Fecal Microbiota Transplantation and Hydrocortisone Ameliorate Intestinal Barrier Dysfunction and Improve Survival in a Rat Model of Cecal Ligation and Puncture-Induced Sepsis

Abstract

Introduction: 

Sepsis is a life-threatening syndrome which can progress to multiple organ dysfunction with high mortality. Intestinal barrier failure exerts a central role in the pathophysiological sequence of events that lead from sepsis to multiple organ dysfunction. The present study investigated the role of hydrocortisone (HC) administration and fecal microbiota transplantation (FMT) in several parameters of the gut barrier integrity, immune activation, and survival, in a model of polymicrobial sepsis in rats.

Methods: 

Forty adults male Wistar rats were randomly divided into four groups: sham (group I), cecal ligation and puncture (CLP) (group II), CLP + HC (2.8 mg/kg, intraperitoneally single dose at 6 h) (group III), and CLP + FMT at 6 h (group IV). At 24 h post-CLP, ileal tissues were harvested for histological and immunohistochemical analyses while endotoxin, IL-6, and IL-10 levels in systemic circulation were determined. In a second experiment the same groups were observed for 7 days for mortality, with daily administration of hydrocortisone (group III) and FMT (group IV) in surviving rats.

Results: 

HC administration and FMT significantly reduced mortality of septic rats by 50%. These interventions totally reversed intestinal mucosal atrophy by increasing villous density and mucosal thickness (μm, mean ± SD: Group I: 620 ± 35, Group II: 411 ± 52, Group III: 622 ± 19, Group IV: 617 ± 44). HC and FMT reduced the apoptotic body count in intestinal crypts whereas these increased the mitotic/apoptotic index. Activated caspase-3 expression in intestinal crypts was significantly reduced by HC or FMT (activated caspase-3 (+) enterocytes/10 crypts, mean ± SD: Group I: 1.6 ± 0.5, Group II: 5.8 ± 2.4, Group III: 3.6 ± 0.9, Group IV: 2.3 ± 0.6). Both treatments increased Paneth cell count and decreased intraepithelial CD3(+) T lymphocytes and inflammatory infiltration of lamina propria to control levels. In the sham group almost the total of intestinal epithelial cells expressed occludin (92 ± 8%) and claudin-1 (98 ± 4%) and CLP reduced this expression to 34 ± 12% for occludin and 35 ± 7% for claudin-1. Administration of HC significantly increased occludin (51 ± 17%) and claudin-1 (77 ± 9%) expression. FMT exerted also a significant restoring effect in tight junction by increasing occludin (56 ± 15%) and claudin-1 (84 ± 7%) expression. The beneficial effects of these treatments on gut barrier function led to significant reduction of systemic endotoxemia (EU/mL, mean ± SD: Group I: 0.93 ± 0.36, Group II: 2.14 ± 1.74, Group III: 1.48 ± 0.53, Group IV: 1.61 ± 0.58), while FMT additionally decreased IL-6 and IL-10 levels.

Conclusion: 

Fecal microbiota transplantation and stress dose hydrocortisone administration in septic rats induce a multifactorial improvement of the gut mechanical and immunological barriers, preventing endotoxemia and leading to improved survival.

INTRODUCTION

Sepsis is a life-threatening syndrome which can progress to multiple organ dysfunction with an estimated in-hospital mortality of above 20%. Despite significant advances in our understanding of the pathophysiology of sepsis, the cornerstone of therapy remains unchanged for decades and consists of antibiotics, fluid resuscitation, vasopressor administration, and supportive care ⁽¹⁾. It is therefore important to investigate additional pathophysiologically-based treatment options in sepsis to improve patient outcomes.

The intestine exerts a central role in the pathophysiological sequence of events that lead from sepsis to multiple organ dysfunction, characterized as the “motor” of sepsis ⁽²⁾. Numerous experimental and clinical studies have demonstrated that the intestinal barrier is compromised in sepsis, leading to bacterial and endotoxin translocation. The excessive presence of endotoxins in the systemic circulation stimulates a systemic inflammatory response, characterized by the release of cytokines and other proinflammatory mediators, which may cause structural and functional deleterious effects on remote organs, resulting in multiple organ dysfunction ⁽³⁾. Current theories suggest that the gut in critical illness and sepsis becomes a pivotal proinflammatory organ, driving the systemic inflammatory response without the need for systemic bacterial translocation ⁽²⁾.

