2-Chlorofatty acids are biomarkers of sepsis mortality and mediators of barrier dysfunction in rats.

Journal of Lipid Research Volume 61, 202

Sepsis represents a significant threat to overall global health, affecting a million Americans annually with mortality ranging from 12% to 25% in the United States (1,2).
In the present study, a rat sepsis model was developed that mimics early stages of infection and treatments of human sepsis, including fluid resuscitation and antibiotic therapy. The cecal slurry (CS) model in these studies improves reproducibility compared with the cecal ligation and puncture model (43). Rats, rather than mice, were chosen as the model animal due to their increased relative amounts of neutrophil MPO, which is more similar to that of humans (44). This model was highly reproducible and effective, leading to complete mortality that could be modulated with antibiotics and fluid resuscitation in a timedependent manner. We demonstrated that, at an early timepoint, plasma free 2-CLFA levels were higher in nonsurviving rats compared with survivors. A potential role of 2-CLFAs as mediators of organ failure is suggested by the finding that in in vivo rat studies, exogenous 2-CLPA increased permeability of Evans blue in the kidney, and in vitro, 2-CLPA causes loss of intestinal epithelial barrier function and vascular cell adhesion molecule-1 (VCAM-1) surface expression.

Materials
Rats were supplied by Envigo (Harlan, Indianapolis, IN). All rats were young adult male Sprague-Dawley weighing between 270 and 330 g. All animals were maintained in a temperature-and humidity-controlled room with a 12 h light/dark cycle and unrestricted access to chow and water. Upon arrival at Saint Louis University, rats were acclimated to the environment for at least a week prior to experiments. Sterile saline for rat experiments was supplied by B. Braun Medical (Bethlehem, PA). Pharmaceutical ceftriaxone was supplied by Pfizer (Hospira, Lake Forest, IL). Evans blue dye was supplied by Sigma-Aldrich (Vienna, Austria; catalog number E2129-10G). Brain heart infusion (BHI) agar was supplied by Sigma-Aldrich (catalog number 70138-500G). 2-CL-[d 4 ]palmitic acid (PA) and 2-CLPA were synthesized as previously described (21

Rat CS studies
All animal experiments were conducted with the approval of the Institutional Animal Care and Use Committee at Saint Louis University. CS was prepared by harvesting the cecal contents of healthy male Sprague-Dawley rats. Rats were euthanized by injecting 0.5 ml Somnasol (390 mg/ml sodium pentobarbital and 50 mg/ml phenytoin sodium) intraperitoneally with a 20 gauge needle followed by thoracotomy. Cecal contents were collected and mixed with sterile water (0.5 ml water per 100 mg cecal content). Water-content mixture was then filtered through two sequential meshes (Newark; 860 m and 180 m) to remove undigested food matter. The filtrate was then mixed thoroughly using a stir bar with sterile 30% glycerol (Sigma-Aldrich; catalog number G7757-1L) in PBS in a 1:1 ratio by volume. The resulting slurry in 15% glycerol was aliquoted into 5 ml cryovials (Corning; catalog number 430656) and placed in plastic cryogenic freezing containers (Thermo, Nalgene; catalog numbers 5100-0036 and 5100-0050) at 80°C. Prior to intraperitoneal administration for sepsis studies, aliquots of CS were thawed quickly in warm water. For certain experiments, CS was "heat-killed" (HK) by incubating aliquots of slurry in an 80°C water bath for 45 min. Slurry was then allowed to cool to room temperature prior to administration.
Rats were administered 15 ml/kg CS or 15% glycerol vehicle control (intraperitoneally) in a total volume of 20 ml/kg, with the remaining 5 ml/kg being sterile saline. At the time of CS administration, animals were administered a concurrent 30 ml/kg dose of subcutaneous sterile saline. Animals were monitored continuously using an adapted clinical scoring system (45)(46)(47). When antibiotics were delivered, 25 mg/kg ceftriaxone (Hospira) in sterile saline was administered intramuscularly in the hind limb in a 1 ml/kg volume. A second subcutaneous 30 ml/kg dose of sterile saline was administered concurrently with the ceftriaxone. For mortality experiments, once the animals were moribund, animals were euthanized using 0.5 ml Somnasol (intraperitoneally) followed by thoracotomy. Following euthanasia, organs were collected and immediately frozen on dry ice. Blood was collected via cardiac puncture, and plasma immediately prepared and then stored at 80°C.
For Evans blue dye studies, 2% Evans blue dye solution was made in sterile PBS. Evans blue [2.5 ml/kg (50 mg/kg)] was administered intravenously through the lateral tail vein using a 25 gauge needle. Thirty minutes following Evans blue administration, rats were euthanized using 0.5 ml Somnasol (intraperitoneally) followed by thoracotomy. Gravity perfusion was performed using PBS at 45 cm. PBS (75 ml) was perfused over 10 min through the right ventricle, followed by 75 ml over 10 min through the left ventricle. Incisions in the right atrium and the right femoral vein were made to allow for perfusate exit from the circulatory system. Following perfusion, lung, liver, and kidney were removed. One hundred milligrams of tissue were homogenized in 1 ml of formamide (Sigma-Aldrich, catalog number 221198-IL) and incubated in a 60°C water bath for 24 h. Tissue was then centrifuged for 30 min at 5,000 g. Supernatant was removed and absorbance analyzed at 620 nm using a spectrophotometer to determine concentration of Evans blue (48).

