Epub ahead of print June 1, 2015 - doi:10.1189/jlb.5MR0315-127R

Review

The future of murine sepsis and trauma research models Philip A. Efron,1 Alicia M. Mohr, Frederick A. Moore, and Lyle L. Moldawer Department of Surgery, University of Florida College of Medicine, Gainesville, Florida, USA RECEIVED MARCH 26, 2015; REVISED APRIL 24, 2015; ACCEPTED MAY 6, 2015. DOI: 10.1189/jlb.5MR0315-127R

ABSTRACT Recent comparisons of the murine and human transcriptome in health and disease have called into question the appropriateness of the use of murine models for human sepsis and trauma research. More specifically, researchers have debated the suitability of mouse models of severe inflammation that is intended for eventual translation to human patients. This mini-review outlines this recent research, as well as specifically defines the arguments for and against murine models of sepsis and trauma research based on these transcriptional studies. In addition, we review newer advancements in murine models of infection and injury and define what we envision as an evolving but viable future for murine studies of sepsis and trauma. J. Leukoc. Biol. 98: 000–000; 2015.

Introduction Sepsis and traumatic injury are 2 major unresolved issues in the field of critical illness, as both continue to be major causes of human morbidity and mortality. For example, sepsis is an increasingly significant problem throughout the world, and infections remain 1 of the top causes of poor outcomes [1, 2]. Severe sepsis and septic shock have estimated in-hospital mortalities of 29–40% and .50%, respectively [3–5]. Even with improvements in patient outcomes as a result of efforts to standardize initial patient care [6, 7], the total number of deaths as a result of sepsis is growing because of its increasing incidence [8]. Trauma is no different; severe traumatic injury is responsible for a major percentage of deaths worldwide [9]. Although advances in critical care have substantially improved the initial mortality associated with sepsis and trauma, many patients who survive the initial injury succumb from complications [10–14]. To date, immune modulation therapy and pharmacotherapeutic agents have proven disappointing in regards to modifying any human outcome, despite promising results from preclinical studies [15] Much of this preclinical work was done in rodent models [15]. Thus, despite decades of promising preclinical and Abbreviations: CCI = chronic critical illness, CLP = cecal ligation and puncture, ENCODE = Encyclopedia of DNA Elements, GG = Glue Grant, ICU = intensive care unit, PICS = persistent inflammation immunosuppression catabolism syndrome, PT = polytrauma, T-H = trauma-hemorrhage, TBI = traumatic brain injury

0741-5400/15/0098-0001 © Society for Leukocyte Biology

clinical investigations that have elucidated individual aspects of the complex pathophysiology present in infection and injury, our understanding of these entities remains incomplete, and few therapies have successfully improved outcome in these critically ill patients [16–18]. It is clear from this that the best methodology to study severe infection or injury has yet to be determined and is in evolution. With over 2 decades of, at worst, failure and, at best, modest improvement in translational research in sepsis and trauma, many have questioned the animal model and the reductionist approach used to address these 2 syndromes. More specifically, the question being asked is if the millions of dollars invested in rodent sepsis and trauma research have literally wasted investigative opportunities and funding [19]. Simultaneously, many have turned to large human and murine datasets by use of a systems approach to compare transcriptional regulation in humans and mice and in some cases, compare them with clinical outcomes. In this era of big data [20–24], relatively recent projects have focused on a comparison of the transcriptional regulation between mouse and humans. For example, the Mouse ENCODE Consortium mapped transcription, DNase I hypersensitivity, transcription-factor binding, chromatin modifications, and replication domains throughout the mouse genome and compared the results with healthy humans [25]. Shay et al. [26] and the Immunologic Genome Consortium examined the murine and human transcriptional response from multiple hematopoietic cell populations in health. The “Inflammation and Host Response to Injury” GG conducted a 10 yr prospective, multi-institutional observational study in human and murine trauma and burns, with the primary aims of describing the epidemiology, proteomic, and leukocyte genomic responses [27, 28]. All of these programs have provided a unique opportunity to compare systematically large swaths of genomic data in healthy patients and mice, as well as after trauma or burns.

ARGUMENTS AGAINST MURINE MODELS At this time, no interventional biologic therapies proven successful in mouse models of sepsis or trauma have been 1. Correspondence: FCCM, Dept. of Surgery, University of Florida College of Medicine, Room 6116, Shands Hospital, 1600 S.W. Archer Rd., Gainesville, FL 32610-0286, USA. E-mail: [email protected]fl.edu

Volume 98, December 2015

Journal of Leukocyte Biology 1

Copyright 2015 by The Society for Leukocyte Biology.

similarly successful in prospective clinical trials. There may be many reasons for this, including technical issues, such as timing of therapy, identification prospectively of patients who would benefit from such therapies, and appropriate dosing regimens. However, there may also be fundamental differences in how the mouse responds to trauma and sepsis compared with humans. In 2013, Seok et al. [29] published the first report directly comparing the leukocyte genome-wide, transcriptional response in humans and mice with traumatic injury, burns, or endotoxin administration. Although human blood leukocytes have similar genomic responses to different inflammatory diseases, the study concluded that on the genome-wide transcriptional level, murine genomic responses are very different, not only with human responses but also with each other. This article created a furor by questioning whether at the level of the transcriptome, murine models recapitulated the human response to trauma [19, 29, 30]. The authors argued that at the level of the transcriptome, the mouse responds fundamentally different to inflammatory states than do humans [19, 29, 30]. These data went far beyond the earlier data by Shay et al. [26], which argued that in health, the human and murine transcriptional responses were similar in most hematopoietic cell populations. In 2014, the Mouse ENCODE Consortium published its genomic findings [25] in healthy mice and came to different conclusions than Shay et al., [26] but in solid organs and tissues. They argued that transcriptional patterns in healthy mouse organs differed more from their human counterparts than they did from other murine tissues and organs [25]. In addition, a large portion of the cis-regulatory network was widely divergent from mouse to human. Although there was more commonality in human and mouse regulatory transcription networks than in transcription patterns themselves, the authors concluded that “multiple lines of evidence (suggest) that gene expression and their underlying regulatory programs have substantially diverged between the human and mouse lineages, although a subset of core regulatory programs are largely conserved” [25]. The concept that mice do not imitate human genomic responses to severe inflammation is not novel. In addition, some have argued that the mouse immune system is "superior" to the human immune system [31]. Mice can live and flourish in vastly different environments than humans, and the authors have speculated that this has led to evolutionary differences in some of the innate and adaptive immune responses of mice and humans [31]. Although Osuchowski et al. [19] clearly indicate that they believe the conclusions of Seok et al. [29] are not justified, the former authors summarize some of the strengths and weaknesses of mouse models (Table 1). The weaknesses include several limitations that can confound an investigator’s results from murine trauma and sepsis models. Inbred, single-strain mice cannot represent the complex genetic diversity of mice, much less humans [35]. Murine work that uses a single sex is also suboptimal at representing the human condition; in fact, the NIH has initiated a new policy requiring a balanced analysis of genders in mouse research [36]. Whereas most human sepsis and trauma patients are not juveniles, we study juvenile mice, a young cohort known to have significantly different immune responses 2