Hydrocortisone (HC) is suggested as a treatment option in septic shock that is non-responding to adequate fluid resuscitation and vasopressor therapy ⁽¹⁾. A recent large randomized controlled trial enrolling 1,241 patients with septic shock, demonstrated a significant survival benefit in those treated with hydrocortisone-plus-fludrocortisone ⁽⁴⁾. Sepsis is known to induce significant alterations of the intestinal microecology equilibrium which affects the integrity of the intestinal mechanical and immunological barrier through their continuous crosstalk ⁽²⁾. Fecal microbiota transplantation (FMT) has been proven efficacious in recurrent pseudomembranous colitis and in several other pathological conditions characterized by intestinal dysbiosis ⁽⁵⁾. The role of hydrocortisone and FMT therapy on the integrity of the gut barrier in sepsis has not been adequately investigated until now.

The present study investigated the role of HC administration and FMT in several parameters of the gut barrier integrity, immune activation and survival, in a model of polymicrobial sepsis in rats.

METHODS

Animals

Forty male albino Wistar rats, weighing 250 g to 320 g, were used. They were housed in stainless-steel cages, three rats per cage, under controlled temperature (23°C) and humidity conditions, with 12-h dark/light cycles, and maintained on standard laboratory diet with tap water ad libidum throughout the experiment, except for an overnight fast before surgery. Animals were randomly divided into four groups: group I: sham operated; group II: Cecal ligation and puncture (CLP); group III: CLP+ hydrocortisone (HC); group IV: CLP+FMT. Two separate studies, with 10 rats assigned in each experimental group, were performed to measure the studied parameters (24 h) and the survival (7 d), respectively. This preclinical animal study took into consideration in its design and execution the recommendations for “minimum quality threshold in preclinical sepsis studies for study design and humane modeling endpoints” ⁽⁶⁾. The sample size for mortality experiments was determined via power analysis with a power of 80%, α level of 0.05, making the assumption for over 50% reduction of mortality of septic rats with the tested interventions. The sample size for the experiment of parameters investigations was determined via power analysis with a power of 80%, α level of 0.05, using values of preliminary experiments with minimum number of rats. The experiments were carried out according to the directive 2010/63/EU of the European Parliament and of the Council on the protection of animals used for scientific purposes, and to the guidelines set forth by the Ethics Committee of Patras University Hospital, Patras, Greece (No: 261070/1336-14/10/2014).

Sepsis animal model and therapeutic interventions

For sepsis induction, a well-established model of polymicrobial sepsis induced by cecal ligation and puncture was used as previously described ⁽⁷⁾. Briefly, rats were anesthetized via intramuscular injection of ketamine (80 mg/kg) and xylazine (10 mg/kg). After skin preparation and disinfection, a 2 cm incision was made through the midline and 75% of the cecum was ligated (the ligation point was in the 3/4 of the distance between the distal pole and the basis of the cecum). After cecal ligation, a perforation was made from the mesenteric to the antimesenteric wall with an 18G needle, and a small amount of feces was extracted from both penetration holes. The cecum was then relocated in the peritoneal cavity and the abdominal incision was closed in two layers with chromic 4-0 cat gut and 4-0 silk. The model of polymicrobial sepsis used in our experiments is highly lethal with 100% mortality in 7 days ⁽⁷⁾. The sham group was treated in an identical manner, with mobilization of the cecum but without ligation and puncture. Immediately after surgery, the rats were resuscitated by subcutaneous injection of saline solution 0.9% (37°C, 5 mL/100 g body weight), received postoperative analgesia with buprenorphin (0.05 mg per kg body weight s.c., which was repeated every 6 h for the first 24 h) and returned to their cages, with free access to water and food. Six hours after induction of sepsis the animals received therapeutic interventions, by intraperitoneal administration of hydrocortisone 2.8 mg/kg (Solu-Cortef; Pfizer, Athens, Greece) in a final volume of 0.5 mL saline solution 0.9% (group III) or by using oral gavage for fecal solution (Fecal Microbiota Transplantation) (group IV). The dose of hydrocortisone was selected to simulate the indicated daily dose for non-responding septic shock that is 200 mg for an average man of 70 kg (e.g., 200 mg/70 kg = 2.8 mg/kg). For simulation purposes, half of the animals in groups I (sham) and II (CLP) received an intraperitoneal injection of 0.5 mL of saline solution 0.9% and half were administered a saline solution at the corresponding volume (0.65 mL) used for FMT by oral gavage. Our previous pilot studies showed that the route of saline administration used for simulation does not affect the results. All experimental animals were re-operated under sterile conditions 24 h after the CLP (groups II, III, and IV) or sham procedure (group I) and samples were obtained according to the experimental protocol. At the end of the experiments, the rats were euthanized by decapitation.