Exogenous 2-CLPA administration
Rats were administered 50 mg/kg 2-CLPA or the molar equivalent of PA intravenously through the lateral tail vein. The fatty acids were administered in 3.5% DMSO in sterile saline in a 1 ml/ kg volume using a 25 gauge needle. Either 30 min or 4 h following fatty acid administration, a blood sample was taken from the lateral tail vein using a 23 gauge needle. Immediately after, 2.5 ml/ kg 2% Evans blue were administered intravenously through the lateral tail vein using a 25 gauge needle. Ten minutes later, rats were euthanized and perfused as described above.

Cell surface expression of adhesion molecules
HSIECs were grown to confluence in 16 mm culture dishes coated with gelatin-based coating solution (catalog number 6950, Cell Biologics). Cells were incubated with 10 M PA or 2-CLPA for 4 h at 37°C in 95% O 2 /5% CO 2 . At the end of the incubation period, the medium was quickly removed, the cells were washed with PBS, immediately fixed in 1% paraformaldehyde, and incubated at 4°C overnight. Cells were washed three times with PBS and then blocked with Tris-buffered saline with 0.1% Tween 20 supplemented with BSA (w/v) and 0.5% fish gelatin (w/v) for 1 h at room temperature. Cell cultures were incubated with mouse anti-intercellular adhesion molecule-1 (ICAM-1) (catalog number sc-53336) or mouse anti-VCAM-1 (catalog number sc-13160, Santa Cruz Biotechnology) at 1:50 dilution for 1 h at room temperature, followed by incubation with horseradish peroxidaseconjugated goat anti-mouse antibody (1:5,000 dilution; Sigma Chemical Co.). Cultures were incubated in the dark for 30 min with 3,3′,5,5′-tetramethylbenzidine liquid substrate. Reactions were stopped by the addition of sulfuric acid, and color development was measured with a microtiter plate spectrophotometer at 450 nm.

Measurement of electrical resistance (permeability)
HSIECs were grown to confluence on Transwell ® polycarbonate membrane cell culture inserts (catalog number 3413, Corning). Resistance across the epithelial monolayer was measured daily, and experiments were performed once resistance was stable over three consecutive days. HSIECs were incubated with 10 M PA or 2-CLPA for up to 6 h. Changes in electrical resistance were measured over time using an epithelial volt ohmmeter (World Precision Instruments).

