Extensive burn injury results in significant pathophysiologic stress accompanied by marked systemic inflammation that may lead to multi-organ failure (1, 2). As such, burns are associated with significant morbidity and mortality leading to approximately 11 million hospitalizations and 300,000 deaths worldwide annually (3). The management of burn patients has evolved to include aggressive fluid resuscitation, early debridement/grafting, and enteral feeding that have collectively improved outcomes (4). Still, the burn patient is highly susceptible to infectious complications such as sepsis and pneumonia, which have deleterious consequences on outcomes and costs. As a correlative, it is thought that sepsis development following a burn injury may be due to increases in intestinal permeability leading to microbe leakage (5–8). Despite this, there remains a paucity of evidence on how to manage compromised gut integrity following severe burns.
Recent advances in sequencing technology have been leveraged to elucidate the complex ecosystem termed the microbiome that lives in and on our bodies. While several studies have examined burn wound outcomes and the skin microbiome (9, 10), relatively few studies have investigated the gut microbiome (GM) following large burn injury. GM shifts are described by observing changes in specific taxonomic levels and by using α-diversity (a measure of species distribution within a sample) and β-diversity (a comparison of microbial composition between groups) matrices. Thus far, one rodent study has shown burn-induced shifts in the GM may allow for proliferation of pathogenic organisms (11). The implications of these changes are still murky but it is thought a pathobiome may contribute to later sepsis development or other forms of multi-organ dysfunction.
While massive vascular leakage and fluid shifts post-burn have been countered with aggressive intravenous (IV) fluid resuscitation, there remains variability in resuscitation strategies and volumes within the burn community (12). Moreover, large amounts of IV fluids can lead to clinical sequelae such as compartment syndromes or acute respiratory distress syndrome. The extent to which fluid levels effect the GM is largely unstudied. To this end, the current study investigates the effect of different resuscitation volumes on the GM and proteins critical for intestinal absorption utilizing an established porcine 40% TBSA burn model (13). We hypothesized that a shift in the GM following burn injury would be modulated by varying fluid resuscitation volumes.
MATERIALS AND METHODS
Research was conducted in compliance with the Animal Welfare Act, the implementing Animal Welfare Regulations, and the principles of the Guide for the Care and Use of Laboratory Animals, National Research Council. The facility's Institutional Animal Care and Use Committee approved all research conducted in this study. The facility where this research was conducted is fully accredited by AAALAC. Seventeen 3-month-old sexually immature female Yorkshire (Sus Scrofa) pigs (weighing 40.9 ± 1.1 kg) were included in this study, which represented a substudy of previously published work (14). Upon arrival to our institute, animals had a minimum 7-day acclimation period, during which they were singly housed, with ad libitum access to water, and fed a consistent commercial laboratory porcine formulated pelleted diet (Laboratory Mini-Pig Grower Diet, Cat# 5081, LabDiet, Richmond, Ind). Animals were randomly allocated to one of three treatments prior to thermal injury: restricted fluid group (None; n = 5), or IV lactated Ringer's solution at 15 mL/kg/d (Low; n = 6), or IV 80 mL/kg/d (High; n = 6).
Thermal injury and follow-up
All animals were given a one-time intramuscular injection of 0.1 ± 0.24 mg/kg Buprenex-HCl Sustained Release (Veterinary Technologies/ZooPharm, Windsor, Colo), which provides analgesia for up to 72 h, immediately prior to the creation of the burn wounds. Creation of 40% TBSA burn full-thickness contact burns was performed as previously described (13). Briefly, animals were anesthetized with an intramuscular injection of tiletamine-zolazepam (Telazol, 6 mg/kg), intubated, and placed on a ventilator with an initial tidal volume at 10 mL/kg, a peak inspiratory pressure of 20 cm H2O, and respiratory rate of 8 to 10 breaths/min. The ventilator was adjusted to achieve an end-tidal PCO2 of 40 ± 5 mm Hg. Animals were maintained on 1% to 3% isoflurane, balance O2 anesthesia. Hair was removed from the dorsum, flanks, and legs using clippers and razors with shaving cream. Intravenous access was created using standard cut-down procedures used to place left and right external jugular vein catheters that were anchored in place and tunneled subcutaneously to the back of the neck. Large (9 × 15 cm) and small (5 × 5 cm) custom-designed brass blocks equipped with a thermocouple were maintained at 100 ± 0.2°C by a temperature controller. Heated probes were placed against the skin for 30 s to produce full-thickness burn injuries. This procedure was repeated until 40% of the TBSA was burned.
