The germfree animal is free of all microorganisms, whereas the disease-free animal may be described as one free of pathogens and the clinical signs that they produce.
Henry L Foster (1) p80
In this perspective, we argue that a great deal more attention should be given to selecting the appropriate animal model and gut pathogen status if the research objective is to translate a drug or technology into military or civilian trauma (2, 3). A model fit for this purpose must carefully consider the balance between precision (e.g., reproducibility) and accuracy (e.g., relevant to the human condition). Notwithstanding the heterogeneity and complexity of human disease and trial design (4, 5), the choice of animal model and its microbiome status may be two factors that influence the success or failure of translation into humans. Currently the failure rate of drug translation is over 90% (6), and the graveyard of clinical trials costs pharmaceutical companies and funding bodies billions of dollars every year (7).
Mouse models have made an enormous contribution to our understanding of the molecular mechanisms underpinning human health and disease (8, 9). However, it is not widely appreciated that the mouse differs from rats, guinea-pigs, rabbits, pigs, and humans in having the ability, under stressful conditions, to enter torpor (2, 10). Torpor is the ability of an animal to “fall off” the standard “mouse-to-elephant curve” and lower their core temperature and metabolic rate to extreme values during times of stress, and return when conditions are more favorable (11). Unlike hibernation, torpor is a more involuntary state that can be triggered at any time, and usually lasts for shorter periods (12). In addition, Bouma et al. (13) showed that torpor itself can profoundly change the animal's immune system by reducing the numbers of circulating leukocytes, lowering complement levels, and change the animal's response to pathogen challenges. This ability of the mouse to enter torpor may be a confounding variable in trauma studies such as hemorrhagic shock, septic shock, spinal cord injury, or cardiac arrest. One example of failed mouse-to-pig translation is the story of hydrogen sulphide for extremis. The initial studies showed that hydrogen sulfide could place a mouse in suspended animation for many hours with apparent complete recovery of function (14). This ability of hydrogen sulfide to induce reversible whole-body arrest received huge interest from the global media and US military. Unfortunately, despite an appropriate experimental design, the mouse study failed to translate to pigs with respect to extremis because the pig does not possess a natural ability to enter states of torpor. The key point here is that small animal models for a particular research objective are not all equal (2).
The topic of small animal models for human translation was also addressed by a 2019 working group convened by the National Institute of General Medical Sciences (NIGMS), Bethesda, Md (15). The group was tasked to evaluate animal models for sepsis research and the best way forward. About 80% of NIGMS-funded sepsis projects (2014–2018) use the mouse model. Strengths primarily centered around the cecal ligation and puncture model in capturing characteristics of short-term infection, whereas the drawbacks were a lack of translation of findings to humans. The field they argued “was hungry to develop, and receive endorsement for, standardized model systems” that mimic the inflammatory response of clinical sepsis (15). Surprisingly, in their 30-page final report, the NIGMS committee not once mentioned the ability of the mouse to depress its metabolism or the gut microbiome as potential confounders of variability and contributors to lack of translatability. In our view, this requires urgent reappraisal, including a re-examination of the 2013 benchmark study of Seok et al. (16) who used a highly inbred C57BL/6 strain to compare the transcriptional responses in peripheral blood to inflammatory injuries to endotoxin and trauma. This study showed little correlation to the human response and was highly controversial (17) (see later).
SPECIFIC PATHOGEN-FREE ANIMALS MAY NOT MIMIC THE CASUALTY ON THE BATTLEFIELD
An additional concern in translation from animals to humans is the right choice of animal husbandry practices, specifically the choice between specific pathogen-free (SPF) variants and conventionally bred animals. SPF manipulation of the microbiome–host relationship is increasingly being recognized as a confounding variable in biomedical research and human translation (3, 9, 18). A wounded soldier on the battlefield, a person undergoing major surgery, or the critically ill, do not have a gut microbiome similar to the highly artificial animals born and raised in a sterile “bubble” or modified thereafter.