Lipid extraction and analyses
Free 2-CLFAs in plasma samples were analyzed by spiking 25 l of plasma with 103 fmol 2-CL-[d 4 ]-PA followed by the addition of 475 l saline. A modified Dole extraction was performed, and the heptane layer was dried down (49,50). Lipids were suspended in 260 l of 85/15 methanol/water with 0.1% formic acid. Total plasma 2-CLFAs were analyzed by spiking 25 l of plasma with 103 fmol 2-CL-[d 4 ]-PA followed by the addition of 155 l water. Two hundred microliters of 1 M NaOH were added, and samples were incubated at 60°C for 2 h. Hydrolysis was stopped by the addition of 120 l of 2 N HCl. After 10 min, lipids were extracted via a modified Dole extraction, and the heptane layer was dried down under nitrogen gas. Lipids were suspended in 260 l of 85/15 methanol/water with 0.1% formic acid (49,51).
Tissue analysis was preceded by pulverizing tissue in liquid nitrogen with a mortar and pestle.  (52). The collected chloroform layer was dried and then suspended in 180 l water. This extract was then subjected to base hydrolysis and subsequently prepared for LC/MS analyses (vide supra). Lipids were suspended in 260 l of 85/15 methanol/water with 0.1% formic acid.
After samples were resuspended in 85/15 methanol/water with 0.1% formic acid, all 2-CLFA analyses were conducted by LC/MS using ESI and selected reaction monitoring following previously described methods on the triple quadrupole mass spectrometer (49,51). LC was performed using a Thermo Fisher Accela LC system with a monolithic C18 column, solvent gradients previously described (49,51). MS was performed with a Thermo Fisher triple quadrupole Quantum Ultra mass spectrometer using XCalibur software (Thermo Fisher).
FFA analyses in tissue samples were conducted by spiking 10 mg pulverized tissue or 50 l plasma with 0.5 g heptadecanoic acid (fatty acid internal standard). A modified Bligh-Dyer extraction was performed, and the chloroform layer was dried down under nitrogen gas (52). Lipids were then suspended in 450 l chloroform. Thirty microliters of the suspended lipids were dried down and 100 l 2.5% diisopropylethylamine and 5% pentafluorobenzyl bromide in acetonitrile were added. Samples were incubated at 45°C for 1 h. Next, samples were twice sequentially suspended in 1 ml ethyl acetate and dried under nitrogen. Samples were then suspended in 80 l of ethyl acetate. FFAs were detected by GC/MS and selected ion monitoring following previously described methods (53).

Statistics
Student's t-test was used to compare two groups. ANOVA with either Dunnett's or Tukey's post hoc test was used for comparisons between three or more groups as indicated. Data are presented as median with whiskers extending to interquartile ranges. *, **, ***, **** are used to designate P-values less than 0.05, 0.01, 0.001, and 0.0001, respectively.

CS sepsis, endotoxemia, and antibiotic treatment
Initial studies examined the time course of mortality following 15 ml/kg CS treatment with concomitant fluids (30 ml/kg subcutaneous saline). This treatment led to mortality in all rats within 26 h with a median time of death of 16.5 h (Fig. 1A). As expected, vehicle treatments did not lead to mortality. To closely mimic clinical sepsis, a 25 mg/kg ceftriaxone dose was administered intramuscularly at selected times following CS treatment, and an additional 30 ml/kg of subcutaneous saline were given concurrently with the ceftriaxone (54,55). Results from these studies showed a time-dependent response in mortality to the concurrent antibiotic and saline administration similar to the mortality risk observed in humans in response to the timing of antibiotic treatment (Fig. 1B) (56). Delaying treatment until 12 h following CS resulted in little change to overall mortality, with only one survivor (1 out of 11). Median time of death was delayed slightly to 18 h compared with the absence of ceftriaxone treatment. In contrast, ceftriaxone treatment 8 h following CS treatment led to improved survivability (4 out of 15 rats survived), and for those rats that did not survive, median time of death was 23 h.
To ensure CS-elicited mortality was due to an active infection (a sepsis model) instead of a systemic inflammatory response to pathogen molecular patterns (an endotoxemia model), some rats received CS that had been heat treated at 80°C for 45 min prior to intraperitoneal injection. To assess the effect of the heat treatment, CS aliquots were plated on BHI agar and incubated at 37°C overnight. Slurry that had only been thawed quickly in warm water resulted in 8.5 × 10 7 colony-forming units (CFU) per milliliter, whereas the heat treatment resulted in 280 CFU/ml, a 300,000-fold reduction in bacterial viability. All rats survived treatments with HK CS (Fig. 1B). An active infection was confirmed in CS-treated rats by demonstrating a 10,000-fold increase in colony-forming units per milliliter in the blood (Fig. 1C) and a millionfold increase in colony-forming units per gram liver tissue in comparison to cultures from vehicle-treated rats (Fig. 1D). Additionally, the time-dependent relationship between antibiotic administration and mortality indicates the role of the active infection. To further verify that ceftriaxone was an appropriate antibiotic in the model, aliquots of diluted CS were plated on BHI agar plates with increasing concentrations of ceftriaxone (Fig. 1E). CS bacterial growth was significantly reduced in a dose-dependent manner in the presence of ceftriaxone, indicating that this was an appropriate antibiotic selection. Taken together, these findings suggest that the mortality seen in the CS model is directly due to an active infection rather than an inflammatory response to pathogen molecular signatures, making this a sepsis model rather than an endotoxemia-like model.