Burn wounds and postinjury follow-up
Wounds were covered with Ioban antimicrobial dressings (3 M, St. Paul, Minn) for the duration of the experiment, which were replaced if wounds were exposed. Animals recovered from anesthesia and were kept in a metabolic cage (15) (dimensions 41’L × 16’ W 44’ H) for monitoring IV fluid intake as previously described. Feed was given once animals were awake and standing independently. Because of the unfamiliarity with the smaller metabolic cage, five animals showed signs of distress (e.g., vocalization, jumping) and were administered intramuscular midazolam (0.1 ± 0.25 mg/kg) for light sedation. Animals were fed the same formulated chow post-burn, and there was no apparent difference in the slightly reduced appetites across groups. Approximately 24 and 48 h following burn injury, animals were sedated with Telazol (6 mg/kg) to collect blood samples and monitor physiological parameters (heart rate, respiratory rate, and rectal temperature). At this time rectal swabs were obtained for sequencing analysis and placed at −80°C until all samples were collected. Upon euthanasia at 48 h, intestine samples from the proximal jejunum were immediately either flash-frozen, or preserved in 10% neutral buffered formalin for a minimum of 48 h, embedded in paraffin wax, and sectioned into 4-μm slices.
At baseline and 48 h, superior mesenteric artery cross-sectional area was visualized with contrast-enhanced computerized tomography (CT) scans. Under anesthesia, 40 mL of Iopamidol (755 mg/mL) was injected into an ear vein catheter. Obtained CT images were transferred to an independent Vitrea 3D workstation (Vitrea Version 6.7.4; Vital Image Inc, Minnetonka, Minn) for measurement. The Vitrea Advanced 3D Vascular: Runoff CT protocol was used for the analysis of the superior mesenteric artery, and the Vessel Probing tool was utilized to probe the superior mesenteric artery 0.5 mm to 1.0 mm in the same anatomic plane as the first junction of the celiac artery.
Proximal jejunum samples embedded in paraffin wax were deparaffinized and stained with wheat germ agglutinin and counterstained with phalloidin and DAPI. Additionally, TUNEL staining was performed (C10617, ThermoFisher Scientific, Waltham, Mass) according to the manufacturer's instructions. Intestine slices were imaged using a Zeiss Observer (Carl Zeiss, Thornwood, NY) and put through automated quantification of colors with ImageJ software version 1.51d (Bethesda, Md). Ensuing images were separated into red, green, and blue channels for quantification of channel intensities.
To quantify total Caspase, SGLT1, AQP1, HSP70, and beta-actin within the intestines Western blots were performed by separating 50 μg on 20% SDS-PAGE gels, and a semidry transfer to nitrocellulose membranes. Secondary antibodies (1:20,000) IRDye 800CW or IRDye 680RD (LI-COR) were applied for compatibility with the Odyssey infrared imaging system (LI-COR) and signal intensity was normalized to beta actin. For detection of tissue cytokines, intestines were lysed in Milliplex lysis buffer (EMD Millipore, Billerica, Mass) and a porcine-specific multiplex kit (Millipore; PCYTMG-23K13PX) was used according to the manufacturer's instructions.
Swabbing was performed using sterile cotton tipped applicators (ThermoFisher Scientific) and stored at −80°C until analysis. Fecal DNA was isolated with the QIAamp DNA Stool Mini Kit (Qiagen, Venlo, The Netherlands) according to the manufacturer's instructions. All samples (except 2 with insufficient DNA yield in the none group) underwent 16S rRNA amplification, library creation, and sequencing using V1_V2 16S Illumina MiSeq 600 pipeline and the 27F/R338 primer pair (Forward: NAGAGTTTGATCMTGGCTCAG; Reverse: NGCTGCCTCCCGTAGGAGT) (16). Sequences were deposited in the NCBI Sequence Read Archive under the bioproject ID#PRJNA573749 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA573749).
Sequencing analysis was performed using QIIME2-2017.10 given its established high performance in microbiome analysis (17). 5,043,315 sequences across all samples were identified after merging and demultiplexing. Sequence quality control, feature table construction, and chimera removal was performed using the DADA2 pipeline (18). A right truncation was performed at 200 bp based on a median quality score value that was > 30, and aligned representative sequences were subsequently determined using FastTree 2 (19). Sample depth for diversity measures was set at 42,100 bp to retain 60.51% of reads and 100% of samples. Alpha (Shannon diversity, Observed Taxonomic Units (OTUs), Faith Diversity, and evenness) and beta (Jaccard, Bray-Curtis, weighted/ unweighted UniFrac) diversity measures were calculated, and visualized using the Emperor program. OTU classification was accomplished using the Greengenes 13_8 database. Taxa were then assigned using the trained classifier and differential abundance was determined.
