin given 30 min pre-LPS or administrated 0, 1, 3, 6, 12 and 24 h after the induction of endotoxemia (Table 1).
Table 1

Serum Glucose, Sodium, Potassium, Calcium, and Phosphate in Timing of Insulin-treated and Endotoxemic Rats after LPS Administration

 

Sham control

24 h after induction of endotoxemia

48 h after induction of endotoxemia

Timing of insulin-treated after LPS

Timing of insulin-treated after LPS

LPS

−30min

0h

1h

3h

6h

12h

LPS

0h

−30min

1h

3h

6h

12h

24h

Glucose (mg/dl)

181 ± 42

157 ± 18 *

140 ± 34*

151 ± 29*

156 ± 35*

143 ± 31*

157 ± 13*

149 ± 37*

169 ± 35

165 ± 27

158 ± 37

161 ± 41

157 ± 37*

167 ± 21

164 ± 36

152 ± 47*

Sodium (mg/dl)

159 ± 11

163 ± 32

157 ± 21

175 ± 12

149 ± 41

163 ± 24

165 ± 31

159 ± 17

172 ± 36

152 ± 31

167 ± 35

162 ± 13

157 ± 21

167 ± 31

157 ± 38

165 ± 27

Potassium (mg/dl)

6.7 ± 0.9

7.3 ± 0.5

6.1 ± 0.7

6.3 ± 0.2

7.8 ± 0.9

5.6 ± 0.4

6.9 ± 0.6

7.3 ± 0.4

8.1 ± 0.3

5.8 ± 0.9

5.2 ± 0.7

7.5 ± 0.6

7.3 ± 0.3

6.5 ± 0.6

5.9 ± 0.5

6.3 ± 0.7

Calcium (mg/dl)

3.7 ± 0.6

2.8 ± 0.2

3.7 ± 0.9

2.9 ± 0.3

3.0 ± 0.2

3.2 ± 0.5

3.4 ± 0.3

3.8 ± 0.4

3.2 ± 0.3

2.8 ± 0.3

3.4 ± 0.3

2.6 ± 0.6

3.7 ± 0.3

4.1 ± 0.2

3.5 ± 0.4

3.8 ± 0.3

Phosphate (mg/dl)

3.4 ± 0.5

3.1 ± 0.2

2.8 ± 0.3

2.4 ± 0.5

3.6 ± 0.3

3.1 ± 0.2

4.1 ± 0.9

4.2 ± 0.7

3.9 ± 0.5

4.7 ± 0.9

3.6 ± 0.7

2.8 ± 0.8

3.8 ± 0.5

3.6 ± 0.9

4.1 ± 0.4

3.9 ± 0.3

Data are the means ± SD

*P < 0.05; significant difference compared with sham control group

Serum Cytokine Levels

Compared with sham control rats, serum concentrations of proinflammatory cytokines IL-1β, IL-6 and TNF-a were significantly increased on 24 and 48 h after induction of endotoxemia. Insulin administrated at 30 min pre-LPS or coadministrated with LPS significantly decreased the serum IL-1β concentration on 24 and 48 h after LPS injection compared with levels in positive LPS controls (P < 0.05; Fig. 1a). Insulin administrated at 1 and 3 h post-LPS significantly decreased the IL-1β concentration 48 h after LPS injection (P < 0.05; Fig. 1a). No statistically significant reduction of IL-1β concentration could be found concerning insulin administrated 6 h after LPS injection. Insulin administrated 12 h after LPS injection attenuated the increase in serum TNF-a, and significantly decreased serum TNF-a at t = 48h compared with levels in positive LPS controls (P < 0.05; Fig. 1b). However, insulin administrated 12 and 24 h after LPS injection lost the ability to reduce serum TNF-a (Fig. 1b). In addition, the action of insulin to significantly decrease serum IL-6 concentrations at t = 24 and 48 h after LPS injection was found only when insulin administrated within 6 h post-LPS (P < 0.05; Fig. 1c).
https://static-content.springer.com/image/art%3A10.1007%2Fs10753-011-9367-8/MediaObjects/10753_2011_9367_Fig1_HTML.gif
Fig. 1

a, b, c, d Change of serum cytokine in each group. *Compared with LPS rats, significant difference was seen on 24 h after induction of endotoxemia (P < 0.05). #Compared with LPS rats, significant difference was seen on 48 h after induction of endotoxemia (P < 0.05).

