Trauma and Hemorrhage Lead to An Elevation in Fecal Hort-Chain Fatty Acids.

Short-chain fatty acids (SCFA) are naturally occurring 1–6 carbon aliphatic-chain carboxylic acids. They are the fermentation products of dietary fiber by the microbiome of the intestines. The human gut produces many of the SCFA through degradation of carbohydrates and amino acids ( 1 – 3 ). The gut microbiome regulates gut barrier function, immunity, and the intake of metabolic nutrients ( 4 ). It has been suggested that SCFA found in the gut could be used as markers for a healthy gut ecosystem ( 4 ) and the measurement of SCFA an index of microbiome health and therefore overall health. Changes in the gut microbiome are associated with inflammation and immune responses ( 1 , 4 – 7 ), Parkinson’s disease ( 8 ), Alzheimer’s disease ( 9 ), inflammatory bowel disease ( 7 ), liver disease ( 10 ), trauma, and hemorrhage ( 11 – 14 ). SCFA supplementation has been suggested for improving outcome in liver disease ( 10 ), endometriosis ( 15 ), and spinal cord trauma ( 16 ). The measurement of SCFA composition may reflect changes in the microbiome and therefore may serve as a biomarker for changes in the microbiome in pathological conditions. We have previously shown that trauma and hemorrhage change the beta diversity of gut bacteria in rat ( 13 ), including an increase in the fecal levels of families Lachnospiraceae, Mogibacteriaceae, and Deferribacteraceae, and decrease in families Barnesiellaceae and Bacteroidaceae. Because SCFA are produced by the gut microbiome, trauma and hemorrhage may change SCFA composition as well. The measurement of SCFA is difficult and performed almost solely by liquid chromatography tandem mass spectroscopy (LC-MS/MS) after conjugation to ring-structured amines for better chromatography and detection by MS/MS. We have recently developed a method using bicycle-amines to measure SCFA in plasma and tissue samples that has better sensitivity and specificity ( 17 ). Here, we used this method to measure 21 SCFA in feces and plasma after trauma and hemorrhage.

All chemicals were purchased at the highest quality of purity supplied by the manufacturers. 2′,2′-Dipyridyl disulfide, tri-phenylphosphene, formic acid, acetic acid, propionic acid, 2-phenyl propionic acid, 3-phenyl propionic acid, butyric acid, 2-methyl propionic acid, 2-phenyl butyric acid, 3-phenyl butyric acid, 4-phenyl butyric acid, pentanoic acid, 2-methyl butyric acid, 3-methyl butyric acid, 4-methyl butyric acid, 2-phenyl pentanoic acid, 4-phenyl pentanoic acid, 5-phenyl pentanoic acid, hexanoic acid, 2-methyl pentanoic acid, 3-methyl pentano-ic acid, 4-methyl pentanoic acid, 5-phenyl hexanoic acid, 6-phenyl hexanoic acid, acetone, dichloromethane, ethyl acetate, 1-butanol, ammonium formate, ammonium acetate, and ammonium carbonate were purchased from MilliporeSigma. 3-Phenyl pentanoic acid and 6-phenyl hexanoic acid were purchase from Smolecule. 4-(Pyrrolidin-1-ylmethyl) benzylamine was purchased from Santa Cruz Biotechnologies. Acetonitrile (MeCN) was purchased from ThermoFisher. Not all Phenyl SCFA are represented as they could not be purchased as standards. This 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 International. All experiments were started between 0800–0900 h in a room separate from home caging. All rats (Sprague-Dawley from Charles Rivers, www.criver.com ) were male and were group housed. The light/dark cycle was 12 h light/12 h dark. The rats ate Laboratory Rodent Diet 5001 ( www.LabDiet.com ). Food and water were given ad libitum . Sprague-Dawley rats (225–250 g) were prepared as previously described ( 13 , 18 – 20 ). Briefly, the rats were anesthetized with 1%–2% isoflurane/100% oxygen through a nosecone and allowed to breathe spontaneously. Cannulas were placed in the left femoral artery and vein for monitoring arterial blood pressure and blood draw. Trauma was performed after a midline incision. The right and medial lobes of the liver received three crushes each using a clamp covered with silastic tubing. A 10-cm section of small intestines was isolated just anterior to the cecum was gently crushed. The intestines and liver were replaced. The abdominal incision was closed in two layers with sutures. Femur fracture was accomplished by dropping six stainless-steel balls (65 g each) stacked together, from 36″ through a guide tube (1″ internal diameter) to impact on a rounded aluminum blade resting on the midright