Effect of Shenfu injection on intestinal mucosal barrier in a rat model of sepsis

a b s t r a c t

Purpose: The effects of Shenfu injection on protecting the Intestinal mucosal barrier were investigated in rats with sepsis.

Methods: Severe sepsis was established by cecal ligation and puncture (CLP) in 30 healthy Sprague-Dawley rats. Twelve rats that received sham surgery received 10 mL/kg of normal saline. Rats with CLP were randomized to receive 10 mL/kg of normal saline (n = 12) and 5 mL/kg Shenfu (n = 12), and 10 received 10 mL/kg Shenfu injection (n = 12) by tail intravenous injection. Rats were killed after 8 hours. Serum levels of tumor necrosis factor ? and interleukin-10, and ileal malondialdehyde and superoxide dismutase activity were measured by enzyme-linked immunosorbent assay. Ileum tissue structures and pathological score were observed by microscopy. Ileal mucosal epithelial cell apoptosis index was calculated by TUNEL assay. Ileal proapoptotic protein Bax, antiapoptotic protein Bcl-2, and tight junction transmembrane protein occludin were measured by immunohistochemistry and immunoblot.

Results: The level of tumor necrosis factor ?, the ileal Malondialdehyde level, ileum pathological score, apoptosis index of ileal mucosal epithelial cells, and Bax protein level were significantly higher, and serum level of interleukin-10, the ileal superoxide dismutase activity, Bcl-2 protein level, Bcl-2/Bax ratio, and occludin protein level were significantly lower in the CLP group than in the Sham group (P b .01 or P b .05). Both low- and high- dose Shenfu significantly ameliorated these changes (P b .01 or P b .05), but high-dose injection achieved more significant improvements than did the low-dose injection (P b .01 or P b .05).

Conclusions: Shenfu injection might ameliorate the mucosal barrier function in a model of sepsis in rats in a dose- dependent manner.

Introduction

Severe sepsis is the most common serious complications leading to septic shock, multiple-organ dysfunction syndrome (MODS), multiple- organ failure, and even death. Currently, treatment of severe sepsis involves fluid resuscitation, administration of vasoactive agents, appropriate antibiotics, ventilatory support, and blood purification; however, the mortality rate from severe sepsis remains as high as 27% to 54.1% [1]. During treatment, the primary focus of infection cannot be identified in roughly one-third of patients, and these patents do not

? Source of support: This study was supported by the Zhejiang province’s Major Disease Prevention and Control of Traditional Chinese Medicine Research Program (2012ZGG001).

?? Conflict of interest: The authors declare that they have no conflict of interest.

* Corresponding author at: ICU, The First Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou 310006, Zhejiang Province, China. Tel.: +86 13 958095268;

fax: +86 21 64085875.

E-mail address: [email protected] (R. Jiang).

¹ These authors contributed equally to this study (co-first author).

respond to antibiotics, leading to the development of MODS [2]. Since 1987, when Border et al [3] reported an association between Bacterial translocation and gut-derived sepsis, many studies have reported that the gut plays important roles in the development of systemic inflamma- tory response syndrome, sepsis, and MODS, and thus, research into the intestinal barrier has received increasing attention.

The mechanical barrier presented by the intestinal mucosal epithe- lium plays a crucial role in preventing the translocation of intestinal bacteria and endotoxins [4]. During severe sepsis, acute injury and dys- function of the intestine can impair this mechanical barrier [5]. Edema of the intestinal mucosal epithelium, increased apoptosis of epithelial cells, shortened tight-junction bands, increased permeability of the intercellular space, atrophy of intestinal mucosa, intestinal motility disorder, and intestinal microcirculation disturbance all contribute to reduce barrier function. Translocation of intestinal flora most often occurs at the mesenteric lymph nodes, liver, and spleen, leading to enterogenous endotoxemia and bacteremia [4].