Two-way ANOVA analysis with Tukey post-testing was performed at each taxonomic level to determine differences in diversity between groups. Western blot and cytokine data were analyzed using a 1-way analysis of variance method (ANOVA) assuming Gaussian distribution. To detect treatment differences Tukey's multiple comparison test was performed. PERMANOVA in QIIME2 was used for beta diversity measure analysis. Unless otherwise stated, values are represented as arithmetic mean ± SEM. Statistical analysis and figure generation was performed using Graphpad Prism 7.0e (Graphpad Software, San Diego, Calif).
Sequencing using the Illumina pipeline identified 5,043,315 total sequences (average of 114,620/sample). Following quality control, and removal of chimeras, 17,052 representative sequences were used for phylogenic tree generation. Eighteen unique phyla were identified with Bacteroidetes (32.3%), Firmicutes (30.9%), Proteobacteria (15.2%), and Spirochaetes (3.2%) being the most prevalent. Forty unique genera were also found, with Prevotella (18.9%), Sulfurimonas (14.3%), Treponema (14.0%), and Ruminococcus (12.9%) representing the four most commonly observed. Phyla and genera distributions for individual animals are shown in Supplemental Figure 1, Supplemental Digital Content 1, http://links.lww.com/SHK/A962. While baseline variability existed, no differences in β-diversity measures unweighted and weighted UniFrac plots (P = 0.781, P = 0.379, respectively), or α-diversity measures Shannon diversity (P = 0.86), evenness (P = 0.80) faith phylogenic diversity (P = 0.44), or observed OTUs (P
= 0.77) were seen between study groups at baseline (Supplemental Figure 2, Supplemental Digital Content 2, http://links.lww.com/SHK/A963).
Burn injury induced significant differences in several phyla which were differentially affected by IV fluids. Proteobacteria were markedly elevated at day 1, but most dramatically in the none group (15.2% vs. 39.5%; P < 0.0001) compared to baseline (Fig. 1A). By day 2, Proteobacteria levels remained elevated in the low group (P = 0.0062). Significant decreases in Firmicutes at day 1 in all groups were also most dramatic in the none group (31.4% vs. 18.7%; P = 0.0197) and became non-significant by day 2 in all groups (Fig. 1, B and C). Non-significant decreases in Bacteroidetes levels were observed on day 1 for the low and none groups that returned to baseline by day 2 (Fig. 1D).
At all lower taxonomic levels, differences based on treatment group and study day were found (Supplemental figure 3, Supplemental Digital Content 3, http://links.lww.com/SHK/A964). Notably, the family Enterobacteriaceae was increased across all time points and treatment groups. At the genus level, a dose- and time-dependent increase in the Bacteroides genus was associated with increased fluid levels, with concurrent decreases in Prevotella. At the species level, increases in Haemophilus parainfluenzae and decreases in Prevotella copri were transiently exacerbated by fluid restriction (Supplemental Figure 3, Supplemental Digital Content 3, http://links.lww.com/SHK/A964).
For α-diversity, a drop in observed taxonomic units (OTU) became significant by day 2 for animals that received fluids (P < 0.022) (Fig. 2A). Similar decreases in Shannon (P = 0.19), faith (P = 0.31), and evenness (P = 0.53) were not statistically significant (Fig. 2, B and D). Interestingly, day 2 Firmicutes (Fig. 2E) levels positively correlated with, OTU (P = 0.0009; r2 = 0.58), Shannon diversity (P < 0.0001; r2 = 0.78), and Evenness (P < 0.0001; r2 = 0.73), while day 2 Proteobacteria (Fig. 2F) inversely correlated with these (P = 0.014; r2 = 0.38, P = 0.0045; r2 = 0.48, P = 0.0071; r2 = 0.44).