Fecal microbiota transplantation

We followed the general steps for the preparation of fresh feces according to the European Consensus Conference for FMT in clinical practice ⁽⁵⁾. As donors for FMT, we used healthy male Wistar rats (with body weight 250 g–320 g), whose diet consisted of standard laboratory diet and tap water. Feces from individual rats were pooled to increase diversity and sample volume. The fecal material was used within 6 h after evacuation. The guidelines recommend 30 g of donor feces per person and transplantation to be diluted with sterile saline with three to five times larger volume of solvent (e.g., 30 g of feces diluted in 150 mL of saline). Assuming an average human weight of 70 kg, this is equivalent to 0.43 g feces per kg bodyweight. Extrapolating these guidelines in our experimental setting, with an average animal weight of 300 g, this is equivalent to 0.13 g of feces diluted in 0.65 mL of saline. This fecal material preparation has been previously used in experimental rat models ⁽⁸⁾. Our fecal collection yielded approximately 25 g of fecal material per pool, which was homogenized in 125 mL saline 0.9% using a standard commercial blender. Rats of group IV were administered 0.65 mL fecal suspension per transplantation (consisting of 0.13 g pooled feces) using an oral gavage feeding tube for rats, with the animals kept in an upright position. The animals were observed for the occurrence of acute complications related to the procedure.

Histologic evaluation

For histologic examination, tissue samples from the terminal ileum were obtained from all animals. The ileal samples were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 4 μm, and stained with hematoxylin and eosin (H&E). In each ileal specimen, several histologic features were evaluated and recorded. These features included indices of intestinal injury (architectural distortion, villous blunting, inflammation of the lamina propria), crypt epithelial mitotic and apoptotic activity, counting of Paneth cells (PCs) and morphometric analyses of the number of villi per centimeter (V/cm) and the total mucosal thickness in micrometers (μm). Two separate observers (DB, VZ) who were unaware of the group assignments performed all histological assessments.

Intestinal injury

All evaluated indices of intestinal injury were scored on a three-point scale from 0 up to 2, based on previous studies ⁽⁹⁾. Architectural distortion included epithelial changes, irregular crypts, crypt loss, and granulation tissue formation and scoring was graded as 0: normal, 1: mild distortion, and 2: moderate to severe distortion. Villous blunting was graded as 0: normal, 1: mild (villous-to-crypt-length ratio of 2:1), and 2: moderate/severe (villous-to-crypt-length ratio of 1:1 to total villous atrophy). Inflammation of the lamina propria was graded as 0: normal inflammatory infiltration of small bowel, 1: mild (mainly focal inflammatory infiltration), 2: moderate to severe (multifocal expanded mixed inflammatory infiltration). At least 20 well-preserved villi were estimated in each sample.

Counting of paneth cells

H&E stain allows the identification of PCs based on their distinctive granule staining pattern. For PC counts, in each sample 20 crypts were evaluated by conventional light microscopy at a constant magnification (×400) and expressed as the number of PCs per crypt.

Morphometric analyses

Ileal mucosa morphometric characteristics were studied by measurements of villus density, defined as the number of V/cm, and villus height in μm. Villus height was measured with a micrometer eyepiece affixed to an Olympus light microscope and at least 20 well-preserved villi were estimated in each sample.