Similar to proinflammatory cytokines, anti-inflammatory cytokines IL-10 was elevated after endotoxemia. Animals received insulin injection at 30 min pre-LPS or coadministrated with LPS showed significant increases in IL-10 immediately on 48 h after LPS injection (P < 0.05; Fig. 1d). No statistically significant augmentation of IL-10 concentration could be found at t = 24h.

Hepatic Cytokine Protein Concentration

LPS administration significantly increased the hepatic IL-1β, TNF-a, IL-6, and IL-10 protein concentration on 48 h after LPS injection. Hepatic IL-1β and TNF-a expression were significantly decreased at t = 48 h by insulin administrated at 30 min pre-LPS. Coadministrated with LPS and 1, 3, 6 h after LPS injection compared with that in positive LPS controls (P < 0.05; Fig. 2a and b), nevertheless was not significantly decreased by insulin administration 6 h after LPS injection. Compared with significantly decreased by insulin administration at 30 min pre-LPS and 0, 1, 3 h after LPS injection, hepatic IL-6 expression showed no significant difference between in rats administrated insulin after 3 h post-LPS injection and in positive LPS controls (Fig. 2c). Hepatic anti-inflammatory cytokine IL-10 was not significantly different in all groups with insulin administrated pre- or post-LPS injection or without insulin administration.
https://static-content.springer.com/image/art%3A10.1007%2Fs10753-011-9367-8/MediaObjects/10753_2011_9367_Fig2_HTML.gif
Fig. 2

a, b, c, d Change of hepatic cytokine in each group. *Compared with LPS rats, significant difference was seen on 48 h after induction of endotoxemia (P < 0.05).

DISCUSSION

In the last decade, intensive insulin therapy or tight glycemic control has become one of the most controversial issues. Not like in clinical situations, animal studies of insulin therapy showed conformable benefits, especially in inhibiting the inflammatory reaction. We guess time of insulin use can be the reason of the differences. This study was designed to compare the inflammatory reactions of different timing of insulin therapy in sepsis. Osuchowski et al. observed plasma concentrations of proinflammatory and anti-inflammatory cytokines in sepsis mice and concluded that either pro- or anti-inflammatory cytokines were reliable in predicting mortality up to 48 h before outcome [9]. The liver, as one of the main metabolic and inflammatory organs, plays a major role after stress. So, in our study IL-1β, TNF-α, IL-6, and IL-10 concentrations in serum and liver within 48 h were chosen as indicators of inflammatory response.

Maintaining normal blood glucose level can explain many of the effects of intensive insulin therapy. However, recent evidence indicates that insulin has anti-inflammatory effects that are independent of controlling hyperglycemia. In our study, the serum glucose level in all groups has no statistical obvious difference. But we observed similar decrease of proinflammatory mediators, comparing to glucose control animal studies. In fact, the challenge that in critically ill patient’s insulin treatment or glucose control, which is more important, has been raised early [10]. Van den Berghe concluded that normoglycemia, rather than the insulin dose per se, was related to the beneficial effects of intensive insulin therapy. In clinical studies, it is impossible to separate the impact of insulin from that of glucose control. A well designed clinical trial will be helpful to promote our understanding of insulin’s role in the critically ill.

Release of IL-1β is increased in animal models of sepsis and in clinical sepsis and septic shock [11]. Elevated IL-1β levels in plasma have been proved to correlate with increased mortality long ago [12]. Here, we find that insulin administrated 30 min pre-LPS or at 0, 1, 3 h after LPS, can significantly decrease the serum IL-1β concentration on 24 and 48 h. Hepatic IL-1β level at 48 h also decreased when insulin was used at −30 min, 0 h, 1 h, or 3 h after LPS injection. But when insulin was used later than 6 h, no such effect was observed. Our findings are consistent with Jeschke et al.’s research in endotoxemic rats [13]. They found that insulin injection together with LPS significantly decreased the hepatic IL-1β mRNA expression on the first and second days after the induction of endotoxemia.