Journal of Leukocyte Biology

Volume 98, December 2015

than the aged [37]. In addition, the primary circulating immune cell of all mice—the lymphocyte—is different than that of humans, whose primary leukocyte in the blood is the neutrophil [32]. Researchers do not currently have the technology to correct severe organ failure reliably in mice, such as the use of ventilators or renal replacement therapy, or to create humane murine models of sepsis and trauma that truly represent the severity of illness present in human ICUs [19, 32]. Add to this that the temporal dynamics of humans and mice are very different, as humans live much longer than mice, and it understandable why attempts to match murine and human data after infection of injury could be fraught with error [38].

ARGUMENTS IN FAVOR OF MURINE MODELS Although few have refuted the validity of the work by Seok et al. [29], many have disputed its conclusions and interpretations [19, 30, 32]. After publication of this article, there has been a subsequent, robust discussion of the data and its analyses. Some of the furor was undoubtedly directed against the discussion and its tone. In addition, a revised analysis of the same human and murine injury dataset came to completely opposite conclusions. Takao and Miyakawa [30], in a more recent publication, reanalyzed the data of Seok et al. [29] and came to the conclusion that, “…gene patterns in mouse models closely recapitulate those in human inflammatory conditions and strongly argue for the utility of mice as animal models of human disorders” [30]. However, the authors limited their analyses not to the entire human and mouse orthologous genome but to 2300 homologous genes whose expression changed after trauma, burns, or endotoxin in mice and humans. Osuchowski et al. [19] went beyond the genome and argued that murine models have recapitulated the human disease at a number of physiologic and organ system levels [19, 29, 30]. These authors have claimed that it is too early to reject murine models when there is such a long history of demonstrating similar physiologic and immunologic responses to human inflammation. There is undoubtedly truth to these conclusions, but one view of all of the data would suggest that mice and humans accomplish the same immunologic responses and functions through the expression of overlapping but different patterns of genes. Such data would coincide with the conclusions reached by the Mouse ENCODE Consortium [25] but differ from the conclusions of Shay et al. [26]. Such a conclusion could also explain why individual targeted therapies that are successful in mice may not be equally effective in humans. There are, however, a number of additional issues that complicate the comparison of murine with human injury models that are more dependent on the differences in the models and their management than the disparate species response. Most murine models have used a polymicrobial sepsis produced by CLP, or trauma produced by hemorrhagic shock and surgical injury [39, 40]. Although these models have been well standardized and are accepted by the scientific community, they are generally assumed to poorly recapitulate human sepsis,

www.jleukbio.org

Efron et al.

Murine sepsis and trauma models

TABLE 1. Partial list of limitations of current murine models used to conduct sepsis and trauma researcha Limitation Strain Sex Age

Total circulating leukocyte comparisons Model and severity mismatch Analysis of responses in a single source

Compartmentalization of the immune-inflammatory response Genome-to-protein gap Mismatching the temporal response patterns

Explanation Single mouse strains cannot be expected to represent the entire genetic diversity of the entire mouse and human population. Single mouse sex studies cannot represent human mixed sex-disease processes or datasets. Mortality is increased significantly in the elderly in sepsis and trauma, and the incidence is increased in the former as well. Typical murine studies use animals that are the age equivalent of an adolescent human and thus, cannot reflect the majority of sepsis patients nor those at the greatest risk for poor outcomes after trauma [32–34]. The composition of circulating leukocytes before and after inflammation varies between mice and humans [32]. Previously accepted models of murine sepsis and trauma mildly imitate the severe and prolonged physiologic effect that occurs in humans. Blood and circulating leukocytes are used for murine-human comparisons as a result of our ability to obtain human blood samples with relative ease. However, inflammation is a systemic phenomenon that affects virtually every organ, and the interspecies comparison of these other organs (i.e., liver) after injury or infection of the host illustrates that they can be quite similar. Changes in the circulation are not tantamount to the changes that occur in other organs and local tissues. Transcription does not equal translation. Twenty-four hours in a mouse is not the equivalent of 24 h in a human being [32].

a

Adapted from Osuchowski et al. [19].

trauma, and their management. In fact, with the use of the same genomic analyses performed by the GG and Mouse ENCODE Consortium, we compared the genomic responses among two models of sepsis, a CLP, and the intraperitoneal injection of cecal slurry [25, 29, 41]. Although both are models of polymicrobial peritoneal sepsis in the same aged mouse strain with fecal flora, the genomic responses differed significantly [41]. In actuality, the differences in genome-wide transcription between the two models of sepsis were greater than a comparison of the CLP (polymicrobial sepsis) with a much more severe, multicompartmental trauma model [41] (unpublished data). These findings are markedly different than the findings of Seok et al. [29], who failed to see any relationship between leukocyte gene-expression patterns from burned mice and mice receiving hemorrhagic shock and surgical injury [29]. In fact, the disparities of these findings with that of Seok et al. [29] argue that the choice of animal models and their timing are essential criteria for comparing results among different murine models and with humans. Although our work indicates a much greater similarity in the total leukocyte genomic response between mouse models of infectious and noninfectious inflammation, it is essential to heed the advice of Osuchowski et al. [19]. These authors emphasized the need for all investigators in the future to consider the relevance of their models to human disease before embarking on sepsis and trauma research in mice (Table 1) [19].