Kidney and liver microcirculatory permeability leakage during CS sepsis
In order to determine that the CS model resulted in systemic endothelial barrier dysfunction, Evans blue extravasation into organs was determined (Fig. 2). Plasma concentrations of Evans blue were unchanged when comparing the vehicle and CS rats, indicating that the Evans blue administration was consistent between groups (data not shown). The septic animals had significantly more Evans blue extravasation systemically than the vehicle animals. Increased Evans blue extravasation was seen in the liver and kidney, but not in the lung ( Fig.  2A-C). Additionally, there was a vast increase in both the concentration of Evans blue present in fluid in the peritoneal cavity (Fig. 2D) and total fluid volume (10-fold more) in CS-treated rats compared with vehicle-treated rats. Only half of the vehicle animals had any obtainable peritoneal fluid. Because the CS-treated and vehicletreated animals received the same total volume per kilogram intraperitoneally, the increase in abdominal fluid in CS-treated rats reflects either an inhibited fluid resorption process or increased leakage into the abdominal cavity. The volume of fluid recovered in the abdominal cavity was greater than the volume injected in the CS animals, therefore it is likely that at least some of the fluid is due to hyperpermeability. Taken together, these results Vehicle treatment was 15 ml/kg 15% glycerol in PBS in a total volume of 20 ml/kg. Rats were also administered 30 ml/kg sterile saline subcutaneously concurrent with CS/vehicle administration. Rats were monitored continuously and euthanized upon reaching a moribund state. B: Conditions were identical to those in A except that either 8 h or 12 h post-CS treatment, rats were administered 25 mg/kg ceftriaxone (ABX) concomitant with subcutaneous 30 ml/kg sterile saline. C: For the nonsurviving rats that received 8 h ABX, as shown in B, whole blood colony-forming units were determined. Student's t-test was used to compare log(CFU) per milliliter of blood. ****P < 0.0001. D: For the nonsurviving rats that received 8 h ABX, as shown in B, colony-forming units in liver were determined. Student's t-test was used to compare log(CFU) per gram of tissue. ****P < 0.0001. E: BHI agar plates were prepared with increasing concentrations of ceftriaxone, and 50 l of a 1,000-fold dilution of CS were plated on the plates. Plates were incubated at 37°C overnight, and colony-forming units were counted. Ceftriaxone (0 mg/l) resulted in innumerable colonies. Graph represents mean ± SEM for n = 4. In some instances, SEM bars are within the data points.
indicate that CS administration leads to systemic endothelial dysfunction.