Shenfu is a traditional Chinese medicine containing extracts of red ginseng (Panax), aconite root (Radix aconiti lateralis preparata), and black monkshood (Aconitum), and when administered clinically, it has

http://dx.doi.org/10.1016/j.ajem.2015.01.056

been reported to restore Yang from collapse, tonify Qi, and relieve de- sertion [6]. Shenfu injection is mainly used for the treatment of infec- tious shock and hemorrhagic shock [7-9]. Adverse reactions associated with Shenfu injection use include occasional tachycardia, al- lergic reactions, Hepatic dysfunction, and Urinary retention [10]. The mechanism of Shenfu action has been reported to involve restoration of hemodynamic stability, increased tissue oxygen partial pressure and oxygen content, improved microcirculation, and tissue metabolism, and thus, Shenfu injection promotes resuscitation from shock [7]. Shenfu injection has also been reported to reduce the serum concentra- tion of Inflammatory mediators, such as tumor necrosis factor (TNF) ? with sepsis, reducing the incidence of MODS, which is otherwise often caused by excessive inflammatory responses [8]. Shenfu has also been demonstrated to improve tissue function and hemodynamic status of heart failure and to exert strong antiendotoxin, anti-inflammatory ef- fects, and acts as an effective oxygen free-radical scavenger [9,11,12]. In our previous study, Shenfu injection was found to quickly improve the hemodynamic status and the abnormal Oxygen metabolism in tis- sues [12]. However, the effect of Shenfu on intestinal mucosal barrier re- mains largely unknown. In this study, a model of severe sepsis was established in rats by cecal ligation and puncture (CLP) [13]. After intra- venous administration of Shenfu, the status of intestinal epithelial cells was assessed in order to assess whether Shenfu protects the intestinal mucosal barrier. We further characterized a variety of physiological and pathological processes of sepsis and recorded marked improve- ments in animals administered Shenfu.

Methods

Animals

All studies were performed with the approval of the animal use and care committee of our hospital. Healthy Sprague-Dawley rats (24 male and 24 female) weighing 200 +- 20 g were provided by the experimen- tal animal center of Zhejiang Chinese Medical University, China (certifi- cation number SCXK [Hu] 2007-000). All rats housed at the experimental animal center of Zhejiang Chinese Medical University (certification number SYXK [Zhe] 2003-0003) at 20?C +- 1?C, 50% to 60% humidity, under a 12:12-hour light/dark cycle, with a ventilation rate of 8 to 15 times/h. Using a randomized complete block design, rats were divided randomly into 4 groups of 12: sham operation group (sham), severe sepsis group (CLP), low-dose Shenfu injection group (LSF), and high-dose Shenfu injection group (HSF).

Animal model establishment and intervention

Severe sepsis was induced by CLP, as previously described [13]. Briefly, all rats were deprived of food, but had free access to water for 12 hours prior to surgery. Anesthesia was induced by intramuscular in- jection of 5% ketamine (0.2 mL/100 g body weight), and rats were fixed on the operating panel. After abdominal hair removal and skin disinfec- tion, a ventral Midline incision (1.5 cm) was made. For the rats in the sham group, the abdominal cavity was opened and closed immediately, but was not ligated or perforated. For the rats in the other 3 groups, the abdominal cavity was opened and the mesentery of the cecum was carefully dissected, and then the root of the cecum was ligated avoiding damage to the ileum and the mesenteric vessels. Punctures were made through the cecum at 3 locations using a 21-gauge needle. The cecum was gently compressed until faces was extruded; the bowel was then returned to the abdomen and the incision was closed. All rats were re- suscitated immediately after surgery by subcutaneous administration of normal saline (5 mL/100 g body weight) [13]. Twelve minutes later, rats in the sham group and CLP group received 10 mL/kg of normal sa- line. Rats in the LSF group received 5 mL/kg Shenfu plus 5 mL/kg of nor- mal saline, whereas HSF-treated animals were administered 10 mL/kg Shenfu. Animal treatment was carried by tail vein injection. Rats were

then housed at 22?C with free access to food and water and were ob- served until they recovered from anesthesia, then every 2 hours until 8 hours after surgery.