Crypt mitotic count and apoptotic body count

The number of mitoses and apoptotic bodies per 10 crypts were also counted. Apoptotic bodies of the cryptal epithelium were identified and tallied using a morphometric analysis, which has been described in detail elsewhere ⁽¹⁰⁾. Apoptotic bodies were defined as rounded vacuoles with fragments of karyorrhectic nuclear debris and were differentiated from small isolated fragments of nuclear chromatin and intraepithelial neutrophils. Apoptotic bodies and mitoses were counted in 30 architecturally successive crypts included in the specimen, regardless of crypt orientation, and their total number was divided by three expressing respectively the number of apoptotic bodies per 10 crypts (referred to as apoptotic body count) or the number of mitoses per 10 crypts (referred to as mitotic count). For each case the ratio of the mitotic count to the apoptotic body count is referred to as the mitotic/apoptotic index.

Immunohistochemistry

Immunohistochemistry was performed on 4-mm-thick formalin-fixed, paraffin-embedded tissue sections mounted on gelatin-coated glass slides. Deparaffination, rehydration, and antigen retrieval were performed in an electric pressure cooker using Trilogy retrieval solution (Cell Marque, Ark) for 30 min. Polyclonal antibodies against occludin (1:80, Zymed Laboratories, San Francisco, Calif), claudin-1 (1:100, Zymed Laboratories, San Francisco, Calif), CD3 (1:100, Dako, Calif), and cleaved (activated) caspase-3 (1:200, Cell Signaling Technology, Danvers, Mass) were used for the primary reaction. The sections were incubated with primary antibodies for 1 h at room temperature, followed by 30 min incubation with Dako EnVision Labelled Polymer (Dako, Calif). Diaminobenzidine (Dako) was used as the chromogen. Nuclei were counterstained with Harris hematoxylin. Occludin, claudin-1, CD3, and cleaved caspase-3 immunohistochemical expression was recorded as present (+) or absent (−). For each case a percentage value of occludin(+) and claudin-1(+) enterocytes was obtained by dividing the number of cells staining positive by the total number of enterocytes lining villi. Also, the number of CD3(+) intraepithelial lymphocytes per 100 intestinal epithelial cells was recorded. Ten well-oriented villi were randomly selected per case, to evaluate occludin, claudin-1 expression, and CD3+ T-cell infiltration. Regarding cleaved caspase-3, its immunohistochemical expression was evaluated in intestinal crypts; caspase-3(+) enterocytes were counted in 30 architecturally successive crypts included in the specimen, regardless of crypt orientation, and their total number was divided by three expressing the number of apoptotic epithelial cells per 10 crypts. Two separate observers (DB, VZ) who were unaware of the group assignments performed all immunohistochemical analyses and when scores between the two observers were discordant, a consensus was achieved by conference at a two-headed microscope.

Serum endotoxin, IL-6, and IL-10 concentrations

For the determination of endotoxin, IL-6, and IL-10 concentrations, the abdominal aorta was punctured and samples 2 mL of blood, were obtained. Endotoxin, IL-6, and IL-10 levels were measured in serum by enzyme-linked immunosorbent assays using commercially available kits, as per the manufacturer's instructions [endotoxin (detection limit 0.005 EU/mL): Abbexa Ltd, Cambridge Science Park, Cambridge, UK; IL-6 (detection limit 21 pg/mL), and IL-10 (detection limit 10 pg/mL): R&D Systems Europe, Abingdon Science Park, Abingdon, UK].

Survival study

The survival studies were conducted with 10 animals in each study group: sham, CLP, CLP+HC, CLP+FMT. Experimental animals were treated as described above and were observed every 12 h after the operation for 7 d and survival was recorded. Hydrocortisone and FMT were administered daily in surviving rats of groups III and IV respectively, starting 6 h post-sepsis induction by CLP. Rats of groups I and II were randomly divided to receive either daily intraperitoneal injections of 0.5 mL of saline solution 0.9% (hydrocortisone simulation) or 0.65 mL of saline solution by oral gavage (FMT simulation). At the end of the day 7 the study ended, and the surviving animals were euthanized by decapitation. Survival was evaluated using Kaplan–Meier survival curves.

Statistical analyses

Data were analyzed using the SPSS statistical package for Windows (version 24.0 SPSS Inc, Chicago, Ill). Results are expressed as mean (SD). Normality of data was tested using the Shapiro–Wilk test. Morphometric parameters were normally distributed with equal variances across groups (Levene test); therefore, comparisons were performed using the one-way ANOVA, followed by Bonferroni post-hoc test. All other parameters studied were not normally distributed, therefore comparisons were performed using the non-parametric analysis of variance (Kruskal–Wallis test) followed by a post-hoc Mann–Whitney U test. Correlations were estimated by a non-parametric Spearman correlation test. All tests were two-tailed and a P value of less than 0.05 was considered significant.