Measures of β-diversity revealed a significant shift in the microbiome following burns regardless of IV fluids (Fig. 3, A–D) for all 4 β-diversity measures studied (Jaccard similarity, P = 0.001, Bray–Curtis, P = 0.002, unweighted UniFrac distance, P = 0.023, and weighted UniFrac distance, P = 0.012). Largely, post-hoc analysis showed differences between baseline and ensuing days, but no further shifts between days 1 and 2. To examine the potential confounding effect of midazolam administration, we compared the day 2 GM β-diversity between animals that received midazolam and those that did not, and found no significant effect of benzodiazepine administration on the GM in terms of either α- or β-diversity (Supplemental Figure 4, Supplemental Digital Content 4, http://links.lww.com/SHK/A965).
High levels of IV fluids led to a non-significant increase in goblet cell area and decrease in wet-to-dry ratio (Supplemental Figure 5A–E, Supplemental Digital Content 5, http://links.lww.com/SHK/A966). As a proxy for gut perfusion, CT analysis of the superior mesenteric artery cross sectional area (Fig. 4, A–C) revealed that burn injury reduced the size of the mesenteric artery (P < 0.0001), which was exacerbated in the none group, but not different between the low and high groups.
Labeling of apoptotic cells with TUNEL staining showed an increase in apoptotic cells within the epithelium and lamina propria of animals that did not receive IV fluids (Fig. 5A). Moreover, quantitative Westerns revealed high levels of fluids reduced the amount of Caspase-3 (P = 0.0139). Western blotting also revealed that high fluid levels were associated with decreased active transporter SGLT1 (P = 0.0213) and non-significant decreases in the passive transporter AQP1 (P = 0.0976). Lastly, IV fluids also increased HSP70 (P = 0.0464) levels in a dose-dependent manner (Fig. 6).
Cytokine analysis showed that low amounts of fluids significantly elevated IL-1α (P = 0.03), and IL-12 (P
= 0.04) levels (Supplemental Table 1, Supplemental Digital Content 6, http://links.lww.com/SHK/A967) within the intestine. Interestingly, SGLT levels were linearly correlated with increased tissue levels of IL-8 (P
= 0.0441; r2 = 0.2964) and IL-10 (P = 0.0490; r2 = 0.2856) (Supplemental Figure 6A, Supplemental Digital Content 7, http://links.lww.com/SHK/A968). Further analysis revealed that Bacteroides levels were positively correlated with tissue IL-18 (P = 0.0166; r2 = 0.4201) (Supplemental Figure 6B, Supplemental Digital Content 7, http://links.lww.com/SHK/A968). Inverse correlations between day 2 Faith diversity and anti-inflammatory cytokines IL-2 (P = 0.034; r2 = 0.35), IL-4 (P = 0.039; r2 = 0.33), and IL-10 (P = 0.039; r2 = 0.33) within the intestine (Supplemental Figure 6C-E, Supplemental Digital Content 7, http://links.lww.com/SHK/A968) were also found.
Recent advances in sequencing technology have yielded a panoply of information on the microbiome in health and disease, yet the microbiome's effect on burn outcomes remains unclear. Early and aggressive fluid resuscitation is a mainstay of burn management: however, there also remains a lack of data on how this treatment strategy influences the GM. To examine these potential effects, we used 16S sequencing of the fecal microbiome and tissue protein analysis in an established porcine model of 40% TBSA thermal injury undergoing different fluid resuscitation strategies. To our knowledge, this is the first report characterizing the gut microbiome of swine following burn injury. Significant perturbations in the GM following injury were demonstrated, including a Proteobacteria-centric community with large shifts β-diversity. Fluids did mitigate burn-induced changes in certain taxa including Proteobacteria phyla and Bacteroides genus.
Following trauma, it has been well established that the GM is altered in both humans (20–23) and animal models (24); however, the implications of these changes are unclear. Burns represent a unique subset of trauma patients due to systemic inflammation, and burn-induced pathophysiology extends to the airway (25), skin (10), and, importantly, gut (11, 26) microbiomes. The acute post-burn phase, the ebb phase, is characterized by tissue hypoperfusion and ischemia which is partially allayed through aggressive fluid resuscitation. Wang et al. (27) illustrated the temporal effect of burn injury on the GM in patients, by showing a distinct microbiome population early (i.e., < 1 mo after injury (Enterococcus and Escherichia)) and late (Bacteroides) in the postinjury period. We present a hyperacute spike in potentially pathogenic organisms (Proteobacteria) immediately following injury, and that high fluid volumes can expedite a reemergence of beneficial organisms such as the Bacteroides genus.