femur of the right leg that was suspended on two aluminum stands, one under the hip and one under the knee. A large hemostat (5-inch tongs) was used to clamp the muscle of the right leg 10 times. The rats were then immediately bled and maintained at a mean arterial pressure of 40 mmHg until 20% of estimated blood volume was removed. Hemorrhage was then discontinued, and blood pressure and heart rate were allowed to freely compensate. In the resuscitation group, fresh whole blood was collected in CDP (at 1:8 ratio) from the femoral artery of an anesthetized donor rat and was transfused at 1-h posttrauma. Blood volume was estimated as 6% of body weight + 0.77. All rats were euthanized at 2 h. Three groups of rats were used (n = 8/group). A sham group with no trauma or hemorrhage, a trauma/hemorrhage group (described above) and a trauma/hemorrhage group that was resuscitated with the shed blood at 1 h after trauma. At 2 h, all groups were euthanized. Just prior to euthanasia, a blood sample (1 mL) was taken over citrate (200 mM), centrifuged and plasma extracted and frozen at −80°C. Fecal samples were collected at time of euthanasia from the colon and frozen in liquid nitrogen. SCFAs were extracted from 100 μL of rat plasma or 100 mg of pulverized rat feces as previously described ( 17 ) with 1 mL of MeCN, vortexed, and the supernatant dried. The dried extraction was then used in the conjugation procedures described below. 8-Bromo-adenosine was used as an internal control for all samples to monitor the extraction and measurement efficiency throughout the process. The same amount of internal control was added to all samples and standards. All reactions were performed in a siliconized glass 12 × 75-mm tube as previously described ( 17 ). Dried extract (plasma or feces), 50 μL of triphenylphosphene (10 mM in MeCN), 50 μL 2,2′-dipyridyl di-sulfide (10 mM in MeCN) and 50 μL of 4-(pyrrolidin-1-ylmethyl) benzylamine (10 mM in MeCN) were combined. The mixture was incubated for 10 min at 60°C, diluted with 2.7 mL ddH 2 O, placed over HLB (Waters) solid phase ex-traction column, washed with 5% MeOH and eluded in 100% MeOH. The eluant was dried by spin dryer. The dried eluant was brought up in 200 μL 5% MeCN, 0.1% formic acid and analyzed by LC-MS/MS (Ultimate 3000-Quantiva, ThermoFisher). SCFA standards were conjugated be the same procedures (with internal controls) for generation of standard curves in nM (plasma) or nM per mg protein (feces). SCFA conjugates were separated by reverse phase chromatography on a Kinetex Polar C18 150 × 2.1 mm column (Phenomenex) at 250 μL/mL as previously described ( 17 ), starting at 5% MeCN, 0.1 formic acid, 12% at 4 min, 20% at 9 min, 100% at 13 min, and back to 5% at 15 min. Mass Spectroscopy (Quantiva, ThermoFisher) was performed using a triple quadruple configuration (Q1, Q2, and Q3) in selective reaction monitoring mode in both Q1 (for section of the parent SCFA conjugates) and Q3 (for selection of daughters). Q2 contained argon gas (1.5 torr) for collision-induced dissociation of parent to daughter. All samples were run at 10 μL in duplicate and standards run in triplicate. Heated-electrospray ionization (vaporizing temperature 100°C) was used to ionize the analytes prior to entering the MS (spray voltage 5,000 V, ion transfer tube 275°C, sheath gas 15, sweep gas 2 and auxiliary gas 5 torr). The precursor and product molecular weights used in the measurement of these SCFA have been previously described in detail ( 17 ). For standard curves, linear regression analysis was performed for calculation of slope, intercept, and correlation coefficient. Lower limit of detection (LOD) was calculated as the 3.3 time the Standard Deviation of the response (y-intercept) divided by the slope. Mean and standard error of the mean (SEM) were calculated for measurement of SCFA extracted from the feces and plasma. % coefficient of variation was calculated as the SD/mean × 100%. Group analysis was performed by ANOVA (sham, trauma/hemorrhage and trauma/hemorrhage/resuscitation) by GraphPad Prism (GraphPad Software, Boston, MA) after log transform. Kolmogorov-Smirnov test was used to show normality after log transform. Dunnett’s post-hoc test was used to compare groups to sham. Significant difference was at P  < 0.05. N = 8/group for feces and 6/group for plasma. Power of the ANOVA was set at a significance (a = 0.05) and power level (1-b = 0.8) and calculated to be n = 5–8/group.