RESULTS

Intestinal histology

The ileal specimens from septic rats of group II demonstrated significant intestinal injury as compared with sham operated (Table 1), consisted of architectural distortion (P < 0.05), villous blunting (P < 0.001), infiltration of the lamina propria by inflammatory cells (P < 0.001), and a trend toward reduction of PCs. Administration of hydrocortisone (group III) or FMT (group IV) significantly reduced architectural distortion (P < 0.01), villous blunting (P < 0.001 and P < 0.01, respectively), inflammatory infiltration of lamina propria (P < 0.001), whereas it increased the number of PCs (P < 0.05 and P < 0.01, respectively).

Table 1 - Indices of intestinal injury and Paneth cell count

	SHAM	CLP	CLP + HC	CLP + FMT
Arch distortion	1.1 ± 0.31∗	1.6 ± 0.51	1§	1§
Villous blunt	0†	1.4 ± 0.51	0||	0.40 ± 0.51§
Inflammation lamina propria	1†	1.7 ± 0.25	1.1 ± 0.21||	1||
Paneth cells	1.45 ± 0.8	1 ± 0.66	1.9 ± 0.87‡	1.95 ± 0.43§

^∗P < 0.05 versus CLP.

^†P < 0.001 versus CLP.

^‡P < 0.05 versus CLP.

^§P < 0.01 versus CLP.

^||P < 0.001 versus CLP.CLP indicates cecal ligation and puncture; FMT, fecal microbiota transplantation; HC, hydrocortisone.

Morphometric analysis

Compared with sham, the septic rats of group II presented intestinal mucosal atrophy with significantly reduced villous density (P < 0.001) and mucosal thickness (P < 0.05), which was reversed after administration of hydrocortisone (P < 0.001 for both parameters) or FMT (P < 0.001 for villous density and P < 0.001 for mucosal thickness) (Figs. 1 and 2).

Fig. 1: Morphometric characteristics of ileal mucosa: villous density and mucosal thickness.

Fig. 2: CLP induced a significant reduction of villous density and mucosal thickness, which were reversed after HC administration or FMT (left panel, H&E). In the right panel, typical mitoses in crypt epithelial cells are indicated by white arrows, apoptotic bodies by black arrows, and cleaved caspase-3(+) epithelial cells by red arrows. CLP indicates cecal ligation and puncture; FMT, fecal microbiota transplantation; H&E, hematoxylin and eosin; HC, hydrocortisone.

Intestinal crypt mitotic and apoptotic activity

Crypt mitotic activity was estimated in H&E sections (mitotic count) and crypt apoptotic activity was estimated by determination of apoptotic body count in H&E sections and by immunohistochemical expression of cleaved caspase-3. In CLP rats of group II, the crypt epithelial apoptosis as estimated by both the methods was significantly increased (P < 0.001 vs. group I, respectively) (Fig. 3). Administration of hydrocortisone significantly reduced apoptotic body count (P < 0.05 vs. group II) and immunohistochemical expression of cleaved caspase-3 (P < 0.01 vs. group II). FMT exerted also a significant antiapoptotic effect by reducing apoptotic body count and cleaved caspase-3 expression in intestinal crypts (P < 0.001 vs. group II, respectively). The ratio of the mitotic count/apoptotic body count (mitotic/apoptotic index) was significantly reduced in CLP rats of group II (P < 0.001 vs. group I) and increased after HC administration (P < 0.05 vs. group II) or FMT (P < 0.001 vs. group II) (Fig. 3). Typical mitotic figures and apoptotic bodies in H&E sections and immunohistochemical staining results for cleaved caspase 3(+) enterocytes in intestinal crypts of CLP rats (group II) are shown in Figure 2.

Fig. 3: Crypt epithelial mitotic and apoptotic activity: mitotic count, apoptotic body count, mitotic/apoptotic index, and cleaved caspase-3 expression.