Understanding GM diversity has the potential to identify diagnostic or therapeutic interventions. Alpha diversity measures represent species diversity within each fecal swab, with indices such as Shannon index accounting for species diversity, evenness accounting for species balance, and Faith indices considering the extent of phylogenetic diversity. One previous rodent model of 30%TBSA burn injury has shown that α-diversity is relatively unchanged in the short-term following burn injury (28). Unlike that study, we saw significant differences in OTUs, which we believe is dependent on the extent of-, and time after injury. They also mentioned qualitative perturbations to one β-diversity measure (differences in community composition amongst groups), which we confirmed with statistical PERMANOVA analyses. Rodent models have also been used to examine the effect of alcohol (29) and age (30). While these studies give valuable insight to burn-induced gut function, the advantage of using a porcine model instead of rodents lies in the similarities between pigs and humans in the cutaneous (31) and intestinal (32) microbiomes along with the structure and healing properties of the skin itself (33). To our knowledge, we believe that we are the first to describe the shift in the GM β-diversity in a porcine model following a large burn injury, which may provide the groundwork for future study into the effect of clinical interventions on the GM.
A long-standing hypothesis for sepsis involves increasing intestinal permeability following burn injury leading to the bacterial translocation (5–8), which may be related to a dysbiotic GM (11, 34). For example, while Bacteroides fragilis (a common anaerobic commensal) rarely causes bacteremia, there are instances of anaerobic sepsis (35). Although the mucus-producing capabilities of B fragilis are generally viewed as beneficial, it has been shown that toxins specific to this anaerobe can cause sepsis in both humans and animals (36, 37). Whether this is a possibility in the indistinct onset of sepsis in burn injury is not clear. However, the Bacteroides genus is generally considered to be a beneficial enteric organism due to its effects on complex sugar breakdown and maintenance of energy requirements (38). Furthermore, given this role Bacteroides plays, it is interesting high IV volumes increased this genus but decreased SGLT expression. Promoting an accelerated shift toward Bacteroides could potentially have widespread effects on burn outcomes due to the increased metabolic demand following injury.
Fluids reduced expression of both active and passive water transporters, which may represent a compensatory mechanism to counteract compromised blood flow as measured by mesenteric artery diameter. We also found decreased apoptosis with IV fluid use and increased heat shock protein in the high fluid group. While limited by a small sample size and moderate variability, these results present circumstantial evidence that instead of undergoing apoptosis, cells in the high fluid group are compensating with heat shock proteins. Prior work has also shown a relationship between apoptosis and Hsp70. Specifically, Yuan et al showed that treatment with sodium arsenite both increases Hsp70 and decreases apoptosis indicating the possibility of a shared mechanism (39). We were surprised to find a lack of correlation between these markers and microbiome characteristics.
While the ebb phase of burn injury is mediated by a milieu of pro-inflammatory cytokines in circulation, herein intestinal cytokine expression was mixed, and the only difference due to fluids was higher IL-1α and IL-12 levels in the low fluid group. An interesting study by Tadros et al. (40) found that directly administering IL-1α following a burn may be protective for mesenteric blood flow and intestinal permeability. In this context, we cannot rule out the possibility of over-resuscitation in the high group (41, 42). The observation of a correlation between IL-18 and the phylum Bacteroidetes is an interesting one, and likely tied to the role of IL-18 in gut integrity, which is implicated in burns (43), and intestinal inflammasome activation (44). While it is enticing to speculate that a specific bacteria within the Bacteroidetes phyla could be responsible for maintaining this balance, we were not able to identify a correlation at a deeper taxonomic level.
The pig GM shares many common species with humans, but does display differences. For example, while the Treponema genus identified in the pig GM (45) is not usually found in human digestive tracts, it has been shown in the guts of hunter-gatherers (46). This brings to light one key variable that dictates the flora of the gut: diet confounds interpretation of human GM studies (47), which can be controlled for in animal models. We found that burn injury induced opposing changes in the genera Bacteroides and Prevotella, which are thought of as products from diets rich in animal protein/fats and plant-based carbohydrates, respectively (48). Moreover, these two genera have distinct production of short chain fatty acids due to fiber utilization (49), and Prevotella may preferentially support weight loss (50). Resuscitation adjuncts that promote Prevotella species levels in the gut may be a worthwhile avenue of research.