REVISION, RATHER THAN REJECTION, OF MURINE MODELS “Strive for progress, not perfection!”—unknown author Most advances in human medical science have not been made from direct human experimentation. In fact, the evaluation of

www.jleukbio.org

new therapeutic advances directly in human subjects is a clear violation of the ethical guidelines embraced by the Nuremberg code and the Declaration of Helsinki, which states that animal testing is a required prerequisite before human testing. There have been multiple examples of how past mouse-to-human translational work has successfully advanced the field of inflammation research and provided benefit to humankind [19]. However, much of the trepidation surrounding the use of murine models for sepsis and trauma research has been a result of our inability to identify a single therapeutic intervention that has translated into a benefit for patients. Some of this can be blamed on an inappropriate understanding of the complexity of severe inflammation in the mammalian host and an over-reliance on reductionist approaches. A modification of Koch’s postulates has often been proposed for the identification of an individual mediator or therapeutic approach for the treatment of sepsis or severe trauma. This has led to an experimental approach based on the linear thinking of the following: that 1) “mediator A produces disease when administered to a healthy animal”; 2) “mediator A is produced in the disease, but not in healthy individuals”; and 3) “preventing mediator A production blocks the expression of the disease.” Based on this reasoning, a multitude of individual mediators has been explored as potential targets for sepsis or trauma interventions. Although such reductionist approaches to experimentation have been absolutely vital to the scientific process and to scientific progress, sepsis and trauma are not "simple and clean" monogenic diseases. Rather, trauma and sepsis are polygenic phenotypes with multiple simultaneous aberrations [27]. This "storm" (whether cytokine or genomic [27]) that occurs with severe infection or injury affects every aspect of mammalian biology— multiple molecules, genes, cells, tissues, compartments, and

Volume 98, December 2015

Journal of Leukocyte Biology 3

organs, as well as the systemic interaction of the individual as a whole. Although hindsight is perfect, it should be of no surprise that we have made only limited therapeutic progress in regards to sepsis and trauma translational research; in the interest of following appropriate scientific methodology (limiting the number of variables), we have sacrificed our capacity to reproduce the complexity of human inflammation appropriately. Ask any ICU physician if there are a multitude of septic middle-/high-school children in their ICUs. The answer will be “no.” Yet, we study 6- to 12-wk-old mice. Ask which trauma patients under the intensivist’s care have poor outcomes, and the physician will tell you individuals with a combination of prolonged hypovolemic shock, multiple severe compartmental injuries, and blood product transfusion. Yet, accepted historical models of murine trauma only induce a transient period of hemorrhagic shock, are not multicompartmental, and rarely consider the effects of massive blood transfusion. The improvement of murine models of sepsis and trauma is neither a novel concept nor is there a single definitive solution. When experiments moved from in vitro studies to animal models, endotoxin challenge in young, healthy mice was obviously a first effort. It has become increasingly clear that more complex and pathologically relevant models, such as CLP in mice, or larger animal hemodynamic models are probably more informative. In fact, a step-wise approach to recapitulating the human condition in rodents, while refraining from inhumane treatment, is the keystone to our making progress in the field. Beginning more than a decade ago, Daniel Remick and his colleagues [42, 43] began promoting the use of antibiotics and crystalloid volume resuscitation after CLP in mice to imitate better the response to severe human infection; a greater infectious nidus could be used, but the mouse could survive the acute phase of sepsis, as is typical for human patients because of aggressive interventional support (Table 2). As opposed to simple LPS injections, the pathologic findings of apoptotic cell death of lymphocytes and gastrointestinal epithelial cells that are one of the hallmarks of human sepsis occur in the CLP animal model [39, 53, 54]. It should be noted, though, that antibiotics and fluids only partly recreate the clinical situation of humans in the ICU. Current models generally give mice enough antibiotics to prolong survival but we are not able to take the same curative approach as we do in humans. However, gene expression studies of lymphocytes in septic mice and in septic patients show very similar findings for the genes that regulate cell death, further reinforcing the appropriateness of the murine CLP model to study human sepsis [55]. Our group too realized that modifications were required in our current murine models to make progress regarding translational research. Our laboratory examined large datasets, such as the GG, to determine what modifications should be implemented by examining those cohorts who had worse outcomes with human severe trauma and apply those conditions to our animal model [27, 28, 56]. Blood transfusion, age, and injury severity are all significantly associated with worsened outcomes in trauma (and sepsis for the former 2 characteristics) [28, 33, 57]. Thus, we have examined blood transfusion in sepsis [57] and aging as it relates to the innate-immune response in sepsis and trauma [34, 41]. In addition, we have created a novel PT model to reflect better the effects of hemorrhagic shock and 4

Journal of Leukocyte Biology

Volume 98, December 2015

multicompartmental injury in humans [40]. In fact, this latter murine model allowed us to expand subsequently on the original Seok et al. [29] publication that failed to demonstrate similar transcriptional regulation between human trauma and an established murine trauma model. Specifically, we could show that increasing the severity of the murine model resulted in better correlations between the genome-wide murine and human leukocyte transcriptomic response to trauma [32]. Although the overall improvement in that correlation was modest, the inflammatory response between the 2 mammals was remarkably similar at the level of the leukocyte transcriptome [29, 32]. The expansion of the severity and resemblance of rodent trauma and sepsis models to improve their similarity to the human condition is not a unique approach to our laboratory. A group from Rutgers-New Jersey Medical School has created a rat model of combined hemorrhagic shock and pulmonary contusion, an extremely common phenotype for patients in trauma ICUs [58]. Even more, they subsequently performed intermittent restraint and repositioning with alarms in the rodent for up to 1 wk, inducing chronic stress in the animal that well recapitulates the environment in a typical ICU [58]. Other authors have combined previous standard models, specifically T-H, followed by polymicrobial sepsis to create a mouse acute lung injury/acute respiratory distress model that attempts to recapitulate what patients experience in the ICU; the incidence of sepsis after trauma is substantial [6, 7, 44, 45, 59]. Researchers have created a standardized, reproducible, and clinically relevant double-hit mouse model induced by a focused blast wave to create chest trauma, followed by polymicrobial sepsis [60]. Others have further expanded immune research to include TBI by use of a clinically relevant murine model that uses an impacting rod directly on the skull [51]. However, even with all of these efforts and novel methodologies, sepsis and trauma murine research is still in a state of evolution and can be improved. For example, researchers still need to work toward creating better mouse pneumonia models for primary sepsis and sepsis subsequent to trauma. The value of these efforts rests with a realization that human sepsis and trauma are much more complicated phenomena than are engendered presently in our current animal models. What remains unanswered is whether the presence of these fundamentally different transcriptional responses in mice, compared with humans, can be overcome by simply making the models more closely imitate the human disease—first raised by Seok et al. [29] and addressed further by Efron and coworkers [32], Osuchowski et al. [19], and Takao and Miyakawa [30]. Some have concluded that even with murine models of trauma and sepsis that more closely replicate the human condition, the differences at the transcriptional level between the 2 mammals are too great to put any reliance on the ability of murine models for human medical research [29, 61, 62]. Unfortunately, there is no easy answer to this question nor is there an easy approach to address the challenge. The other difficulty facing murine models of trauma or sepsis is in replicating the timing of the host response. In addition to improving the intensity of the rodent models, many investigators have shifted their focus from the acute phase of trauma and

www.jleukbio.org

Efron et al.