Immunohistochemical results for occludin, claudin-1, and intraepithelial CD3(+) T-lymphocytes

In the CLP group II, the expression of the tight junction (TJ) proteins occludin and claudin-1 in intestinal epithelial cells was significantly decreased (P < 0.001 vs. group I). Administration of HC (group III) or FMT (group IV) significantly increased occludin (P < 0.05 and P < 0.01, respectively) and claudin-1 expression (P < 0.001, respectively) compared with group II (Fig. 4). Regarding intraepithelial CD3(+) T-lymphocytes, a significant increase was demonstrated in group II (P < 0.001 vs. group I), falling to control levels in groups III and IV (P < 0.001 vs. group II, respectively) (Fig. 4). Representative microphotographs of immunohistochemical stain for occludin, claudin-1, and CD3 in all experimental groups are shown in Figure 5.

Fig. 4: Semiquantitative immunohistochemical results of enterocytes’ occludin and claudin-1 expression and intraepithelial CD3(+) T-lymphocyte count.

Fig. 5: Representative photomicrographs of immunohistochemical stain for CD3, occludin, and claudin-1 in experimental groups: CLP increased CD3(+) intraepithelial T-lymphocytes and significantly depleted occludin expression from intestinal epithelial cells (green arrows) and decreased claudin-1 expression mainly in the middle and lower third of the villi. HC administration and FMT decreased mucosal infiltration by CD3(+) T-lymphocytes and significantly restored occludin and claudin-1 expression in numerous enterocytes. Black arrows indicate restored occludin expression at the apical part of enterocytes and red arrows at the lateral membrane. CLP indicates cecal ligation and puncture; FMT, fecal microbiota transplantation; HC, hydrocortisone.

Systemic endotoxin, IL-6, and IL-10 concentrations

Endotoxin levels in systemic circulation were significantly increased in septic CLP rats (P < 0.01 vs. group I). Administration of hydrocortisone or FMT significantly reduced systemic endotoxemia (P < 0.05 vs. group II, respectively) (Fig. 6). Levels of the proinflammatory cytokine IL-6 and the anti-inflammatory cytokine IL-10 were both significantly increased in septic rats of group II compared with sham (P < 0.05, respectively). Administration of hydrocortisone led to a further increase of both cytokines (P < 0.01, as compared with group II, respectively). In contrast, FMT induced a significant decrease of both IL-6 and IL-10 to control levels (P < 0.01 vs. group II, respectively) (Fig. 6).

Fig. 6: Serum levels of endotoxin, IL-6, and IL-10.

Correlations

The intestinal barrier integrity indices of occludin expression, apoptotic body count, and mitotic/apoptotic index were significantly correlated with systemic endotoxemia, IL-6 and IL-10 levels (supplementary Figure, https://links.lww.com/SHK/B53). Specifically, occludin expression negatively correlated with endotoxin (P < 0.001), IL-6 (P < 0.05), and IL-10 (P < 0.05) concentrations, apoptotic body count positively correlated with endotoxin (P < 0.001), IL-6 (P < 0.05), and IL-10 (P < 0.05) and the mitotic/apoptotic index negatively correlated with endotoxin, IL-6, and IL-10 levels (P < 0.05, respectively). Serum levels of IL-6 and IL-10 did not correlate to systemic endotoxemia (P = 0.087, respectively).

Survival

All the animals in the CLP group died within the 7-day period (50% died at 24 h), while 7-day survival rates in groups III and IV were 50% respectively (P < 0.001) (Fig. 7).

Fig. 7: Kaplan–Meier survival curves for experimental groups.

DISCUSSION

Sepsis is defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection, with a mortality rate of above 20% ⁽¹¹⁾. In the present experimental study, it was demonstrated that HC administration and FMT led to 50% reduction of mortality in septic rats. Current pathophysiological theories highlight the pivotal role of the gut not only as reservoir of pathogens, but as a proinflammatory organ per se that drives the systemic inflammatory response associated with multiple organ dysfunction syndrome ⁽²⁾. The gut barrier function is comprised by three major lines of defence: the biological barrier (gut microbiota), the immune barrier, and the mechanical barrier (epithelial cells and their interconnections) ^{(12, 13)}. The present study was designed to investigate the effects of the tested therapeutic interventions on several parameters of gut barrier integrity in sepsis.