SPF animals were introduced into biomedical research in the early 1960s to minimize disease or infection as an unwanted variable in experimental design (1). In 1962, Lane-Petter (19) summarized the continual frustration of the times when he wrote: “This has led, for many years, to a demand for “better” animals; animals that are “disease-free,” ’healthy,” or otherwise free of the customary drawbacks”. Three different methods of breeding were examined: Germ-free animals or “life in a bubble”; SPF animals, which originally were born from germ-free stock or cesarean aseptic techniques, then the colony was exposed to an environment free of infectious organisms (not all) that may otherwise interfere with an investigator's research objectives, and conventional healthy animals (sometimes called “dirty” animals), which were bred in open cages in a controlled, health-monitored and more natural environment of antigenic exposures, and indigenous gut flora (19, 20) (Table 1).
Today, another confounding variable is the wide SPF heterogeneity among the major vendor breeding facilities and institutions. For example, Charles River and Taconic Biosciences do not exclude opportunistic organisms beta-hemolytic Streptococcus, Staphylococcus aureus, Pseudomonas aeruginosa, Klebsiella spp., and Helicobacter spp. (21), whereas they are excluded at Harvard Medical School, Cornell University, and our own Australian Institute of Tropical Health and Medicine, James Cook University, Australia (22). Table 2 summarizes the exclusion criteria used by some of the major vendors and universities.
WHAT YOU START WITH MATTERS
The animal's indigenous/baseline gut microbiota matters because the experiment itself (handling, anesthesia, surgery, treatment, antibiotics) elicits a stress response that can dramatically alter the type and abundance of intestinal flora. Typically, there are around 100 trillion microbes (bacteria, archaea, viruses, and fungi) in the mammalian gut that have coevolved with the host's genome for millions of years (23–25). Our survival is therefore the product of two genomes that coordinate the breakdown and assimilation of foodstuffs, immune system programming, inflammatory control, and the maintenance of health and wellbeing (23–25). Immune modulation of the gut-brain axis has received much attention recently as it influences chronic disease susceptibility, circadian rhythm, psychological behavior, and stress tolerances (26–31). Moreover, it is increasingly being recognized that in the acute experimental or trauma setting, the gut flora can transform a shift toward a “pathobiome” and poor outcomes (32). This is particularly apparent when the gut wall is breached following ischemia, infection, or mechanical injury (23, 24, 33, 34). In a recent study of severe trauma patients, Howard et al. (35) showed the microbiome can change its composition and relative abundance of species in the first 72 h after injury. Nicholson et al. (36) also showed in traumatic brain injury patients that significant changes in the microbiome composition occur as early as 2 h after injury, and more recently after administering transfusion therapy in trauma patients (37). Thus, it appears trauma can lead to a rapid depletion of health-promoting microbes and contribute to secondary injury progression associated with systemic inflammation, coagulopathy, immune deficiencies, and infection (25, 38). Selecting the right animal model and gut pathogen status that best suits your research objectives is critically important in trauma and shock research.
WE WERE WARNED
Concerns with SPF versus conventionally bred animals has a long history. In the early- to mid-1960s, microbiologist Dubos et al. (39–41) studied germ-free, SPF, and “normal” wild adult mice from the same genetic origin, and were among the first to show that changes in microbiota were associated with differences in growth rate, efficiency in utilizing food, social interactions, maternal care, resistance to infection and toxins, immune function, and stress (40). Moreover, when germ-free or SPF mice were housed with normal mice these changes reverted back to their “normal” states. These early observations appear to have been lost over the years, and we believe every young investigator or university course in animal research and ethics should revisit these studies before they undertake biomedical research.