Murine sepsis and trauma models

TABLE 2. Examples of novel rodent models of sepsis and trauma Model name CLP T-H model followed by CLP CLP followed by intraperitoneal fungal infection Zymosan PT T-H model with chronic restraint Pseudofracture with/without T-H TBI Burn injury followed by infection

Modifications to make more similar to human condition Antibiotics and crystalloid resuscitation [42, 43] 2-Hit model of hemorrhagic shock followed by polymicrobial sepsis, usually 24 h later [44, 45] Nonlethal polymicrobial sepsis followed by an intraperitoneal Candida infection 4 d later [46, 47] Although not novel in and of itself, the authors’ methodology of examination of the rodents and their emphasis on examination of the chronic effects of inflammation are novel [48]. Ninety minutes of hemorrhagic shock are followed by multicompartmental injuries (long bone fracture and cecectomy), creating a much higher injury severity score for the mouse [32, 40]. Hemorrhagic shock combined with pulmonary contusion and subsequent chronic restraint of the rodent [49] Recreates the features of long bone fractures without breaking the native bone. A bilateral muscle crush injury to the hindlimbs is performed followed by injection of a bone solution into these injured muscles [50]. Clinically relevant murine model that uses an impacting rod directly on the skull [51] Clinically relevant severe burn model that subsequently subjects the mouse to a secondary infection, such as pneumonia. These secondary infections and morbidity and mortality to them are common in human burn patients [52].

sepsis to the chronic response. Acute death from sepsis and trauma is no longer the major cause of death in the ICU [13]. More commonly, we are finding patients surviving in the short term to go on to CCI and PICS [13]. Murine models have been notoriously poor at recapitulating CCI. Most existing trauma models study only the 1st 72–96 h after the response [40], and a 30% burn injury in the mouse will be healed completely by 14 d, where the host inflammatory response in patients with equivalent burns will last months and not days [63]. Several attempts have been made to create models of CCI. One such approach has been the sterile administration of zymosan, a fungal cell-wall product [48]. This produces a triphasic response associated with early mortality, a transient recovery period followed ultimately by death from multiple organ failure, 2–3 wk after induction [48]. Although on the surface, the timing and the phenotype of the inflammation recapitulate human CCI, a detailed investigation has revealed that the model is one of persistent hyperinflammation, without evidence of a simultaneous immune suppression, as is seen in humans with CCI and PICS [49]. Other approaches involve nonlethal peritonitis with a nidus of necrotic tissue created by a nonlethal CLP. We have shown that an LD10 model of CLP produces a PICS-like syndrome in mice that demonstrates persistent inflammation and immunosuppression for up to 8 wk [64]. However, the model actually transitions from polymicrobial sepsis to a walled-off peritoneal abscess with necrotic tissue at its center and no evidence of systemic dissemination of bacteria. The Hotchkiss laboratory [46, 47] has sought to recreate and examine CCI through a similar nonlethal CLP, followed by an intravenous Candida infection. This obviates some of the challenges associated with a longterm abscess and superimposes a nosocomial infection that uses an opportunistic organism [46, 47]. Although Candida may not be the most common secondary infection in ICUs, it is more common than appreciated previously [46, 47] and does provide a reasonable murine model.

www.jleukbio.org

"REVERSE-TRANSLATIONAL" MURINE RESEARCH Traditional sepsis and trauma research in mice have followed a specific paradigm (Fig. 1). Investigators would identify a general problem in human medicine—in our case, the extremely poor outcomes that occur after sepsis and trauma. The researchers would subsequently perform experiments in a mouse model that were essentially aimed to replicate the findings in human injury or infection. This model seemed appropriate based on the understanding of the disease process at the time, and the model accurately reflected an aspect of inflammation or immunity in humans. Then, an appropriate reductionist approach was undertaken in this murine model, usually following Koch’s postulates, often with impressive results. The laboratory would then attempt to translate its work back into human patients—bench to bedside! Uniformly, the human work has failed, yielding either negative or marginal findings, and leaving researchers and reviewers confounded and disgruntled. The reasoning behind the lack of success for this traditional approach in sepsis and trauma research was discussed previously: 1) a lack of comprehension of the complexity of the inflammatory storm related to severe infection or injury; 2) a lack of understanding of the profoundly exacerbated and prolonged effect that sepsis and trauma had on the host immune system; and 3) the desire to identify a single "magic bullet" or treatment. Current investigators have the luxury of standing on the shoulders of these previous individuals who have laid the foundation for our present research. We benefit not only from their previous work but also from our contemporary technology that is able to evaluate complex systems biology. However, funding agencies and big pharma have become reticent to continue the same approach to sepsis and trauma research, especially as it relates to rodent experimentation. Investigators will have to prove that their methodology and/or rodent model are clinically relevant—an

Volume 98, December 2015

Journal of Leukocyte Biology 5

Figure 1. Comparison of previously used methodology for and our proposed future approach to rodent bench and human bedside sepsis and trauma research. (A) The traditional paradigm of rodent sepsis and trauma research, which would eventually progress to human translational investigations. (B) Our proposal for future rodent sepsis and trauma studies. We recommend that investigators use “reverse translation” to prove that their methodology and/or rodent model is clinically relevant. This would be followed eventually by translational research from the rodent back to the human.

approach we have termed reverse translation (Fig. 1). This will involve not only determining a generalized medical problem in humans but also identifying a more specific phenotype or dysfunction in human patients before working with animal models. Subsequently, the investigators will have to "reverse translate" this work to the rodent—bedside to bench—confirming that the mouse model used indeed replicates the human condition at the mechanistic level and therefore, is a useful study approach. Only then can specific hypothesis-based work and interventions be performed, as required for drug development. Failure to demonstrate that the specific condition and hypothesis proposed in the mouse are mechanistically similar to humans will only lead to results that cannot be ultimately translated. Although this is an oversimplified description of what would normally be a more complex pathway to human research, we anticipate that this will become the standard for future studies, requiring investigators to justify their animal model and mechanistic approach. Many models, such as the PT model, as well as the chronic restraint stress after injury and shock model, have already delineated this in their publications [32, 40, 49].