The role of corticosteroids in sepsis is being investigated in an attempt to effectively suppress hyper-inflammation. Recent research efforts have focused on the role of stress doses of hydrocortisone based on the findings of endogenous steroids insufficiency in critical illness. Two large-scale randomized controlled trials on hydrocortisone in patients with septic shock (Adjunctive Glucocorticoid Therapy in Patients with Septic Shock and Activated Protein C and Corticosteroids for Human Septic Shock), demonstrated beneficial effects on morbidity with an excellent safety profile, with the Activated Protein C and Corticosteroids for Human Septic Shock study showing a survival benefit ^{(4, 14)}. However, the role of HC administration on the critical parameter of gut barrier function has not been investigated until now. Our study demonstrated that stress doses of HC in septic rats reverse intestinal mucosal atrophy by increasing villous density and mucosal thickness. Improvement of mucosal atrophy may be explained by a significant reduction of crypt epithelial apoptosis and increase of the mitotic/apoptotic index. The balance between cell proliferation and death in intestinal crypts is crucial for epithelial homeostasis because the total of epithelial cells lining villi originate from stem cells located in the proliferation zone of the crypt. Improvement of the mitotic/apoptotic index indicates an improved regenerative capacity of the intestinal epithelium to injurious insults ⁽¹⁵⁾. Previous mechanistic studies have shown that glucocorticoids prevent intestinal apoptosis through control of intestinal Signal transducer and activator of transcription 1 and tumor necrosis factor-induced inflammation ⁽¹⁶⁾. In the present study, HC also ameliorated other parameters of intestinal injury namely architectural distortion and villous blunting. The protective effect of HC against mucosal injury has been previously demonstrated in intestinal ischemia-reperfusion ⁽¹⁷⁾. Most importantly, the intestinal mechanical barrier was strengthened by HC administration through increased enterocytes’ occludin and claudin-1 expression, thus restricting the passage of intraluminal pathogens and/or their products (endotoxin) through the paracellular space. Previous in vitro studies with intestinal epithelial cells have shown that HC increases the expression of genes responsible for TJ formation ⁽¹⁸⁾. Regarding the investigated parameters of the intestinal immunological barrier, HC increased the mucosal count of PCs, which are considered key effectors in the host immune response to enteric pathogens by secreting a broad spectrum of antimicrobial peptides. In addition, HC decreased the intraepithelial CD3(+) T lymphocytes and the inflammatory infiltration of lamina propria, probably as a result of decreased bacterial translocation ⁽¹⁹⁾. This multifactorial improvement of gut barrier function resulted in significant prevention of systemic endotoxemia in septic rats. Despite prevention of systemic endotoxemia, the proinflammatory cytokine IL-6 and the anti-inflammatory IL-10 were both increased. An HC-induced promotion of IL-10 secretion has been previously described in patients with severe sepsis, accompanied by reduction of the pro-inflammatory cytokine IL-6 ⁽²⁰⁾. The increase in IL-6 levels observed in our study indicates a compensatory increase of IL-6 to IL-10 enhancement by HC. Alternatively, the low dose of HC (stress dose) administered once in our study (6 h post-CLP) before cytokine measurement (24 h post-CLP) may not be adequate to control an initial compensatory increase of IL-6, which may decline later.