Free 2-CLFA plasma levels associate with rat sepsis mortality
Because ceftriaxone treatment 8 h post-CS administration leads to both survivor and nonsurvivor outcomes, this model was investigated further to determine that rat sepsis plasma 2-CLFA levels associated with mortality similar to previous studies demonstrating plasma free and total 2-CLFA levels assessed from specimen collection taken during admission in the ICU (defined in that study as day 0) associate with human sepsis mortality (30). In rats treated with CS and ceftriaxone 8 h post-CS treatment, we observed a 73% mortality rate with median death at 23 h and the earliest death at 16 h. Blood samples were taken at 12 h pre-CS/vehicle and 8 h post-CS/vehicle, just prior to ceftriaxone treatment. There was no significant increase in plasma levels of 2-CLFA molecular species in rats sampled 12 h before and 8 h following vehicle-treatment in either free or esterified pools (Fig. 3A-D). Both 2-CLPA and 2-CLSA levels in esterified pools were significantly Fig. 2. Evans blue dye extravasation following CS administration. Rats were administered 15 ml/kg CS or vehicle in 20 ml/kg total volume. Seven and one-half hours later, 2.5 ml/kg 2% Evans blue dye in PBS were injected in the rat lateral tail vein, and Evans blue extravasation in lung (A), liver (B), and kidney (C) were measured as described in the Materials and Methods. In D, any abdominal fluid that was present in the peritoneal cavity was collected following euthanasia, and the volume was recorded and absorbance at 620 nm was measured. Data are presented as median with whiskers extending to interquartile ranges. Student's t-test was used to compare groups. *P < 0.05, ***P < 0.001, ****P < 0.0001. , and esterified 2-CLSA (D) were measured as described in "Materials and Methods." ANOVA with Tukey's post hoc test was used for multiple comparisons between all groups. Data are presented as median with whiskers extending to interquartile ranges. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 between indicated groups. increased in CS-treated rats compared with vehicle-treated rats in both survivors and nonsurvivors (Fig. 3C, D). Free 2-CLSA was elevated in both survivors and nonsurvivors compared with vehicle-treated rats, but free 2-CLPA was only elevated in nonsurvivors. (Fig. 3A, B). In addition, free 2-CLPA and 2-CLSA plasma levels were significantly elevated in nonsurvivor rats compared with survivor rats (Fig.  3A, B). Plasma levels of free 2-CLSA, but not free 2-CLPA or esterified 2-CLPA or 2-CLSA, were elevated (compared with vehicle-treated) in rats treated with nonfatal HK CS (Fig. 3A-D). Furthermore, free 2-CLFA plasma levels in HK CS rats were significantly lower than those measured in CS nonsurvivors and were similar to those levels in rat CS survivors (Fig. 3A, B). Both free and esterified levels of 2-CLFA were analyzed in the CS survivor cohort on a daily basis following the sampling at 8 h post-CS treatment, which revealed 2-CLFAs peaked at 8 h post-CS (day 0, Fig.  4) and returned to pre-CS levels 1 day afterward.

Temporal course of 2-CLFA levels in plasma and organs in response to CS sepsis
Separate cohorts of rats were treated with CS to examine plasma and organ 2-CLFA levels at 4, 8, 12, and 20 h post-CS treatment. In this study, 25 mg/kg ceftriaxone and an additional 30 ml/kg saline were administered 8 h posttreatment, such that analyses 12 and 20 h post-CS treatment included this additional intervention. Plasma levels of both free 2-CLPA and 2-CLSA were significantly elevated at 4 and 8 h following CS treatment compared with vehicle treatment (assessed 8 h following vehicle treatment) and were decreased to near vehicle-treatment levels 12 h following CS treatment (Fig. 5A, B). In comparison, plasma levels of both esterified 2-CLPA and 2-CLSA remained elevated 12 h following CS treatment, and while esterified 2-CLPA returned to baseline levels 20 h following CS treatment, plasma esterified 2-CLSA remained elevated (Fig. 6A, B). To investigate the possibility that antibiotic administration directly resulted in the decrease in free plasma 2-CLFA from 8 to 12 h, free 2-CLFA levels in the 8 h tail vein bleeds were compared with the 12 h tail vein bleeds (from the survival experiment). Those animals received neither antibiotics nor fluids at 8 h, and there were significantly less free 2-CLFAs, indicating that the decrease in plasma free 2-CLFA levels from 8 to 12 h is not a result of rescue therapy (supplemental Fig. S1). Free 2-CLFA levels in the lung of CS-treated rats were significantly elevated at all timepoints evaluated compared with vehicle-treated rats (Fig.  5C, D). In contrast, esterified 2-CLFA levels in the lung remained unchanged compared with vehicle-treated levels (Fig. 6C, D). Free 2-CLPA levels were only elevated 4 h post-CS treatment in the kidney, while free 2-CLSA was elevated 4 h and 8 h post-CS (Fig. 5E, F). Esterified 2-CLSA in the kidney was significantly elevated at 8 h post-CS, but not at other timepoints. Esterified 2-CLPA was not elevated in the kidney at any of the assessed time intervals (Fig. 6E, F).
Only free liver 2-CLSA levels, not 2-CLPA levels, were elevated 4-20 h post-CS treatment (Fig. 7A, B). However, both esterified 2-CLPA and 2-CLSA levels were elevated 4-12 h post-CS treatment in the liver (Fig. 8A, B). Free 2-CLFA levels were elevated 4 and 8 h post-CS treatment in the ileum (Fig. 7C, D) and 4-20 h post-CS in the colon (Fig. 7E, F). Ileum and colon esterified 2-CLFAs were elevated 4-20 h post-CS (Fig. 8C-F). Free and esterified 2-CLFA levels were also elevated in the spleen 4 and 8 h following CS treatment but were only elevated in the heart esterified pool at 12 and 20 h post-CS treatment (supplemental Figs. S2, S3). To verify that the free 2-CLFA's increase in plasma, liver, and lungs is specific to the chlorinated fatty acids, and not their nonchlorinated analogs, free palmitic and stearic acid were determined (supplemental Fig. S4). Lungs had no increase at any timepoint in either nonchlorinated fatty acid molecular species, while liver and plasma had an increase in both molecular species at 20 h, presumably as a metabolic response associated with the rats' decreased eating activity.