AUTHORSHIP

CONCLUDING REMARKS

DISCLOSURES

All authors contributed to writing the manuscript.

ACKNOWLEDGMENTS The GG database was supported by the Inflammation and the Host Response to Injury Large Scale Collaborative Research Program (GG U54 GM062119), awarded to Dr. Ronald G. Tompkins, Massachusetts General Hospital, by the U.S. National Institutes of Health (NIH) National Institute of General Medical Sciences (NIGMS). The work represents a secondary use of this public database, and the conclusions and discussion are the authors’ and do not necessarily represent the views of either Massachusetts General Hospital or the NIGMS. This work was also supported by NIH Grants R01 GM40586-24 and R01 GM-081923-06, awarded by the NIGMS. A.M.M. was supported by the NIH NIGMS, Grant R01 GM105893-01A1. In addition, P.A.E. was supported by P30 AG028740 from the NIH National Institute on Aging and R01 GM113945-01 (NIGMS). Finally, P.A.E., F.A.M., and L.L.M. were supported by P50 GM111152-01 (NIGMS).

No conflict of, or competing, interests have been declared.

In summary, current investigators must be willing to learn from the past to improve the future. Murine sepsis and trauma research is far from over, but it has to evolve. We are required to improve or adapt the well-established models of the past to increase their clinical relevancy and to demonstrate this scientific appropriateness to the audience/reviewers before implementation of vast rodent experimentation. It is only in this “future of murine sepsis and trauma research models” that we will be able to unravel the complexity of the human inflammatory response and identify modifications of the immune system that will improve the outcomes of the next generation of critically ill patients. 6

Journal of Leukocyte Biology

Volume 98, December 2015

REFERENCES

1. Mares, C. A., Sharma, J., Ojeda, S. S., Li, Q., Campos, J. A., Morris, E. G., Coalson, J. J., Teale, J. M. (2010) Attenuated response of aged mice to respiratory Francisella novicida is characterized by reduced cell death and absence of subsequent hypercytokinemia. PLoS ONE 5, e14088. 2. Turnbull, I. R., Clark, A. T., Stromberg, P. E., Dixon, D. J., Woolsey, C. A., Davis, C. G., Hotchkiss, R. S., Buchman, T. G., Coopersmith, C. M. (2009) Effects of aging on the immunopathologic response to sepsis. Crit. Care Med. 37, 1018–1023. 3. Sands, K. E., Bates, D. W., Lanken, P. N., Graman, P. S., Hibberd, P. L., Kahn, K. L., Parsonnet, J., Panzer, R., Orav, E. J., Snydman, D. R., Black, E., Schwartz, J. S., Moore, R., Johnson, Jr., B. L., Platt, R.; Academic

www.jleukbio.org

Efron et al.

4.

5.

6. 7. 8. 9.

10.

11. 12. 13.

14.

15. 16. 17. 18.

19.

20. 21. 22.

23. 24.

25.

Medical Center Consortium Sepsis Project Working Group. (1997) Epidemiology of sepsis syndrome in 8 academic medical centers. JAMA 278, 234–240. Starr, M. E., Ueda, J., Yamamoto, S., Evers, B. M., Saito, H. (2011) The effects of aging on pulmonary oxidative damage, protein nitration, and extracellular superoxide dismutase down-regulation during systemic inflammation. Free Radic. Biol. Med. 50, 371–380. Dombrovskiy, V. Y., Martin, A. A., Sunderram, J., Paz, H. L. (2007) Rapid increase in hospitalization and mortality rates for severe sepsis in the United States: a trend analysis from 1993 to 2003. Crit. Care Med. 35, 1244–1250. Moore, L. J., McKinley, B. A., Turner, K. L., Todd, S. R., Sucher, J. F., Valdivia, A., Sailors, R. M., Kao, L. S., Moore, F. A. (2011) The epidemiology of sepsis in general surgery patients. J. Trauma 70, 672–680. Moore, L. J., Moore, F. A., Todd, S. R., Jones, S. L., Turner, K. L., Bass, B. L. (2010) Sepsis in general surgery: the 2005–2007 national surgical quality improvement program perspective. Arch. Surg. 145, 695–700. Xiao, H., Siddiqui, J., Remick, D. G. (2006) Mechanisms of mortality in early and late sepsis. Infect. Immun. 74, 5227–5235. Menzel, C. L., Pfeifer, R., Darwiche, S. S., Kobbe, P., Gill, R., Shapiro, R. A., Loughran, P., Vodovotz, Y., Scott, M. J., Zenati, M. S., Billiar, T. R., Pape, H. C. (2011) Models of lower extremity damage in mice: time course of organ damage and immune response. J. Surg. Res. 166, e149–e156. Probst, C., Pape, H. C., Hildebrand, F., Regel, G., Mahlke, L., Giannoudis, P., Krettek, C., Grotz, M. R. (2009) 30 years of polytrauma care: an analysis of the change in strategies and results of 4849 cases treated at a single institution. Injury 40, 77–83. Probst, C., Zelle, B. A., Sittaro, N. A., Lohse, R., Krettek, C., Pape, H. C. (2009) Late death after multiple severe trauma: when does it occur and what are the causes? J. Trauma 66, 1212–1217. Sasser, S. M., Varghese, M., Joshipura, M., Kellermann, A. (2006) Preventing death and disability through the timely provision of prehospital trauma care. Bull. World Health Organ. 84, 507. Gentile, L. F., Cuenca, A. G., Efron, P. A., Ang, D., Bihorac, A., McKinley, B. A., Moldawer, L. L., Moore, F. A. (2012) Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care. J. Trauma Acute Care Surg. 72, 1491–1501. Vanzant, E. L., Lopez, C. M., Ozrazgat-Baslanti, T., Ungaro, R., Davis, R., Cuenca, A. G., Gentile, L. F., Nacionales, D. C., Cuenca, A. L., Bihorac, A., Leeuwenburgh, C., Lanz, J., Baker, H. V., McKinley, B., Moldawer, L. L., Moore, F. A., Efron, P. A. (2014) Persistent inflammation, immunosuppression, and catabolism syndrome after severe blunt trauma. J. Trauma Acute Care Surg. 76, 21–30. Artenstein, A. W., Higgins, T. L., Opal, S. M. (2013) Sepsis and scientific revolutions. Crit. Care Med. 41, 2770–2772. Baue, A. E. (1997) Multiple organ failure, multiple organ dysfunction syndrome, and systemic inflammatory response syndrome. Why no magic bullets? Arch. Surg. 132, 703–707. Baue, A. E. (2003) Sepsis, systemic inflammatory response syndrome, multiple organ dysfunction syndrome, and multiple organ failure: are trauma surgeons lumpers or splitters? J. Trauma 55, 997–998. Napolitano, L. M., Fabian, T. C., Kelly, K. M., Bailey, J. A., Block, E. F., Langholff, W., Enny, C., Corwin, H. L. (2008) Improved survival of critically ill trauma patients treated with recombinant human erythropoietin. J. Trauma 65, 285–297, discussion 297–299. Osuchowski, M. F., Remick, D. G., Lederer, J. A., Lang, C. H., Aasen, A. O., Aibiki, M., Azevedo, L. C., Bahrami, S., Boros, M., Cooney, R., Cuzzocrea, S., Jiang, Y., Junger, W. G., Hirasawa, H., Hotchkiss, R. S., Li, X. A., Radermacher, P., Redl, H., Salomao, R., Soebandrio, A., Thiemermann, C., Vincent, J. L., Ward, P., Yao, Y. M., Yu, H. P., Zingarelli, B., Chaudry, I. H. (2014) Abandon the mouse research ship? Not just yet! Shock 41, 463–475. Chute, C. G., Ullman-Cullere, M., Wood, G. M., Lin, S. M., He, M., Pathak, J. (2013) Some experiences and opportunities for big data in translational research. Genet. Med. 15, 802–809. Kannry, J. L. Williams, M. S. (2013) Integration of genomics into the electronic health record: mapping terra incognita. Genet. Med. 15, 757–760. Welch, B. M. Kawamoto, K. (2013) The need for clinical decision support integrated with the electronic health record for the clinical application of whole genome sequencing information. J. Pers. Med. 3, 306–325. Collins, F. S., Varmus, H. (2015) A new initiative on precision medicine. N. Engl. J. Med. 372, 793–795. Overby, C. L., Tarczy-Hornoch, P., Hoath, J. I., Kalet, I. J., Veenstra, D. L. (2010) Feasibility of incorporating genomic knowledge into electronic medical records for pharmacogenomic clinical decision support. BMC Bioinformatics 11 (Suppl 9), S10. Yue, F., Cheng, Y., Breschi, A., Vierstra, J., Wu, W., Ryba, T., Sandstrom, R., Ma, Z., Davis, C., Pope, B. D., Shen, Y., Pervouchine, D. D., Djebali, S., Thurman, R. E., Kaul, R., Rynes, E., Kirilusha, A., Marinov, G. K., Williams, B. A., Trout, D., Amrhein, H., Fisher-Aylor, K., Antoshechkin, I.,

