Unique Hepatic Responses to Burn, Sepsis and Trauma: The Adaptability of Innate Immunity in the Face of Different Stimuli

The systemic immune response is considered to be a core response to external insults upon mammalian systems, and consists of a wide variety of effects, including dilation of the vasculature, hypermetabolism and lean body mass loss, acute phase protein synthesis, and macrophage activation., which, if untreated, can eventually lead to multiple organ failure and death [1]. Because of its central role in the metabolic aspects of this systemic response, as well as protein synthesis and innate immune activation, the liver represents a central player in this response whose transcriptional activity following injury is an indicator of critical pathways which are affected by the response, as well as a biomarker for the long term resolution of the host response. Traditionally, the acute phase of the inflammatory response is considered to be critically dangerous, however recent evidence suggests that the long term response to injuries can also result in serious patient complications, including mortality [2]. In order to understand the long term effects of injuries on pathways involved in the inflammatory response, we utilized sepsis and/or burn injury in rats, to assess their whole liver transcriptional response to multiple different injury conditions over the course of 8 days. We used statistical and clustering algorithms to investigate the genetic signatures of hepatic response to this double-injury model and evaluated the progression of the response in a temporal manner, in order to resolve critical pathways involved in the long term, as well as patterns of resolution in each injury.

The rat model receiving a 20 % total body surface area burn (B) followed 2 days later by cecal ligation and puncture (CLP) to induce sepsis was used. Control groups were given a sham-burn (S) treatment, which is identical to burn treatment but the animals were immersed in 37°C water instead of boiling water, or sham cecal ligation and puncture treatment (SCLP), which utilized an identical procedure to the CLP treatment, but did not ligate or puncture the cecum (SCLP condition). Animals were sacrificed at different time points during the first 10 days to collect liver samples for microarray analysis.

Co-expressed and injury-responsive gene groups in the microarray analysis were identified between CLP and SCLP using a “consensus clustering” approach [3] , which selected for temporal variations in the gene expression profiles of both control and treatment groups. Principle component analysis was then used to identify key differences in temporal progression between CLP and SCLP over the entire time course. In order to assess the combined effects of the BCLP injury, a technique known as SAM (Significant Analysis of Microarrays) was utilized [4] to determine a set of significantly varying genes following injury. This method uses multiple gene specific t-tests to verify the significance of the fold change of the double hit injury relative to sepsis alone, followed by a false discovery rate test, which aims to exclude genes whose variations are the result of random chance. 

The consensus clustering method identified a total of 3 major responses in the CLP condition when compared directly to the SCLP control. These clusters are all strongly related to the inflammatory response, with two of the major clusters representing co-expressed sets of pro inflammatory genes, while the third cluster represents co-expressed anti inflammatory genes. Principle component analysis of the concatenated data set (Figure 1) reveals that CLP and SCLP have polar opposite responses along the first, major principle component, while they have almost identical responses along the second. Analysis of genes contributing to each of the principle components reveals that principle component 1 is mapping out a completely unique inflammatory response profile for each injury: the septic injury (CLP) reveals an early pro inflammatory response followed by severe immunosuppression, while the trauma injury (SCLP) shows the opposite dynamics, with early immunosuppression followed by severe pro inflammatory activity.

 The SAM analysis identified a total of 957 genes that were either reduced in expression level or increased by a factor of 2 between the CLP condition, and the CLP condition influenced by prior burn injuries. Critical genes can be seen in Figure 2, which divides the functional differences into metabolic and immune changes: one can observe an up regulation in amino acid degradation over the first day of the response, as well as an early immunosuppression that quickly resolves into increased innate immune function. The long term response appears to be affected in its method of recuperation through a shift towards amino acid biosynthesis and wound repair.  

Figure 1: Principle component analysis of the progression of the long term CLP response compared to SCLP control	Figure 2: Critical genes altered by the burn injury over the progression of the subsequent CLP response

In conclusion, the differences in the responses of the innate immune response to a variety of stimuli necessitate a new paradigm of studying these conditions: inflammation is not a single condition, but rather a collection of responses unique to the injury. By utilizing transcriptional analysis, we were able to observe that the trauma injury has a completely different hepatic response to the septic injury, while burn priming does not appear to change the fundamental underlying response. Clinical strategies aiming to mitigate the disruptive effects of SIRS should therefore be tailored to account for the adaptability of the innate immune system in the face of various stimuli.

References:

1 . Beal, A. L. and F. B. Cerra, Multiple organ failure syndrome in the 1990s. Systemic inflammatory response and organ dysfunction. Jama, 1994. 271(3): p. 226-33.

2 . Sasse, Kent C., et al., Long-term survival after intensive care unit admission with sepsis. Critical Care Medicine, 1995. 23(6): p. 1040-1047.

3 . Nguyen, T. T., R. S. Nowakowski, and I. P. Androulakis, Unsupervised selection of highly coexpressed and noncoexpressed genes using a consensus clustering approach. Omics, 2009. 13(3): p. 219-37.

4 . Tusher, Virginia Goss, Robert Tibshirani, and Gilbert Chu, Significance analysis of microarrays applied to the ionizing radiation response. Proceedings of the National Academy of Sciences, 2001. 98(9): p. 5116-5121.