We had a similar experience in 2018 when we switched from conventional to SPF rats for our military-funded research at James Cook University, and found that SPF animals displayed an abnormal hemodynamic, hematological, and bleeding phenotype in response to anesthesia and minor surgery (18). SPF rats had significantly lower pulse pressure (38% decrease), arrhythmias and prolonged QTc (27% increase) compared with conventional rats. No arrhythmias were found in conventional rats. SPF rats had significantly higher white cell, monocyte, neutrophil and lymphocyte counts, and were hyperfibrinolytic, indicated by EXTEM maximum lysis > 15% (18). When we reverted back to the conventional animal husbandry methods, basic physiology and the stress response to surgical and hemorrhagic trauma returned to established norms (18).
Similarly, a landmark study of Beura et al. (28) demonstrated that “standard” SPF adult mice have “immature” immune systems and were more prone to infection than conventionally-bred mice. They found their laboratory mice lacked effector-differentiated and mucosally distributed memory T cells, which was different from free-living barn populations of feral mice and pet store mice. Moreover, cohousing of SPF mice with pet store mice reversed the problem, and produced mice with immune systems closer to adult humans (28). Similarly, Rosshart et al. (42) showed SPF-type mice reconstituted with natural microbiota exhibited reduced inflammation and increased survival following influenza virus infection, and displayed improved resistance against colorectal tumorigenesis, while others have exploited the SPF idea and developed experimental inflammatory disease in mouse models (43). More recently, Rosshart et al. (44) showed that laboratory born to wild mice (called wildings) have natural microbiota more closely mimics human immune responses with enhanced validity and reproducibility. These new findings may help to explain why the widely cited study of Seok et al. (16) found little correlation between mouse and human inflammatory responses to infection. This is important because animal rights organizations continue to use Seok's study to buttress their arguments on the irrelevance of mouse models to human health. Again, it was disappointing the final report from the 2019 NIGMS working group failed to consider the gut-brain axis in their discussions on the applicability of the mouse model to study sepsis (see above).
THE PATH FORWARD
In this perspective, we are not recommending introducing wild or pet store animals into a standard breeding colony at the risk of introducing unwanted infectious vectors. However, we are suggesting the best chance of translational success will be found by using animal models that reflect a more natural state from a conventional animal facility with routine microbiome profiling, standard health screening, and ethical practices (3). Unfortunately, we don’t think it is plausible to fully “humanize” an animal model given the staggering microbial gene diversity in humans (44, 45). A path forward may be the introduction of an animal microbial/pathogen exclusion status statement at the end of each scientific publication (3). For example, it could read: A list of pathogens excluded in animals supporting the conclusions of this study is available by contacting the author(s) and/or institutional data hub (with an appropriate URL). This information is already available at most, if not all, research institutions and universities. The strategy may also help to address the subtle differences in experimental results within a single laboratory or between laboratories around the world.
1. Foster HL. Housing of disease-free vertebrates. Ann N Y Acad Sci
2. Dobson GP. The August Krogh Principle: Seeking Unity in Diversity. Shock
42 (5):480, 2014.
3. Dobson GP, Letson HL, Biros E, Morris JL. Specific pathogen-free
(SPF) animal status as a variable in biomedical research: Have we come full circle? EBioMedicine
4. Van Norman GA. Phase II trials in drug development and adaptive trial design. JACC Basic Transl Sci
4 (3):428–437, 2019.
5. Downing NS, Shah ND, Aminawung JA, Pease AM, Zeitoun JD, Krumholz HM, Ross JS. Postmarket safety events among novel therapeutics approved by the US Food and Drug Administration Between 2001 and 2010. JAMA
317 (18):1854–1863, 2017.
6. Hay M, Thomas DW, Craighead JL, Economides C, Rosenthal J. Clinical development success rates for investigational drugs. Nat Biotechnol
32 (1):40–51, 2014.
7. Moore TJ, Zhang H, Anderson G, Alexander GC. Estimated costs of pivotal trials for novel therapeutic agents approved by the US Food and Drug Administration, 2015-2016. JAMA Intern Med
178 (11):1451–1457, 2018.