www.jleukbio.org

26.

27.

28.

29.

30. 31. 32.

33.

34.

35. 36.

Murine sepsis and trauma models

DeSalvo, G., See, L. H., Fastuca, M., Drenkow, J., Zaleski, C., Dobin, A., Prieto, P., Lagarde, J., Bussotti, G., Tanzer, A., Denas, O., Li, K., Bender, M. A., Zhang, M., Byron, R., Groudine, M. T., McCleary, D., Pham, L., Ye, Z., Kuan, S., Edsall, L., Wu, Y. C., Rasmussen, M. D., Bansal, M. S., Kellis, M., Keller, C. A., Morrissey, C. S., Mishra, T., Jain, D., Dogan, N., Harris, R. S., Cayting, P., Kawli, T., Boyle, A. P., Euskirchen, G., Kundaje, A., Lin, S., Lin, Y., Jansen, C., Malladi, V. S., Cline, M. S., Erickson, D. T., Kirkup, V. M., Learned, K., Sloan, C. A., Rosenbloom, K. R., Lacerda de Sousa, B., Beal, K., Pignatelli, M., Flicek, P., Lian, J., Kahveci, T., Lee, D., Kent, W. J., Ramalho Santos, M., Herrero, J., Notredame, C., Johnson, A., Vong, S., Lee, K., Bates, D., Neri, F., Diegel, M., Canfield, T., Sabo, P. J., Wilken, M. S., Reh, T. A., Giste, E., Shafer, A., Kutyavin, T., Haugen, E., Dunn, D., Reynolds, A. P., Neph, S., Humbert, R., Hansen, R. S., De Bruijn, M., Selleri, L., Rudensky, A., Josefowicz, S., Samstein, R., Eichler, E. E., Orkin, S. H., Levasseur, D., Papayannopoulou, T., Chang, K. H., Skoultchi, A., Gosh, S., Disteche, C., Treuting, P., Wang, Y., Weiss, M. J., Blobel, G. A., Cao, X., Zhong, S., Wang, T., Good, P. J., Lowdon, R. F., Adams, L. B., Zhou, X. Q., Pazin, M. J., Feingold, E. A., Wold, B., Taylor, J., Mortazavi, A., Weissman, S. M., Stamatoyannopoulos, J. A., Snyder, M. P., Guigo, R., Gingeras, T. R., Gilbert, D. M., Hardison, R. C., Beer, M. A., Ren, B.; Mouse ENCODE Consortium. (2014) A comparative encyclopedia of DNA elements in the mouse genome. Nature 515, 355–364. Shay, T., Jojic, V., Zuk, O., Rothamel, K., Puyraimond-Zemmour, D., Feng, T., Wakamatsu, E., Benoist, C., Koller, D., Regev, A.; ImmGen Consortium. (2013) Conservation and divergence in the transcriptional programs of the human and mouse immune systems. Proc. Natl. Acad. Sci. USA 110, 2946–2951. Xiao, W., Mindrinos, M. N., Seok, J., Cuschieri, J., Cuenca, A. G., Gao, H., Hayden, D. L., Hennessy, L., Moore, E. E., Minei, J. P., Bankey, P. E., Johnson, J. L., Sperry, J., Nathens, A. B., Billiar, T. R., West, M. A., Brownstein, B. H., Mason, P. H., Baker, H. V., Finnerty, C. C., Jeschke, ´ M. G., Lopez, M. C., Klein, M. B., Gamelli, R. L., Gibran, N. S., Arnoldo, B., Xu, W., Zhang, Y., Calvano, S. E., McDonald-Smith, G. P., Schoenfeld, D. A., Storey, J. D., Cobb, J. P., Warren, H. S., Moldawer, L. L., Herndon, D. N., Lowry, S. F., Maier, R. V., Davis, R. W., Tompkins, R. G.; Inflammation and Host Response to Injury Large-Scale Collaborative Research Program. (2011) A genomic storm in critically injured humans. J. Exp. Med. 208, 2581–2590. Cuschieri, J., Johnson, J. L., Sperry, J., West, M. A., Moore, E. E., Minei, J. P., Bankey, P. E., Nathens, A. B., Cuenca, A. G., Efron, P. A., Hennessy, L., Xiao, W., Mindrinos, M. N., McDonald-Smith, G. P., Mason, P. H., Billiar, T. R., Schoenfeld, D. A., Warren, H. S., Cobb, J. P., Moldawer, L. L., Davis, R. W., Maier, R. V., Tompkins, R. G.; Inflammation and Host Response to Injury, Large Scale Collaborative Research Program. (2012) Benchmarking outcomes in the critically injured trauma patient and the effect of implementing standard operating procedures. Ann. Surg. 255, 993–999. Seok, J., Warren, H. S., Cuenca, A. G., Mindrinos, M. N., Baker, H. V., Xu, W., Richards, D. R., McDonald-Smith, G. P., Gao, H., Hennessy, L., Finnerty, C. C., Lopez, C. M., Honari, S., Moore, E. E., Minei, J. P., Cuschieri, J., Bankey, P. E., Johnson, J. L., Sperry, J., Nathens, A. B., Billiar, T. R., West, M. A., Jeschke, M. G., Klein, M. B., Gamelli, R. L., Gibran, N. S., Brownstein, B. H., Miller-Graziano, C., Calvano, S. E., Mason, P. H., Cobb, J. P., Rahme, L. G., Lowry, S. F., Maier, R. V., Moldawer, L. L., Herndon, D. N., Davis, R. W., Xiao, W., Tompkins, R. G. (2013) Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc. Natl. Acad. Sci. USA. 110, 3507–3512. Takao, K., Miyakawa, T. (2015) Genomic responses in mouse models greatly mimic human inflammatory diseases. Proc. Natl. Acad. Sci. USA 112, 1167–1172. Mestas, J., Hughes, C. C. (2004) Of mice and not men: differences between mouse and human immunology. J. Immunol. 