Twenty-one different SCFA and phenyl SCFA were measured after conjugation to 4PyBA (Table 1 ). All were separated by chromatography and measured by MS/MS. Good separation of all SCFAs (and isomers) and the phenyl-SCFA (and isomers) was achieved and shown in Figure 1 A and 2 A, respectively. Separation of the isomers is critical in the chromatography as the measurement of each SCFA is dependent on the molecular weight, and isomers have the same molecular weight. Each peak was identified by running single standards of each SCFA. Single ion monitoring (Fig. 1 , B–D, and 2 , B–D) demonstrate the ability to identify and measure each isomer separately. Short chain fatty acids measured Table 1 shows all short chain fatty acids that could be measured in the assay. 2-, 3-, and 4-phenyl hexanoic acid could not be measured as those chemical standards could not be purchased as standards. Chromatograph showing separation of propionic, butyric, pentanoic, and hexanoic acid and all their isomers . (A) shows the total ion monitoring of all four SCFA and their isomers. Single ion monitoring of (B) propionic acid, (C) butyric acid, (D) pentanoic acid, and (E) hexanoic acid and their respective isomers demonstrate separation of different chemical entities. Chemical structures are associated with their respective peak of interest. SCFA, short-chain fatty acid. Chromatograph showing separation of phenyl-propionic, -butyric, -pentanoic, and -hexanoic acid and all their isomers . (A) shows the total ion monitoring of all four phenyl-SCFA and their isomers. Single ion monitoring of (B) phenyl-propionic acid, (C) phenyl-butyric acid, (D) phenyl-pentanoic acid, and (E) phenyl-hexanoic acid and their respective isomers demonstrate separation of different chemical entities. Chemical structures are associated with their respective peak of interest. SCFA, short-chain fatty acid. Examples of chromatographs (single ion monitoring) from feces and plasma are shown in Figure 3 with the corresponding molecular structure represented by each peak. All SCFA were successfully separated for measurement. Only 2- and 4-phenyl butyric acid could not be separate chromatographically (Fig. 2 C) so the data in the peak represents both isomers. Multiple ion monitoring was used to select and identify each SCFA in feces and plasma (Fig. 3 ) and the area under the curves were measured. Standard curves were generated for each of the 21 SCFA using concentrations between 1 nM and 10 μM. The standard curves showed good linearity (r 2  > 0.99). Accuracy for each standard curve was calculated as the LOD and was between 40 and 229 nM for all SCFA, and precision for each analyte at 1 μM for a 10 mL sample was below 7% as has been previously published ( 17 ). Chromatographs showing the ability to detect SCFA and phenyl-SCFA in rat feces and plasma . (A–D) shows single ion monitoring of SCFA in feces. E–H shows single ion monitoring of SCFA in plasma. Phenylated version of SCFA were found in feces (I and K) and plasma (J and L). Chemical structures are associated with their respective peak of interest. SCFA, short-chain fatty acid Of the 21 SCFA measured in feces, 11 were measurable above the lower limit of detection (Fig. 4 ). Of these, five demonstrated a significant rise after trauma and hemorrhage as compared to Sham. These included propionic, 2-methyl propionic, pentanoic, 4-methyl pentanoic, and 3-phenyl propionic acid. The addition of resuscitation significantly elevated propionic, butyric, 2-methyl propionic, pentanoic, hexanoic, and 4-methyl pentanoic acid as compared to sham. 2- and 3-Methyl butyric and 2/4-phenyl butyric showed no change to trauma/hemorrhage or resuscitation. Changes in fecal SCFA 2 h after sham, trauma and hemorrhage (T/Hem) or trauma, hemorrhage and resuscitation (T/H/Res ). Values represent mean ± SEM. * =  P  < 0.05 by ANOVA followed by Dunnett’s test compared to sham. N = 8/group. We also measured SCFAs in the plasma 2 h after trauma and hemorrhage at the same time the rats were euthanized, and the feces was collected. Of the 21 SCFAs, nine could be measured above the LOD. These include propionic, butyric, 2-methyl propionic, 3-phenyl propionic, 2- and 3-methyl butyric, hexanoic acid, and 2/4-phenyl butyric acid. However, trauma, hemorrhage or the addition of resuscitation did not significantly affect the level of these SCFA in plasma (Fig. 5 ). Changes in plasma SCFA 2 h after sham, trauma and hemorrhage (T/Hem) or trauma, hemorrhage and resuscitation (T/H/Res ). Values represent mean ± SEM. N = 6/group.