Exogenous 2-CLFA causes renal permeability barrier dysfunction
Thirty minutes following 50 mg/kg 2-CLPA administration (intravenous via lateral tail vein) the average plasma 2-CLPA levels reached 100 nM (Fig. 9A). This corresponds with the upper level of plasma levels of 2-CLFA observed in human sepsis patients (30). At this same timepoint (30 min Fig. 4. Survivor plasma 2-CLFA levels return to baseline within 1 day. Blood samples were taken from rats 12 h prior to (prebleeds) and 8 h after receiving CS as well as at indicated times in rats that survived CS treatments. Plasma levels of 2-CLFA species including free 2-CLPA (A), free 2-CLSA (B), esterified 2-CLPA (C), and esterified 2-CLSA (D) were measured as described in the Materials and Methods. ANOVA with Dunnett's post hoc test was used for comparisons between all days to the prebleed. All data are presented as median with whiskers extending to interquartile ranges. **P < 0.01 and ***P < 0.001 for comparisons to day 1.
following treatment), 2-CLPA treatment, compared with PA treatment, caused an increase in Evans blue extravasation in the kidneys (Fig. 9B), indicating loss of the permeability barrier in the kidney. In order to determine whether this effect on the kidney persisted for a longer time period following the fatty acid administration, Evans blue extravasation was determined 4 h following fatty acid treatment. At 4 h, plasma 2-CLPA levels remained elevated compared with PA-treated rats (Fig. 9C), albeit the levels had decreased compared with sampling at 30 min (Fig. 9A). Kidney Evans blue extravasation remained elevated in 2-CLPA-treated rats compared with PA-treated rats, but to a lesser degree (Fig. 9D).

2-CLFA causes loss of intestinal epithelial barrier function and epithelial adhesion molecule surface expression
Because CS treatment resulted in increased Evans blue extravasation suggesting systemic endothelial dysfunction and because we previously reported mesenteric microcirculation is sensitive to 2-CLPA treatment in intravital studies (35), further studies examined the impact of physiological levels of 2-CLPA on human intestinal epithelial cells. Previous studies suggest that 10 M 2-CLPA is a concentration reached at sites of inflammation (22,28,31). This concentration of 2-CLPA was shown to elicit VCAM-1 surface expression on human intestinal epithelial cells (Fig. 10A). In comparison, no effect was observed with the nonhalogenated fatty acid, PA. Further studies showed a significant loss of barrier function, as measured by changes in electrical resistance, of these cells over 6 h when treated with 2-CLPA (Fig. 10B).