172, 2731–2738. Gentile, L. F., Nacionales, D. C., Lopez, M. C., Vanzant, E., Cuenca, A., Cuenca, A. G., Ungaro, R., Baslanti, T. O., McKinley, B. A., Bihorac, A., Cuschieri, J., Maier, R. V., Moore, F. A., Leeuwenburgh, C., Baker, H. V., Moldawer, L. L., Efron, P. A. (2014) A better understanding of why murine models of trauma do not recapitulate the human syndrome. Crit. Care Med. 42, 1406–1413. Vanzant, E. L., Hilton, R. E., Lopez, C. M., Zhang, J., Ungaro, R. F., Gentile, L. F., Szpila, B. E., Maier, R. V., Cuschieri, J., Bihorac, A., Leeuwenburgh, C., Moore, F. A., Baker, H. V., Moldawer, L. L., Brakenridge, S. C., Efron, P. A.; Inflammation and Host Response to Injury Investigators. (2015) Advanced age is associated with worsened outcomes and a unique genomic response in severely injured patients with hemorrhagic shock. Crit. Care 19, 77. Nacionales, D. C., Gentile, L. F., Vanzant, E., Lopez, M. C., Cuenca, A., Cuenca, A. G., Ungaro, R., Li, Y., Baslanti, T. O., Bihorac, A., Moore, F. A., Baker, H. V., Leeuwenburgh, C., Moldawer, L. L., Efron, P. A. (2014) Aged mice are unable to mount an effective myeloid response to sepsis. J. Immunol. 192, 612–622. Fink, M. P. (2014) Animal models of sepsis. Virulence 5, 143–153. Clayton, J. A., Collins, F. S. (2014) Policy: NIH to balance sex in cell and animal studies. Nature 509, 282–283.

Volume 98, December 2015

Journal of Leukocyte Biology 7

37. Turnbull, I. R., Wlzorek, J. J., Osborne, D., Hotchkiss, R. S., Coopersmith, C. M., Buchman, T. G. (2003) Effects of age on mortality and antibiotic efficacy in cecal ligation and puncture. Shock 19, 310–313. 38. Perlman, H., Budinger, G. R., Ward, P. A. (2013) Humanizing the mouse: in defense of murine models of critical illness. Am. J. Respir. Crit. Care Med. 187, 898–900. 39. Cuenca, A. G., Delano, M. J., Kelly-Scumpia, K. M., Moldawer, L. L., Efron, P. A. (2010) Cecal ligation and puncture. Curr. Protoc. Immunol. Chapter 19, Unit 19.13. 40. Gentile, L. F., Nacionales, D. C., Cuenca, A. G., Armbruster, M., Ungaro, R. F., Abouhamze, A. S., Lopez, C., Baker, H. V., Moore, F. A., Ang, D. N., Efron, P. A. (2013) Identification and description of a novel murine model for polytrauma and shock. Crit. Care Med. 41, 1075–1085. 41. Gentile, L. F., Nacionales, D. C., Lopez, M. C., Vanzant, E., Cuenca, A., Cuenca, A. G., Ungaro, R., Szpila, B. E., Larson, S., Joseph, A., Moore, F. A., Leeuwenburgh, C., Baker, H. V., Moldawer, L. L., Efron, P. A. (2014) Protective immunity and defects in the neonatal and elderly immune response to sepsis. J. Immunol. 192, 3156–3165. 42. Newcomb, D., Bolgos, G., Green, L., Remick, D. G. (1998) Antibiotic treatment influences outcome in murine sepsis: mediators of increased morbidity. Shock 10, 110–117. 43. Osuchowski, M. F., Craciun, F., Weixelbaumer, K. M., Duffy, E. R., Remick, D. G. (2012) Sepsis chronically in MARS: systemic cytokine responses are always mixed regardless of the outcome, magnitude, or phase of sepsis. J. Immunol. 189, 4648–4656. 44. Stephan, R. N., Kupper, T. S., Geha, A. S., Baue, A. E., Chaudry, I. H. (1987) Hemorrhage without tissue trauma produces immunosuppression and enhances susceptibility to sepsis. Arch. Surg. 122, 62–68. 45. Monaghan, S. F., Thakkar, R. K., Heffernan, D. S., Huang, X., Chung, C. S., Lomas-Neira, J., Cioffi, W. G., Ayala, A. (2012) Mechanisms of indirect acute lung injury: a novel role for the coinhibitory receptor, programmed death-1. Ann. Surg. 255, 158–164. 46. Muenzer, J. T., Davis, C. G., Chang, K., Schmidt, R. E., Dunne, W. M., Coopersmith, C. M., Hotchkiss, R. S. (2010) Characterization and modulation of the immunosuppressive phase of sepsis. Infect. Immun. 78, 1582–1592. 47. Unsinger, J., Burnham, C. A., McDonough, J., Morre, M., Prakash, P. S., Caldwell, C. C., Dunne, Jr., W. M., Hotchkiss, R. S. (2012) Interleukin-7 ameliorates immune dysfunction and improves survival in a 2-hit model of fungal sepsis. J. Infect. Dis. 206, 606–616. 48. Hill, N. E., Saeed, S., Phadke, R., Ellis, M. J., Chambers, D., Wilson, D. R., Castells, J., Morel, J., Freysennet, D. G., Brett, S. J., Murphy, K. G., Singer, M. (2015) Detailed characterization of a long-term rodent model of critical illness and recovery. Crit. Care Med. 43, e84–e96. 49. Von Asmuth, E. J., Maessen, J. G., van der Linden, C. J., Buurman, W. A. (1990) Tumour necrosis factor alpha (TNF-alpha) and interleukin 6 in a zymosan-induced shock model. Scand. J. Immunol. 32, 313–319. 50. Gore, A. V., Bible, L. E., Livingston, D. H., Mohr, A. M., Sifri, Z. C. (2015) Can mesenchymal stem cells reverse chronic stress-induced impairment of lung healing following traumatic injury? J. Trauma and Acute Care Surg. 78, 767–772. 51. Schwulst, S. J., Trahanas, D. M., Saber, R., Perlman, H. (2013) Traumatic brain injury-induced alterations in peripheral immunity. J. Trauma Acute Care Surg. 75, 780–788. 52. Ipaktchi, K., Mattar, A., Niederbichler, A. D., Kim, J., Hoesel, L. M., Hemmila, M. R., Su, G. L., Remick, D. G., Wang, S. C., Arbabi, S. (2007)