Specimen collection and Bacterial cultures

Eight hours after surgery, all live rats were anesthetized with 5% ke- tamine. The heart rate (HR) was measured as follows: needle electrodes were inserted into both upper extremities and Left lower extremity and connected to the RM-6280 Physiologic Recording System (Xingwan, Guangdong, China). Rectal temperature (T) was monitored by ther- mometer; heart blood samples were taken as follows: hair was removed and skin was disinfected in the area under the xiphoid, then 2 mL of blood was removed from the heart by syringe. The blood was trans- ferred to a culture bottle, then the abdominal cavity was opened and swabbed with a sterile swab stick to collect the peritoneal exudates. Bacterial cultures of the peritoneal swabs and blood culture specimens were made at the Clinical Pathogenic Microbiological Laboratory of the First Affiliated Hospital of Zhejiang Chinese Medicine.

Serology

Before administration of anesthesia, 2 mL blood sample was taken from the abdominal aorta of all animals and white blood cell (WBC), serum alanine aminotransferase (ALT), aspartate aminotransferase , Blood urea nitrogen , creatinine (Cr), and creatine kinase isoenzyme-MB (CK-MB) levels or activity were measured. Serum TNF-? and interleukin (IL)-10 were measured using enzyme-linked im- munosorbent assay (Shanghai Bioleaf Biotech Co Ltd, Shanghai, China), according to the manufacturer’s instructions.

Ileal tissue inflammation score

A 2-cm section of terminal ileum was also taken for further analysis. Ileal tissues were sliced, hematoxylin and eosin stained and observed by microscopy (Carl Zeiss AG, Oberkochen, Germany). Tissue structures were observed by experienced pathologist, and Chiu’s scores were cal- culated as previously described [14]. The detailed Chiu’s pathological scoring criteria are summarized in Supplementary Table 1.

Ileal tissue malondialdehyde and superoxide dismutase measurement

Tissue samples of 200 mg were homogenized in a 0.9% NaCl solution (NaCl/liver tissue, 9:1, vol/vol) at 4?C. Tissue homogenates were centri- fuged for 10 minutes at 1300g, and the clear upper supernatants were collected for Malondialdehyde and Superoxide dismutase analysis. Both MDA and SOD were determined by the appropriate enzyme-linked immunosorbent assay (Shanghai Bioleaf Biotech Co, Ltd) using a spectrophotometer.

Apoptosis detection of Ileal mucosal cells

The apoptosis index (AI) of the myocardium was assessed by the TUNEL method using a TUNEL staining kit (Maxim, Beijing, China), ac- cording to the manufacturer’s instructions. Stained cells were counted in six random visual fields. Apoptosis index was defined as the number of stained cells per 100 cells.

Immunohistochemistry

We performed immunohistochemistry on ileal tissues to determine protein expression levels of both Bax and Bcl-2. The following primary antibodies were used: (1:2000; Abcam, Cambridge, UK). The secondary antibody used was a biotinylated goat antirabbit IgG (Beyotime Institute of Biotechnology, Shanghai, China). Sections were visualized with di- aminobenzidine (Shanghai Jierdun Biotech Co Ltd, Shanghai, China)

Table 1

Comparison of Biochemical parameters in rats 8 hours after surgery (x +- SD)

Group	HR (beats/min)	T (?C)	WBC (x 10⁹)	ALT (U/L)	AST (U/L)	BUN (mmol/L)	Cr (umol/L)	CK-MB (U/L)
Sham (n = 12)	327.2 +- 35.6	37.3 +- 0.2	6.2 +- 1.5	34.9 +- 3.2	37.2 +- 1.5	7.3 +- 0.7	53.7 +- 4.5	19.1 +- 3.3
CLP (n = 11)	552.3 +- 27.4?	38.7 +- 0.2?	13.6 +- 1.1?	95.4 +- 2.5?	98.7 +- 3.2?	11.3 +- 1.3?	115.0 +- 11.5?	37.1 +- 2.1?
LSF (n = 12)	–	–	–	92.3 +- 3.4	99.5 +- 4.3	10.4 +- 1.3	113.9 +- 12.3	38.3 +- 2.6
HSF (n = 11)	–	–	–	94.2 +- 3.7	96.3 +- 3.5	11.3 +- 3.2	116.3 +- 11.6	37.3 +- 1.7

* P b .05 vs sham group.

and counterstained with hematoxylin and eosin. Immunohistochemical images were captured with a digital camera (Nikon, Tokyo, Japan) and analyzed using the IMS imaging processing system (Media Cybernetics, New York, NY). Regions positively stained for Bax and Bcl-2 were count- ed and analyzed in 10 randomly selected nonoverlapping fields of view, as a semiquantitative parameter to reflect Bax and Bcl-2 expression.