DISCUSSION
Systemic endothelial barrier dysfunction leads to the development of organ injury in sepsis, but to date, no treatments targeting the endothelial barrier dysfunction in sepsis have been successful (6,7,57). A better understanding of mediators and mechanisms in sepsis-elicited barrier dysfunction could provide targets for future therapies. Neutrophils and MPO are involved in sepsis-related organ injury (8-13, 41, 42). MPO-derived HOCl can target membrane plasmalogens, leading to the formation of 2-CLFALD (20), which then may be oxidized to the more stable metabolite, 2-CLFA (22). Our group and others have shown that 2-CLFA and 2-CLFALD induce endothelial permeability and NETosis in vitro and in vivo (33,(35)(36)(37)58). Additionally, plasma 2-CLFA levels on day of admission associate with development of ARDS and 30 day mortality in human sepsis patients (30). This study was designed to investigate chlorinated lipids in an animal model of sepsis. Sepsis induced by CS administration was lethal in the absence of antibiotic rescue therapy. Some protection was afforded by antibiotic therapy, but even with this marginal therapy, microcirculatory dysfunction was observed using organ Evans blue extravasation as an indicator of microcirculatory function. The mortality observed in the CS model was a result of an active infection, as demonstrated by: the time-dependent increase in survival with antibiotic treatment, the presence of proliferating bacteria in blood and liver homogenates in CS-administered rats, and a complete lack of mortality in rats administered HK CS.
When antibiotics were administered 8 h following CS, survival improved from 0% to 26.6%. At 8 h following CS treatment, nonsurvivors had higher plasma free 2-CLFA levels compared with survivors. These data are similar to our previous findings showing plasma free and total 2-CLFA levels associate with mortality in human sepsis (30). Plasma free 2-CLFA concentrations were not statistically different at 8 h in comparisons between CS survivors and rats treated with HK CS, a treatment that did not lead to death. It should be recognized that animal studies limit sequential blood draws, and we spaced our blood sampling during these studies according to animal welfare guidelines. Additional information may be gleaned from temporal blood sampling for the analysis of plasma 2-CLFA levels in stratifying survivors versus nonsurvivors. For example, it is possible that times prior to 8 h may provide additional information, particularly because we did see increases in 2-CLFA in the earliest timepoint we assessed at 4 h. Significant increases in both free and esterified 2-CLFA were also found in kidney, liver, lung, spleen, colon, and ileum of rats treated with CS compared with vehicle. The absence of an increase in nonchlorinated palmitic and stearic acid in the liver, lungs, and plasma confirms that the increase in 2-CLFAs is not the result of a global increase in all fatty acids but has some specificity for an MPO-mediated mechanism. While our studies in humans implicated 2-CLPA as a prognostic indicator and potential mediator of ARDS in sepsis (30), the CS model under the conditions employed did not lead to lung injury, as indicated by no significant increases in lung Evans blue extravasation and other measures, including no increase in both the lung wet/dry weight and protein in bronchoalveolar lavage fluid (data not shown). Others have also indicated that intestinal sepsis models in rodents do not lead to lung injury (59). Interestingly, though, elevations in free 2-CLFAs were demonstrated in the lungs of CS-treated rats. It could be speculated that the rodent lung microvascular system may not be as sensitive to mediators of injury, such as 2-CLFAs, in intestinal sepsis models. Alternatively, there may be a directionality aspect of 2-CLFA-elicited barrier effects in the lung microcirculatory system, for in a pneumonia-sepsis context, 2-CLFA would presumably be generated within the alveolar space.
Dysfunctional neutrophils contribute to the microvascular dysfunction, tissue damage, and organ injury in sepsis (10,11). Neutrophil enzymes, such as neutrophil elastase and MPO have been directly implicated in microcirculatory defects (60,61). The studies herein expand our understanding of neutrophil-mediated pathogenesis by examining the role of MPO-derived 2-CLFA. Free 2-CLFA has been previously shown to elicit neutrophil NETosis, cause mesenteric microcirculatory dysfunction, and elicit changes in endothelial storage granule mobilization leading to adhesion molecule surface expression and barrier dysfunction (30,(35)(36)(37). Thus, we propose that the observed increases in free 2-CLFAs may be one of the neutrophil-derived inflammatory mediators