8

Journal of Leukocyte Biology

Volume 98, December 2015

53. 54. 55. 56.

57.

58.

59.

60.

61. 62. 63. 64.

Attenuating burn wound inflammation improves pulmonary function and survival in a burn-pneumonia model. Crit. Care Med. 35, 2139–2144. Muenzer, J. T., Davis, C. G., Dunne, B. S., Unsinger, J., Dunne, W. M., Hotchkiss, R. S. (2006) Pneumonia after cecal ligation and puncture: a clinically relevant “two-hit” model of sepsis. Shock 26, 565–570. Remick, D. G., Garg, S. J., Newcomb, D. E., Wollenberg, G., Huie, T. K., Bolgos, G. L. (1998) Exogenous interleukin-10 fails to decrease the mortality or morbidity of sepsis. Crit. Care Med. 26, 895–904. Hotchkiss, R. S., Monneret, G., Payen, D. (2013) Immunosuppression in sepsis: a novel understanding of the disorder and a new therapeutic approach. Lancet Infect. Dis. 13, 260–268. Brakenridge, S. C., Phelan, H. A., Henley, S. S., Golden, R. M., Kashner, T. M., Eastman, A. E., Sperry, J. L., Harbrecht, B. G., Moore, E. E., Cuschieri, J., Maier, R. V., Minei, J. P.; Inflammation and the Host Response to Injury Investigators. (2011) Early blood product and crystalloid volume resuscitation: risk association with multiple organ dysfunction after severe blunt traumatic injury. J. Trauma 71, 299–305. Nacionales, D. C., Cuenca, A. G., Ungaro, R., Gentile, L. F., Joiner, D., Satoh, M., Lomas-Neira, J., Ayala, A., Bihorac, A., Delano, M. J., Ang, D. N., Efron, P. A. (2012) The acute immunological response to blood transfusion is influenced by polymicrobial sepsis. Shock 38, 598–606. Bible, L. E., Pasupuleti, L. V., Gore, A. V., Sifri, Z. C., Kannan, K. B., Mohr, A. M. (2015) Chronic restraint stress after injury and shock is associated with persistent anemia despite prolonged elevation in erythropoietin levels. J. Trauma Acute Care Surg. In press. Wafaisade, A., Lefering, R., Bouillon, B., Sakka, S. G., Thamm, O. C., Paffrath, T., Neugebauer, E., Maegele, M.; Trauma Registry of the German Society for Trauma Surgery. (2011) Epidemiology and risk factors of sepsis after multiple trauma: an analysis of 29,829 patients from the Trauma Registry of the German Society for Trauma Surgery. Crit. Care Med. 39, 621–628. Amara, U., Kalbitz, M., Perl, M., Flierl, M. A., Rittirsch, D., Weiss, M., Schneider, M., Gebhard, F., Huber-Lang, M. (2010) Early expression changes of complement regulatory proteins and C5A receptor (CD88) on leukocytes after multiple injury in humans. Shock 33, 568–575. Warren, H. S., Tompkins, R. G., Moldawer, L. L., Seok, J., Xu, W., Mindrinos, M. N., Maier, R. V., Xiao, W., Davis, R. W. (2015) Mice are not men. Proc. Natl. Acad. Sci. USA 112, E345. Tompkins, R. G., Warren, H. S., Mindrinos, M. N., Xiao, W., Davis, R. W. (2013) Reply to Osterburg et al.: to study human inflammatory diseases in humans. Proc. Natl. Acad. Sci. USA 110, E3371. Abdullahi, A., Amini-Nik, S., Jeschke, M. G. (2014) Animal models in burn research. Cell. Mol. Life Sci. 71, 3241–3255. Delano, M. J., Scumpia, P. O., Weinstein, J. S., Coco, D., Nagaraj, S., Kelly-Scumpia, K. M., O’Malley, K. A., Wynn, J. L., Antonenko, S., AlQuran, S. Z., Swan, R., Chung, C. S., Atkinson, M. A., Ramphal, R., Gabrilovich, D. I., Reeves, W. H., Ayala, A., Phillips, J., Laface, D., Heyworth, P. G., Clare-Salzler, M., Moldawer, L. L. (2007) MyD88dependent expansion of an immature GR-1(+)CD11b(+) population induces T cell suppression and Th2 polarization in sepsis. J. Exp. Med. 204, 1463–1474.

KEY WORDS: translation immunity •

www.jleukbio.org