Immunoblot

Total protein was extracted from 20 mg ileal tissues and lysed in 200 ul RIPA lysis buffer. Protein Samples (40 ug) were subjected to 10 % sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Merck Millipore, Billerica, MA). Membranes were then blocked and stained with anti- occludin polyclonal antibody (1:500; Cell Signaling Technology, Danvers, MA). After washing, membranes were incubated with secondary antibodies (Cell Signaling Technology). Bands were visualized using enhanced chemiluminescence (Pierce, Rockford, IL). ss-Actin (1:500; Abcam) was used as the internal control and was detected on the same membrane. Bio-Rad Gel imaging system and Quantity One software (Her- cules, CA) were used to analyze the grayscale of bands.

Statistical analysis

Continuous data are expressed as mean +- SD, and comparison of the means among groups was performed using 1-way analysis of variance. Comparisons between groups were performed using t tests. Statistical analyses were performed using SPSS 17.0 (SPSS Inc, Chicago, IL). P values less than .05 were considered significant.

Results

rat model of sep”>Establishment of a rat model of sepsis

During the 8-hour postoperative observation period, no rats died in the sham group or LSF group. One rat died in the CLP group and 1 rat died in the HSF group, corresponding to Death rates of 0, 0, 8.33%, and protective effects of Shenfu in”>8.33%, respectively, which did not differ significantly between the groups. Rats in the sham group behaved normally, were sensitive, and had glossy hair, normal stool, and no hyperemia and edema in the intes- tine. Live rats in the CLP group behaved differently, with delayed re- sponses, poor appetite, loose stool, lusterless hair, and bloody ascites, and the abdominal cavity had a strong fishy odor, with hyperemia and edema in the intestine, and dark purple ligated cecum. Live rats in the

LSF and HSF groups did not noticeably differ. These animals behaved normally, with normal appetite, but shapeless stool, lackluster hair, and bloody ascites, and the abdominal cavity had a strong fishy odor, with hyperemia and edema in the intestine, and the distal end of the ligated cecum was dark purple.

Assessment of sepsis and organ dysfunction

Eight hours after surgery, the HR, T, and WBC counts and levels of serum ALT, AST, BUN, Cr, and CK-MB were significantly higher in the CLP group than in the sham group (P b .05; Table 1). Eight hours after surgery, bacterial cultures of blood and peritoneal exudates revealed no pathogenic bacteria in the sham group, whereas in the CLP group, blood culture was positive for Enterococcus faecalis, Escherichia coli, and Proteus mirabilis, and peritoneal exudates culture was positive for Bacillus proteus vulgaris, P mirabilis, E faecalis, and E coli (Table 1).

In comparison to the sham group, the level of serum TNF-? (P b .01) and ileal MDA (P b .01) were higher, and serum IL-10 (P b .05) and ileal SOD activity (P b .01) were lower in the CLP group (Table 2). In the sham group, the ileal mucosal villi were orderly arranged, the structure of surrounding blood vessels was normal, there was no obvious bleeding, the muscle fibers were orderly arranged, and the serosa was normal. In the CLP group, the ileal mucosal villi were damaged; swelling, ulcer- ation, and bleeding were observed around blood vessels; and fractured basal layer was observed. The Chiu’s pathological score of the ileum was significantly higher in the CLP group than in the sham group (P b .01; Table 3 and Fig. 1).

TUNEL assay revealed that only a few ileal mucosal epithelial cells were apoptotic in the sham group, whereas in the CLP group, extensive apoptosis was observed, occurring mainly in the mucosal column like epithelia and submucosal cells, and the AI in the CLP group was signifi- cantly higher than that in the sham group (P b .01; Table 3).

In ileum tissues from rats in the CLP group, expression of the antiapoptotic protein Bcl-2 was significantly lower (P b .05), expression of the proapoptotic protein Bax was significantly higher (P b .05; Fig. 2), and expression of tight junction protein occludin was significantly lower (P b .01) than in ileum tissues from rats in the sham group (Fig. 3).