8. Osuchowski MF, Remick DG, Lederer JA, Lang CH, Aasen AO, Aibiki M, Azevedo LC, Bahrami S, Boros M, Cooney R, et al. Abandon the Mouse Research Ship? Not just Yet!. Shock
41 (6):463–475, 2014.
9. Masopust D, Sivula CP, Jameson SC. Of mice, dirty mice and men: using mice to understand human immunology. J Immunol
10. Schubert KA, Boerema AS, Vaanholt LM, de Boer SF, Strijkstra AM, Daan S. Daily torpor in mice: high foraging costs trigger energy-saving hypothermia. Biol Lett
6 (1):132–135, 2010.
11. Dobson GP. Organ arrest, protection and preservation: natural hibernation to cardiac surgery: a review. Comp Biochem Physiol Part B Biochem Mol Biol
12. Ruf T, Geiser F. Daily torpor and hibernation in birds and mammals. Biol Rev
90 (3):891–926, 2015.
13. Bouma HR, Carey HV, Kroese GM. Hibernation: the immune system at rest? J Leukoc Biol
88 (4):619–624, 2010.
14. Blackstone E, Roth MB. Suspended animation-like state protects mice from lethal hypoxia. Shock
27 (4):370–372, 2007.
15. Younger JG and Kraft MC (Co-Chairs). NAGMSC Working Group on Sepsis: Final Report. National Institutes of Health, 2019. Available at: https://www.nigms.nih.gov/News/reports/Documents/nagmsc-working-group-on-sepsis-final-report.pdf
. Accessed November 7, 2019.
16. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, Xu W, Richards DR, McDonald-Smith GP, Gao H, Hennessy L, et al. Genomic responses in mouse models poorly mimic human inflammatory diseases. Proc Natl Acad Sci USA
110 (9):3507–3512, 2013.
17. Drake AC. Of mice and men; what rodent
models don’t tell us. Cell Mol Immun
18. Letson HL, Morris JL, Biros E, Dobson GP. Conventional and specific-pathogen free rats respond differently to anesthesia and surgical trauma
. Sci Rep
9 (1):9399, 2019.
19. Lane-Petter W. Provision of pathogen-free animals. Proc R Soc Med
20. Franklin CL, Ericsson AC. Microbiota and reproducibility of rodent
models. Lab Anim (NY)
46 (4):114–122, 2017.
22. Harvard Medical School. List of Excluded Rodent
Disease Agents. Center for Animal Resources and Comparative Medicine, 2019. Available at: http://www.kudosconcepts.com/samples/arcm/documents/html_pages/excluded_rodent_disease_agents.htm
). (Accessed September 23, 2019).
23. Ley RE, Peterson DA, Gordon JI. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell
24. Li D, Chen H, Mao B, Yang Q, Zhao J, Gu Z, Zhang H, Chen YQ, Chen W. Microbial biogeography and core microbiota of the rat digestive tract. Sci Rep
25. Otani S, Chihade DB, Coopersmith CM. Critical illness and the role of the microbiome
. Acute Med Surg
6 (2):91–94, 2019.
26. Sudo N. M. Lite JF, Cryan M. Microbiome
, HPA axis and production of endocrine hormones in the gut. Springer, Microbiology Endocrinology: The Microbiota-Gut Brain Axis in Health and Disease. 817
. New York:2014.
27. Mayer EAK, Gupta T A. Gut/brain axis and the microbiota. J Clin Invest
125 (3):926–938, 2015.
28. Beura LK, Hamilton SE, Bi K, Schenkel JM, Odumade OA, Casey KA, Thompson EA, Fraser KA, Rosato PC, Filali-Mouhim A, et al. Normalizing the environment recapitulates adult human immune traits in laboratory mice. Nature
29. Schott EM, Farnsworth CW, Grier A, Lillis JA, Soniwala S, Dadourian GH, Bell RD, Doolittle ML, Villani DA, Awad H, et al. Targeting the gut microbiome
to treat the osteoarthritis of obesity. JCI Insight
3 (8):pii:95997, 2018.