The main limitation of this study lies in the lack of insight into location-specific changes in the microbiome, as small versus large intestinal variation exists (32). Given that we used fecal swabs, our results inform effects of burns on GM of the colon. While this is in contrast to the histological and molecular analyses from the jejunum (and thus may explain the lack of correlations between GM and proteins), we feel it represents a more feasible target for clinical translation. The other limitation of our study lies in the limited number of animals used in this study which hampered statistical power. Moreover, the lack of non-burn shams in our study complicates the interpretation of data where preburn measures are not available (i.e., the molecular data). However, this also highlights the potential noninvasive diagnostic power of the GM, which could be used to study the independent effect of resuscitation on gut flora. Finally, given our data showing a significant effect of resuscitation strategy on superior mesenteric artery diameter, further work is needed to determine whether resuscitation influences the GM directly, or is rather as a consequence of perfusion.
Early and aggressive fluid resuscitation is a hallmark of burn treatment. The utility of the Brooke versus the Parkland formulas is a hotly contested area. This study is the first to show that the GM is altered following a large burn injury in pigs and that the GM may be influenced by the resuscitation strategy used. We went on to show that varying fluid resuscitation strategies influence key functional proteins in the small intestine. Given that the GM is altered following a burn, further mechanistic study is needed to determine the GM effects on burn outcomes. Such future studies that manipulate the microbiome using, for example, fecal transplant or antibiotics could examine how the GM influences outcomes following burn injury in hopes of elucidating therapeutic and diagnostic targets the critical role that the GM plays in burn physiology and resuscitation outcomes. The interaction between the GM and burn outcomes remains elusive and, future preclinical and clinical trials of resuscitation strategies should include GM analysis.
The authors thank the US Army Institute of Surgical Research Veterinary Support Branch for their support.
The assertions and opinions contained herein are solely those of the authors and do not represent those of the United States Army or the Department of Defense.
1. Kraft R, Herndon DN, Finnerty CC, Shahrokhi S, Jeschke MG. Occurrence of multiorgan dysfunction in pediatric burn patients: incidence and clinical outcome. Ann Surg
259 (2):381–387, 2014.
2. Jeschke M, Chinkes D, Finnerty C. The pathophysiologic response to severe burn injury. Ann Surg
248 (3):387–401, 2008.
3. Peck MD. Epidemiology of burns throughout the world. Part I: distribution and risk factors. Burns
37 (7):1087–1100, 2011.
4. Williams FN, Herndon DN, Jeschke MG. The hypermetabolic response to burn injury and interventions to modify this response. Clin Plast Surg
36 (4):583–596, 2009.
5. Jones WG 2nd, Barber AE, Minei JP, Fahey TJ 3rd, Shires GT 3rd, Shires GT. Differential pathophysiology of bacterial translocation after thermal injury and sepsis. Ann Surg
214 (1):24–30, 1991.
6. Jones WG, Minei JP, Barber AE, Rayburn JL, Fahey TJ, Shires TGI, Shires TG. Bacterial translocation and intestinal atrophy after thermal injury and burn wound sepsis. Ann Surg
211 (4):399–405, 1990.
7. Latorre M, Krishnareddy S, Freedberg DE. Microbiome as mediator: do systemic infections start in the gut? World J Gastroenterol
21 (37):10487–10492, 2015.
8. Yao YM, Yu Y, Sheng ZY, Tian HM, Wang YP, Lu LR, Yu Y. Role of gut-derived endotoxaemia and bacterial translocation in rats after thermal injury: effects of selective decontamination of the digestive tract. Burns
21 (8):580–585, 1995.
9. Johnson TR, Gómez BI, McIntyre MK, Dubick MA, Christy RJ, Nicholson SE, Burmeister DM. The cutaneous microbiome and wounds: new molecular targets to promote wound healing. Int J Mol Sci
19 (9):E2699, 2018.
10. Plichta JK, Gao X, Lin H, Dong Q, Toh E, Nelson DE, Gamelli RL, Grice EA, Radek KA. Cutaneous burn injury promotes shifts in the bacterial microbiome in autologous donor skin. Shock
48 (4):441–448, 2017.
11. Earley ZM, Akhtar S, Green SJ, Naqib A, Khan O, Cannon AR, Hammer AM, Morris NL, Li X, Eberhardt JM, et al. Burn injury alters the intestinal microbiome and increases gut permeability and bacterial translocation. PLoS One
10 (7):1–16, 2015.
12. Cartotto R, Greenhalgh DG, Cancio C. Burn state of the science: fluid resuscitation. J Burn Care Res
38 (3):e596–e604, 2017.