IL-6 level is a useful tool to predict mortality and identify patients who are at risk for development of MODS [14, 15]. Barkhausen et al. [16] reported injection of insulin in 1 hour after trauma/hemorrhage and daily in a murine two-hit model diminished serum IL-6 and its mRNA expression in liver tissue. In the present study, we confirmed similar findings that timely administration of low-dose insulin can decrease the levels of IL-6 in both serum and liver. Delayed administration of insulin did not have such effect. In a similar model, hepatic IL-6 mRNA increased by almost 100-fold compared to normal and insulin significantly decreased hepatic IL-6 mRNA on days 2 and 7 when injected at a dose of 5 IU/kg body weight together with LPS [13].

TNF-α was early known to be one of the most important inflammatory cytokines in septic patients [17]. Activation of the TNF-α receptor results in stimulation of NF-κB signaling via Ikkb, and can then upregulate IL-6 and IL-1β. Insulin therapy up to 6 h significantly inhibits TNF-α in this study. IL-10 is an anti-inflammatory cytokine produced by macrophages and lymphocytes. Leonidou et al. concluded that serum IL-10 levels are prognostic factors of hospital mortality in severe sepsis patients [18]. Our study showed little effect of insulin on IL-10 level.

In our study, delayed injection of insulin (later than 6 h) cannot reduce the levels of IL-1β, IL-6, and TNF-α in both serum and liver at t = 48 h. Why does different timing of insulin therapy have diverse effects in modulating inflammatory cytokines? Insulin resistance caused by trauma, burn, infection, or stress could possibly explain this question. Studies suggest that several pathways are involved in insulin resistance during sepsis. Inactivation of Akt is thought to be the main mechanism leading to insulin resistance, while the activation of phosphatidylinositol 3-kinase (PI3K)/protein kinase B (Akt) pathway by insulin has been regarded as one of the key factors in insulin therapy. Kidd et al. proved that insulin reduces LPS-induced inflammation in mice in a PI3K/Akt-dependent manner without affecting blood glucose levels. Insulin resistance that occurred hours after LPS injection can directly or indirectly influence the anti-inflammatory effect of insulin therapy. That can explain the early administration of insulin therapy may bring better anti-inflammatory effect. In most of the animal studies related to insulin therapy in sepsis, insulin was given immediately after animal model was established, which means before the onset of insulin resistance. So the anti-inflammatory effect could easily be achieved.

In the LPS group, high levels of IL-1β, IL-6, and TNF-α can surely contribute to the pathogenesis of insulin resistance [19]. In clinical situations, insulin is usually used when hyperglycemia occurred, which sometimes as a result of insulin resistance. Honiden et al. conducted a clinical trial comparing timing of intensive insulin therapy in critically ill patients [7]. Their data suggests that insulin therapy within 48 h of ICU admission is associated with better outcomes. These differences can be explained by complex situations in clinics or adverse effects of insulin resistance. Although the situation in clinical trial is quite different from that in animal study, our research may still support the early use of insulin in sepsis.

A main defect of our study is that we did not evaluate the insulin sensitivity of the experiment animals. A subanalysis of Van den Berghe’s study suggests that IIT may improve insulin sensitivity [8]. A recent clinical trial in severely burned children proved that intensive insulin therapy can improves insulin sensitivity [20]. We guess insulin therapy, though not changing blood glucose, may also involved in improvement of insulin sensitivity. Further study should be focused on dosage of insulin: the insulin dosage that can play the most power in modulating inflammatory cytokines or even reverse insulin resistance while not causing hypoglycemia.

In conclusion, this is the first animal study concerning the time point of insulin therapy in sepsis. Thirty minutes pre-LPS injection or within 6 h after induction of endotoxemia, the administration of low-dose insulin, without changing blood glucose, showed exact effects on downregulation of proinflammatory cytokines, in both serum and liver. Delayed administration of insulin had no such good effect, which may be related to insulin resistance. Further studies, including evaluating the severity of insulin resistance, are needed to explain the differences and determine the optimal dose of insulin in hyperglycemia during sepsis.

Acknowledgment

This study was supported by the National Natural Science Foundation in China (No. 30801086).

Copyright information

© Springer Science+Business Media, LLC 2011