contributing to organ dysfunction in the rat CS model. This is further supported herein by the demonstration that exogenous 2-CLPA treatment leading to plasma levels of 2-CLFA observed in human sepsis (30) elicits kidney Evans blue extravasation. Our sepsis model also led to severe intraperitoneal fluid loss and elevated levels of free 2-CLFAs in the ileum and colon, suggesting that 2-CLFA may contribute to leakage in the mesenteric microcirculation similar to that observed with exogenously applied 2-CLFA (35). Furthermore, in vitro studies now show the profound effect of 2-CLFA on intestinal epithelial barrier function, suggesting the CS treatment-elicited 2-CLFA elevations may contribute to gut barrier dysfunction, which may further exacerbate sepsis by increased intestinal microbiome flux into the peritoneum.
In contrast to the biological effects of free 2-CLFA, esterified 2-CLFA may represent a historical cumulative production of 2-CLFA. No evidence suggests the 2-CLFA esterified pool has biological activity. Esterifying 2-CLFA may be one way to detoxify free 2-CLFA. Esterified 2-CLFA may also be the method of transport of 2-CLFA, potentially through HDL-mediated transport of complex lipids in the Data are presented as median with whiskers extending to interquartile ranges. ANOVA with Dunnett's post hoc test was used for comparisons between all timepoints to vehicle 8 h. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. rat. Interestingly, the free/esterified 2-CLFA ratio in the liver at 8 h was extraordinarily skewed toward the esterified pool, compared with other organs, suggesting that the liver might be the destination for 2-CLFA transport. The finding that free 2-CLFA levels, and not esterified 2-CLFA levels, are elevated in nonsurvivor CS-treated rats compared with survivors suggests that the conversion of free 2-CLFA into esterified 2-CLFA pools may be limiting in rats that succumb to CS sepsis. Because the incorporation of fatty acids into esterified pools requires conversion of fatty acid to acyl Conditions are identical to those described in Fig. 5. Organs were collected and free 2-CLFA species quantified as described in the Materials and Methods. Data are presented as median with whiskers extending to interquartile ranges. ANOVA with Dunnett's post hoc test was used for comparisons between all timepoints to vehicle 8 h. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Fig. 9. 2-CLPA causes Evans blue extravasation in the kidney. A, C: Plasma free 2-CLPA levels from rats treated with either 2-CLPA or PA at indicated times posttreatment. B, D: Evans blue extravasation in the kidney at indicated times posttreatment measured as described in the Materials and Methods. Data are presented as median with whiskers extending to interquartile ranges. Student's t-test was used to compare groups. *P < 0.05 and ****P < 0.0001.
CoA, an ATP-dependent process, it is plausible that nonsurvivor metabolic capacity to produce 2-CLFA CoA intermediates is limiting. Additionally, the inability to convert free 2-CLFA to esterified 2-CLFA in nonsurvivors may be due to the profound microcirculatory collapse during sepsis, reducing transit of free 2-CLFA from sites of production to key lipid metabolizing organs (e.g., the liver). Although not determined in the current study, previous studies with endothelial cells and human neutrophils suggest that 2-CLPA, produced from 2-chlorohexadecanal oxidation, is incorporated into phosphatidylcholine, phosphatidylethanolamine, cholesteryl ester, and triacylglycerol pools (21).
These studies demonstrate that the CS sepsis model with antibiotic resuscitation represents a good model to examine survivors and nonsurvivors of sepsis and to evaluate these cohorts for biomarkers and mediators of the multiorgan injury leading to sepsis mortality. Plasma free 2-CLFAs associated with mortality in this rat sepsis model, and additional evidence suggests that 2-CLFAs elicit kidney and gut dysfunction. It will likely be beneficial to develop targets to modulate the role of chlorinated lipids in sepsis. Although MPO inhibition should reduce 2-CLFA, it is not therapeutically advisable due to MPO's bactericidal role. It is known that PPAR- regulates the metabolism of 2-CLFA in the liver (62), and PPAR- has also been shown to be important in sepsis survival (63)(64)(65)(66). Alternatively, understanding the mechanisms how 2-CLFA elicits permeability barrier dysfunction may lead to therapeutic targets directed at this action. In the future, devising mechanisms to block the deleterious impact of chlorinated lipids on endothelial, epithelial, and neutrophil function during sepsis may lead to new therapeutic interventions.

Data availability
All data are contained within the article.