13. Burmeister DM, McIntyre MK, Baker BA, Rizzo JA, Brown A, Natesan S, Chung KK, Christy RJ. Impact of isolated burns on major organs. Shock
14. Gómez BI, He C, Chao T, Dubick MA, Burmeister DM. Effect of intravenous fluid volumes on the adrenal glucocorticoid reponse after burn injury in swine
. J Burn Care Res
39 (5):652–660, 2018.
15. Gómez BI, McIntyre MK, Gurney JM, Chung KK, Cancio LC, Dubick MA, Burmeister DM. Enteral resuscitation with oral rehydration solution to reduce acute kidney injury in burn victims: Evidence from a porcine model. PLoS One
13 (5):E0195615, 2018.
16. Fouhy F, Clooney AG, Stanton C, Claesson MJ, Cotter PD. 16S rRNA gene sequencing of mock microbial populations-impact of DNA extraction method, primer choice and sequencing platform. BMC Microbiol
16 (1):1–13, 2016.
17. Allali I, Arnold JW, Roach J, Cadenas MB, Butz N, Hassan HM, Koci M, Ballou A, Mendoza M, Ali R, et al. A comparison of sequencing platforms and bioinformatics pipelines for compositional analysis of the gut microbiome. BMC Microbiol
17 (1):1–16, 2017.
18. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nat Methods
13 (7):581–583, 2016.
19. Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximum-likelihood trees for large alignments. PLoS One
5 (3):e9490, 2010.
20. 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 in trauma: significant changes in microbial diversity occur early after severe injury. Trauma Surg Acute Care Open
2 (1):e000108, 2017.
21. Rogers MB, Firek B, Shi M, Yeh A, Brower-Sinning R, Aveson V, Kohl BL, Fabio A, Carcillo JA, Morowitz MJ. Disruption of the microbiota across multiple body sites in critically ill children. Microbiome
4 (1):66, 2016.
22. Yeh A, Rogers MB, Firek B, Neal MD, Zuckerbraun BS, Morowitz MJ. Dysbiosis across multiple body sites in critically ill adult surgical patients. Shock
46 (6):649–654, 2016.
23. Nicholson S, Burmeister D, Johnson T, Zou Y, Lai Z, Scroggins S, DeRosa M, Jonas R, 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.
24. Nicholson SE, Merrill D, Zhu C, Burmeister DM, Zou Y, Lai Z, Darlington DN, Lewis AM, Newton L, Scroggins S, et al. Polytrauma independent of therapeutic intervention alters the gastrointestinal microbiome. Am J Surg
216 (4):699–705, 2018.
25. Walsh DM, McCullough SD, Yourstone S, Jones SW, Cairns BA, Jones CD, Jaspers I, Diaz-Sanchez D. Alterations in airway microbiota in patients with PAO2/FiO2 ratio < 300 after burn and inhalation injury. PLoS One
12 (3):1–20, 2017.
26. Shimizu K, Ogura H, Asahara T, Nomoto K, Matsushima A, Hayakawa K, Ikegawa H, Tasaki O, Kuwagata Y, Shimazu T. Gut microbiota and environment in patients with major burns: a preliminary report. Burns
41 (3):e28–e33, 2015.
27. Wang X, Yang J, Tian F, Zhang L, Lei Q, Jiang T, Zhou J, Yuan S, Wang J, Feng Z, et al. Gut microbiota trajectory in patients with severe burn: a time series study. J Crit Care
28. Huang G, Sun K, Yin S, Jiang B, Chen Y, Gong Y, Chen Y, Yang Z, Chen J, Yuan Z, et al. Burn injury leads to increase in relative abundance of opportunistic pathogens in the rat gastrointestinal microbiome. Front Microbiol
8 (Jul):1–9, 2017.
29. Hammer AM, Morris NL, Earley ZM, Choudhry MA. The first line of defense: the effects of alcohol on post-burn intestinal barrier, immune cells, and microbiome. Alcohol Res
37 (2):209–222, 2015.
30. Wheatley EG, Curtis BJ, Hulsebus HJ, Boe DM, Najarro K, Ir D, Robertson CE, Choudhry MA, Frank DN, Kovacs EJ. Advanced age impairs intestinal antimicrobial peptide response and worsens fecal microbiome dysbiosis following burn injury in mice. Shock
53 (1):71–77, 2020.
31. McIntyre MK, Peacock TJ, Akers KS, Burmeister DM. Initial characterization of the pig skin bacteriome and its effect on in vitro models of wound healing. PLoS One
11 (11):e0166176, 2016.