SCFA are the major metabolic product of gut bacteria after anaerobic fermentation of dietary fiber. Propionate, butyrate, pentanoate, and hexanoate are the end products of enzymatic digestion of carbohydrates and amino acids, and beta-oxidation in the mitochondria. Their isomers (branch chain fatty acids) are formed during the catabolism of branched-chain amino acids such as valine, leucine, iso-leucine ( 21 ). Phenylated SCFA are derived from phenylalanine, phenylacetyglycine, and the involvement of medium-chain acyl-CoA dehydrogenase ( 22 ). Bacteria that favor production of various SCFA include akkermansia Muciniphila and eubacterium halli for propionate, faecalibacterium prausnitzii, eubacterium rectale, eubacterium halli , and ruminococcus bromi for butyrate ( 23 , 24 ) and bracilus fragilis for phenylpropionic acid ( 25 ). Prebiotic manipulation of diet can shift the production of SCFA by gut bacteria for health benefits ( 4 ). Because SCFA are produced by the gut microbiome, they have been implicated in normal and pathological conditions that affect the microbiome. The use of SCFA as biomarkers or therapy is gaining favor ( 7 , 26 – 28 ) and the literature has many examples. Butyrate has been shown to protect against endometriosis through G-protein–coupled receptors, histone deacetylases, and a GTPase activating protein, RAP1GAP ( 15 ). In Parkinson’s Patients, plasma levels of propionic acid correlated with motor symptom severity, whereas plasma levels of butyrate and valerate (pentanoate) were associated with cognitive decline ( 8 ). SCFAs have anti-inflammatory effects through G protein-coupled receptor pathway (GPR41, 43, and 109a) and histone acetylase enabling it to alter the genetic profile of immune cells ( 7 ). Loss of micro-organisms that produce butyrate can potentially result in a proinflammatory response and alter the host response to infection or injury ( 29 , 30 ). Patients with inflammatory bowel disease show a decrease in the butyrate-producing bacteria, faecalibacterium prausnitzii, and butyrate treatment has a positive clinical impact on these patients ( 27 , 28 ). Additionally, colonic butyrate levels have also been shown to be depleted after antibiotic exposure or critical illness ( 31 ). Propionate has been shown to significantly attenuate cardiac hypertrophy, fibrosis, vascular dysfunction, and hypertension in hypertensive mice ( 32 ). 3-Phenylpropionic acid facilitates intestinal epithelial barrier function ( 25 ) and muscle growth ( 33 ). Most reports state that SCFA are produced predominately by gut bacteria and their solubility allows for movement into the blood stream for distribution throughout the body ( 1 , 4 , 5 , 24 , 34 – 36 ). We have recently reported that the SCFA and phenyl SCFA can be found in rat RBC and platelets, and in swine liver ( 17 ), and include propionic, butyric (and isobutyric), pentanoic (and its two isomers) and hexanoic acid, 2- and 3-phenyl propionic, and 2-,3-, 4-phenyl butyric acid. The presence of these SCFA in tissues suggest that tissues and organs take up circulating levels bacterial-derived SCFA or produce the SCFA themselves. In this study, we show that severe trauma and hemorrhage elevate SCFA in feces at 2 h. This finding correlates with the changes seen in beta diversity of the microbiome in colon of rats 2 h after severe trauma and hemorrhage ( 13 ). We previously have shown that trauma and hemorrhage in rat lead to an increase in the fecal levels of families Lachnospiraceae , Mogibacteriaceae , and Deferribacteraceae , and decrease in families Barnesiellaceae and Bacteroidaceae ( 13 ). In rabbit, hemorrhage led to a significant elevation of Clostidiales and Bacteroidales , and a fall in Lactobacillus at 24 h ( 14 ). In rat, beta-adrenergic blockade significantly changed the alpha-diversity 7 and 14 days after trauma and hemorrhage ( 12 ). These studies show that trauma and/or hemorrhage can alter the microbiome. The data in this report show that trauma and hemorrhage change the levels of SCFA in feces. This suggests that changes in SCFAs could be used as a biomarker to monitor changes in the microbiome in normal or pathologic conditions. However, we did not see a see changes in plasma SCFA at 2 h. This is disappointing as plasma levels of SCFA would be an easier biomarker to measure clinically. The 2-h time point may be too early in the time course following trauma and hemorrhage to detect SCFA changes in the plasma, and extending the time course might allow us to observe plasma changes as the majority of SCFA are made by the gut microbiome ( 13 ). It was interesting that the addition of resuscitation elevated butyric and hexanoic acid. This could be due to reperfusion of the gut microvasculature; restoration of gut integrity and promoting gut bacteria to produce and secrete these two SCFA. In conclusion, trauma and hemorrhage led to an elevation in propionic, 2-methyl propionic, pentanoic, 4-methyl pentanoic, and 3-phenyl propionic acid in rat feces. Furthermore, the addition of resuscitation significantly elevated butyric and hexanoic acid as compared to Sham. These changes may reflect similar changes in the beta diversity of gut bacterial of our previous report.