Previous studies have shown that the intestinal microflora equilibrium is disrupted in sepsis and microbiota represents another important therapeutic target in this patient population ^{(21, 22)}. Microbiota displays important metabolic, immunologic, and gut protective functions. Metabolically, microbiota ferments carbohydrates, and indigestible oligosaccharides and synthesizes short chain fatty acids such as butyrate, propionate, and acetate, which are rich sources of energy for the intestinal epithelium ⁽²³⁾. Immunologically, the gut microbiota contribute to gut immunomodulation interacting with both the innate and adaptive immune systems, through production of pathogen-associated molecular patterns, which are recognized by specific receptors of intestinal immune cells ⁽²⁴⁾. In addition, intestinal microbiota prevents growth of potentially pathogenic bacteria through antagonism for nutrients and exerting colonization resistance. Through all these functions, gut microbiota is in a continuous cross talk with the intestinal epithelium and exerts pivotal role in preserving the integrity of the intestinal epithelium. It has been shown that gut microbiota affects intestinal epithelial cell turnover, apoptosis, and the expression and function of TJ ⁽²⁵⁾. Previous attempts to reverse microbiota alterations in sepsis by selective digestive decontamination or administration of probiotics, prebiotics and synbiotics, reported negative or contradictory results regarding mortality reduction ⁽²⁾. FMT has been established as a highly effective therapy for recurrent Clostridium difficile infection and is under investigation for several other intestinal and extra-intestinal diseases, such as inflammatory bowel diseases, irritable bowel syndrome, obesity, nonalcoholic hepatic steatosis, and diabetes ^{(26, 27)}. To the best of our knowledge, this is the first study investigating the effect of FMT in sepsis. According to our results, FMT provided a survival benefit in septic rats reducing mortality by 50%. This effect was associated with an improvement of multiple levels of gut barrier integrity: First, FMT prevented intestinal injury and mucosal atrophy, which may be attributed to a reduction in crypt epithelial apoptosis with concomitant increase of the epithelial regenerative capacity, as expressed by the crypt mitotic/apoptotic index. The antiapoptotic and trophic effect of FMT on the intestinal mucosa has been previously demonstrated in chemotherapy-induced mucositis and was mediated through inhibition of nuclear factor kappa-light-chain-enhancer of activated B cells activity ⁽²⁸⁾. Second, in our study FMT positively affected the integrity of the paracellular barrier by preventing the sepsis-induced significant decrease of occludin and claudin-1 expression. This effect may be attributed to the beneficial effects on multiple metabolic pathways, antioxidant action, anti-inflammatory action, and inhibition of nuclear factor kappa-light-chain-enhancer of activated B cells ⁽²⁹⁾. An additional mechanism may be the reduction of systemic endotoxin and IL-6 levels observed in our study, because these stimuli have been shown to negatively regulate occludin expression and function in the intestinal tissue ^{(12, 30)}. Third, FMT beneficially affected the immunological intestinal barrier by increasing the number of PCs and decreasing the submucosal inflammation in the lamina propria and intraepithelial CD3(+) T-lymphocytes. Microbiota exerts a pivotal role in orchestrating the innate and adaptive immune response of the intestinal immune system through secretion of diverse substances like short chain fatty acids, fragelin, lipopolysaccharide, and sphingolipids. These substances modulate B-cell switch to IgA producing plasma cells, induce T-cell differentiation to T-helper 17 cells, and promote T regulatory cell proliferation and differentiation ⁽²⁴⁾. The global preventive effect of FMT on the sepsis-induced gut barrier dysfunction, resulted in significant prevention of systemic endotoxemia and cytokinemia. In correlation analyses, intestinal barrier indices of occludin expression, crypt epithelial apoptosis, and mitotic/apoptotic index significantly correlated with systemic endotoxemia, IL-6, and IL-10 levels, whereas IL-6 and IL-10 levels did not correlate to systemic endotoxemia. Taken together, our findings indicate that the lowering effect of FMT on cytokine levels might by attributed, at least partly, to inhibition of cytokine production by the intestine through improvement of its barrier function that attenuates its role as a proinflammatory organ in sepsis ⁽²⁸⁾.

The present experimental study presents some limitations: First, we did not perform an analysis of microbiota changes. The rationale for FMT utilization in sepsis was the reversal of intestinal microbiota dysbiosis, and FMT was proven effective in restoring gut barrier function in septic rats. However, the effect of this therapeutic intervention on the intestinal microflora alterations was not investigated. Second, we used a highly lethal model of sepsis with 100% mortality in 7 days, to investigate the potential beneficial effects of the tested interventions. Our findings could not be readily extrapolated in less severe grades of sepsis. Third, investigated parameters of gut barrier function and inflammatory response were evaluated only at one time point (24 h post CLP); therefore, this study does not provide information on their longitudinal dynamics which have important pathophysiological value. Fourth, considering the positive effect of both tested interventions on gut barrier function and mortality of septic rats, a combination of these treatments could potentially exert a synergistic beneficial effect, but this was not investigated in the present study.

In conclusion, our results show that FMT and hydrocortisone administration decrease mortality in sepsis. This positive effect is associated with a multifactorial improvement of the gut mechanical and immunological barriers leading to the prevention of systemic endotoxemia. FMT additionally normalized sepsis-induced cytokinemia. Considering the high rates of mortality in sepsis these encouraging findings should be tested further in clinical studies.