32. Crespo-Piazuelo D, Estellé J, Revilla M, Criado-Mesas L, Ramayo-Caldas Y, Óvilo C, Fernández AI, Ballester M, Folch JM. Characterization of bacterial microbiota compositions along the intestinal tract in pigs and their interactions and functions. Sci Rep
8 (1):1–12, 2018.
33. Sullivan TP, Eaglstein WH, Davis SC, Mertz P. The pig as a model for human wound healing. Wound Repair Regen
9 (2):66–76, 2001.
34. Ito I, Bhopale KK, Kobayashi M, Finnerty CC, Herndon DN, Suzuki F. Lamina propria group 2 innate lymphoid cells impair the antibacterial defense of burned mice to enterococcal translocation. J Leukoc Biol
102 (6):1451–1460, 2017.
35. Wareham DW, Wilks M, Ahmed D, Brazier JS, Millar M. Anaerobic sepsis due to multidrug-resistant Bacteroides fragilis: microbiological cure and clinical response with linezolid therapy. Clin Infect Dis
40 (7):e67–e68, 2005.
36. Nakamura I, Aoki K, Miura Y, Yamaguchi T, Matsumoto T. Fatal sepsis caused by multidrug-resistant Bacteroides fragilis, harboring a cfiA gene and an upstream insertion sequence element, in Japan. Anaerobe
37. Choi VM, Herrou J, Hecht AL, Teoh WP, Turner JR, Crosson S, Bubeck Wardenburg J. Activation of Bacteroides fragilis toxin by a novel bacterial protease contributes to anaerobic sepsis in mice. Nat Med
22 (5):563–567, 2016.
38. Wexler HM. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol Rev
20 (4):593–621, 2007.
39. Yuan ZQ, Peng YZ, Li XL, Huang YS, Yang ZC. Induction of heat shock protein 70 by sodium arsenite attenuates burn-induced intestinal injury in severe burned rats. Burns
34 (2):247–253, 2008.
40. Tadros T, Traber D, Heggers J, Herndon D. Effects of interleukin-1alpha administration on intestinal ischemia and reperfusion injury, mucosal permeability, and bacterial translocation in burn and sepsis. Ann Surg
237 (1):101–109, 2003.
41. Rex S. Burn injuries. Curr Opin Crit Care
18 (6):671–676, 2012.
42. Dulhunty JM, Boots RJ, Rudd MJ, Muller MJ, Lipman J. Increased fluid resuscitation can lead to adverse outcomes in major-burn injured patients, but low mortality is achievable. Burns
34 (8):1090–1097, 2008.
43. Cannon A, Akhtar S, Hammer A, Morris N, Javorski M, Li X, Kennedy R, Gamelli R, Choudhry M. Effects of mesalamine treatment on gut barrier integrity after burn injury. J Burn Care Res
37 (5):283–292, 2016.
44. Nowarski R, Jackson R, Gagliani N, Zoete MR, de Palm NW, Bailis W, Low JS, Harman CCD, Graham M, Elinav E, et al. Epithelial IL-18 equilibrium controls barrier function in colitis. Cell
45. Holman DB, Brunelle BW, Trachsel J, Allen HK. Meta-analysis to define a core microbiota in the swine
2 (3):e00004–e00017, 2017.
46. Obregon-tito AJ, Tito RY, Metcalf J, Sankaranarayanan K, Clemente JC, Ursell LK, Xu ZZ, Treuren W, Van Knight R, Gaffney PM, et al. Subsistence strategies in traditional societies distinguish gut microbiomes. Nat Commun
6 (6505):1–9, 2015.
47. Conlon MA, Bird AR. The impact of diet and lifestyle on gut microbiota and human health. Nutrients
48. Wu GD, Chen J, Hoffmann C, Bittinger K, Chen Y, Keilbaugh SA, Bewtra M, Knights D, Walters WA, Knight R, et al. Linking long-term dietary patterns with gut microbial enterotypes. Science
334 (6052):105–109, 2011.
49. Chen T, Long W, Zhang C, Liu S, Zhao L, Hamaker B. Fiber-utilizing capacity varies in Prevotella- versus Bacteroides-dominated gut microbiota. Sci Rep
7 (1):2594, 2017.
50. Hjorth MF, Roager HM, Larsen TM, Poulsen SK, Licht TR, Bahl MI, Zohar Y, Astrup A. Pre-treatment microbial Prevotella-to-Bacteroides ratio, determines body fat loss success during a 6-month randomized controlled diet intervention. Int J Obes
42 (2):580–583, 2018.