30. Masri S, Sassone-Corsi P. The emerging link between cancer, metabolism, and circadian rhythms. Nat Med
24 (12):1795–1803, 2018.
31. Morris JL, Letson HL, Gillman R, Hazratwala K, Wilkinson M, McEwen P, Dobson GP. The CNS theory of osteoarthritis: opportunities beyond the joint. Semin Arthritis Rheum
piiS0049-172 (19):30012–30015, 2019.
32. Akrami K, Sweeney DA. The microbiome
of the critically ill patient. Curr Opin Crit Care
24 (1):49–54, 2018.
33. Clark JA, Coopersmith CM. Intestinal crosstalk: a new paradigm for understanding the gut as the “motor” of critical illness. Shock
28 (4):384–393, 2007.
34. Dobson GP. Addressing the global burden of trauma
in major surgery. Front Surg
35. Howard BM, Kornblith LZ, Christie SA, Conroy AS, Nelson MF, Campion EM, Callcut RA, Calfee CS, Lamere BJ, Fadrosh DW, et al. Characterizing the gut microbiome
: significant changes in microbial diversity occur early after severe injury. Trauma Surg Acute Care Open
2 (1):e000108, 2017.
36. Nicholson SE, Watts LT, Burmeister DM, Merrill D, Scroggins S, Zou Y, Lai Z, Grandhi R, Lewis AM, Newton LM, et al. Moderate traumatic brain injury alters the gastrointestinal microbiome
in a time-dependent manner. Shock
52 (2):240–248, 2019.
37. Nicholson SE, Burmeister DM, Johnson TR, Zou Y, Lai Z, Scroggins S, DeRosa M, Jonas RB, Merrill D, Zhu C, et al. A prospective study in severely injured patients reveals an altered gut microbiome
is associated with transfusion volume. J Trauma Acute Care Surg
86 (4):573–582, 2019.
38. Zhu CS, Grandhi R, Patterson TT, Nicholson SE. A review of traumatic brain injury and the gut microbiome
: insights into novel mechanisms of secondary brain injury and promising targets for neuroprotection. Brain Sci
8 (6):E113, 2018.
39. Dubos RJ, Shaedler RW. The effect of the intestinal flora on the growth rate of mice, and on their susceptibility to experimental infections. J Exp Med
40. Dubos R, Savage D, Schaedler R. Biological Freudianism. Lasting effects of early environmental influences. Pediatrics
38 (5):789–800, 1966.
41. Tannock GW. Commentary: remembrance of microbes past. Int J Epidemiol
34 (1):13–15, 2005.
42. Rosshart SP, Vassallo BG, Angeletti D, Hutchinson DS, Morgan AP, Takeda K, Hickman HD, McCulloch JA, Badger JH, Ajami NJ, et al. Wild mouse gut microbiota promotes host fitness and improves disease resistance. Cell
171 (5):1015–1028, 2017.
43. Hufeldt MR, Nielsen DS, Vogensen FK, Midtvedt T, Hansen AK. Variation in the gut microbiota of laboratory mice is related to both genetic and environmental factors. Comp Med
44. Rosshart SP, Herz J, Vassallo BG, Hunter A, Wall MK, Badger JH, McCulloch JA, Anastasakis DG, Sarshad AA, Leonardi I, et al. Laboratory mice born to wild mice have natural microbiota and model human immune responses. Science
2019; 365 (6452): pii: eaaw4361.
45. Tierney BT, Yang Z, Luber JM, Beaudin M, Wibowo MC, Baek C, Mehlenbacher E, Patel CJ, Kostic AD. The landscape of genetic content in the gut and oral human microbiome
. Cell Host Microbe
26 (2):283–95e8, 2019.