Heat Shock Protein 70 and the relationship to glutamine in the critically ill child

Abstract

Background: Specific nutrients such as glutamine have an important role in the management of critically ill patients, modulating the host response to stress, improving outcome. The mechanisms of action of glutamine are thought to be through the release of heat shock protein 70 (HSP70), which helps orchestrate an appropriate inflammatory response to critical illness. HSP70 release stimulates immune activation, maintenance of cell homeostasis, cellular protection, preventing apoptotic cell death and is associated with increased survival following severe trauma. However, much of the work considering the use of glutamine and HSP70 in critical illness has been completed in adults. The focus of the current research was therefore to investigate HSP70 release and its relationship to glutamine and inflammatory markers in healthy adults and paediatric volunteers using a whole blood endotoxin model. In addition to this, describing relationships between glutamine, HSP70 and inflammatory mediators, in a cohort of critically ill children. Methodology: An in vitro whole blood endotoxin stimulation model using lipopolysaccharide (LPS) was adapted for use in healthy adult (n=18) and paediatric volunteers (n=25) to investigate the effect of glutamine supplementation on the release of HSP70 and inflammatory mediators (IL-6, IL-8, IL-1β, IL-10, TNF-α). Plasma from children with acute meningococcal disease (MD) (n=143) and during convalescence (n=78) was analysed for levels of HSP70, inflammatory cytokines (IL-6, IL-8, IL-10, TNF-α) and glutamine. Following a clinical folder review, transcriptomic analysis of differential gene expression between acute disease and convalescence was carried out in a subset of children. Results: A whole blood stimulation model was effective in measuring the inflammatory response to endotoxin in vitro over an incubation period of 4 to 24 hours. Glutamine supplementation significantly increased HSP70 expression over time in healthy adult and paediatric volunteers compared to unsupplemented controls (p<0.05). HSP70 levels in vitro rose over time and was correlated with inflammatory mediators (IL-1β, TNF-α, IL-8) in adults, and IL-6, TNF-α in paediatric samples at 4 hours, following glutamine supplementation. There was no relationship between HSP70 and IL-10 in either adults or children. Glutamine supplementation significantly attenuated the release of IL-8 at 24 hours (p<0.05) and in the paediatric in vitro endotoxin model significantly increased TNF-α release at 4 hours (p<0.005). Although there were no other significant differences in the release of inflammatory mediators (IL 6, TNF-α, IL10) between conditions, glutamine supplementation appeared to attenuate the release of inflammatory mediators in adults in contrast to promoting the release of inflammatory mediators in children. In patients with MD, HSP70 was significantly increased (p=<0.0001) in the acute phase [26.7ng/ml; SD± 79.95 (median 5.7ng/ml; range 0.09 – 600.5ng/ml)] vs. convalescence [mean 3.16ng/ml ±SD; 5.67 (median 1.045ng/ml range 0.001 – 37.31ng/ml)]. Mean glutamine levels during the acute phase were 0.31mmol/l; ±SD 0.13 (median 0.31mmol/l; range 0.50 – 0.64) and for convalescence 0.40mmol/l; ±SD 0.14 (median 0.40mmol/l; range 0.72 – 0.64). During the acute phase of illness, glutamine levels were 52% below the normal range, (using 0.6mmol/l as the lower reference range). In convalescent samples, taken on average 55 days later (35 – 135 days), glutamine levels were found to be 26% below normal range. Results from the transcriptomic analysis show that genes associated with HSP70 and inflammatory signalling pathways, glycolysis and gluconeogenesis are significantly upregulated and those relating to adaptive immunity are downregulated during the acute phase of meningococcal disease compared to convalescence. There were numerous correlations between plasma glutamine and HSP70 with genes associated with p38MAPK signalling pathway during the acute phase of meningococcal disease, suggesting that glutamine and HSP70 could mediate the inflammatory response to infection. Conclusion: Glutamine supplementation increased the release of HSP70 in an in vitro endotoxin model in children and adults. Children with MD during the acute phase of their illness have significant glutamine depletion, occurring early during the course of the disease, with a concomitant increase in plasma levels of HSP70 and inflammatory mediators. A transcriptomic analysis of differential gene expression showed that genes associated with HSP70 (HSPA1A and HSPA1B) were differentially upregulated during the acute phase of illness. HSP70 genes and plasma levels of HSP70 were correlated, which was suggestive of a role in meningococcal disease. Relationships were found with plasma levels of HSP70 and glutamine and genes associated with p38MAPK signalling pathway during the acute phase of meningococcal disease, suggesting that glutamine and HSP70 could mediate the inflammatory response to infection. Future work would aim to understand the effect of glutamine supplementation on gene expression during critical illness, especially relative to the differential expression of genes associated with HSP70, inflammatory signalling pathways and those relating to the antigen presentation pathway.