The protective effects of Shenfu injection on the intestinal mucosal barrier of rats with sepsis

In comparison to the CLP group, the level of serum TNF-? and ileal MDA level was significantly lower in both the LSF group (P b .01 and P b .05, respectively) and the HSF group (P b .01), and the levels of

Table 2

Comparisons of TNF-? and IL-10 in serum, and MDA and SOD in ileal tissue (x +- SD)

Group	TNF-? (ug/mL)	IL-10 (ug/mL)	MDA (pg/L)	SOD (pg/L)
Sham (n = 12)	13.99 +- 2.85??	35.60 +- 7.50?	12.28 +- 1.21??	123.28 +- 11.39??
CLP (n = 11)	89.71 +- 14.05	28.27 +- 3.15	25.51 +- 1.29	62.94 +- 9.71
LSF (n = 12)	44.69 +- 4.80??	67.41 +- 6.81??	22.26 +- 1.61?	84.38 +- 1.61??
HSF (n = 11)	29.13 +- 4.69??,+	78.97 +- 3.46??,+	16.48 +- 2.39??,???	107.772.39??,+
* P b .05 vs CLP group. ?? P b .01 vs CLP group.

??? P b .05 vs LSF group.

⁺ P b .01 vs LSF group.

Table 3

Ileal Chiu’s scores and ileal mucosal epithelial cell AI (x +- SD)

Group	Chiu’s score	AI
Sham (n = 12)	0.3 +- 0.30??	5.39 +- 1.45??
CLP (n = 11)	3.8 +- 0.42	27.21 +- 1.55
LSF (n = 12)	3.3 +- 0.48?	18.20 +- 1.29??
HSF (n = 11)	2.5 +- 0.52??,???	11.68 +- 1.45??,???
* P b .05 vs CLP group. ?? P b .01 vs CLP group.

??? P b .01 vs LSF group.

serum IL-10 and ileal SOD activity were higher in both the LSF and HSF groups (P b .01). In the HSF group, IL-10 levels and ileal SOD activity were higher and TNF-? levels and ileal MDA level were more substantially lower than those in the LSP group (P b .01, P b .01, P b .01, and P b .05, respectively; Table 2). However, treatment with Shenfu injection resulted in a slight decrease of the sepsis-induced elevated WBC count observed in the CLP group (CLP: 13.4 +- 0.78 x 10⁹; LSF: 12.9 +- 0.45 x 10⁹; HSF: 12.4 +- 0.80 x 10⁹). Although a

significant reduction was observed with high Shenfu injection-dosed animals compared with the CLP group, a Complete recovery was not reached (HSF: 12.4 +- 0.80 x 10⁹; sham: 5.8 +- 0.38 x 10⁹).

In the LSF and HSF group rats, slight damage to the ileal mucosal villi, local necrosis, and some bleeding were observed. The Chiu’s pathologi- cal score of the ileum was significantly higher in the LSF group (P b .05) and HSF group (P b .01) than that in the CLP group. Also, the Chiu’s pathological score of the ileum of animals in the HSF group was signifi- cantly lower than that in the LSF group (P b .01; Table 3 and Fig. 1).

In addition, the AIs of ileal mucosal epithelial cells in LSF and HSF group samples were significantly lower than those in CLP group samples (P b .01). The AIs of ileal mucosal epithelial cells in HSF group samples were also significantly lower than those in LSF group samples (P b .01; Table 3).

Expression of the proapoptotic protein Bax was significantly lower (all P values b .05), and the expressions of the antiapoptotic protein Bcl-2 and tight junction protein occludin were significantly higher in

LSF and HSF group animals in comparison to the CLP group (all P values b .05). Expression of Bcl-2 and Bax was significantly higher and lower (all P values b .05) (Fig. 2) and occludin was significantly higher (P b .01), respectively, in the HSF group than those in the LSF group (Fig. 3).

Adverse reactions

Eight hours after surgery, blood samples were collected from rats in the LSF and HSF groups, and the circulating levels of ALT, AST, BUN, Cr, and CK-MB did not differ significantly from those in the CLP group (Table 1; P N .05).

Discussion

In this study, a model of severe sepsis was established in rats by CLP [13], and the influence of administration on intestinal epithelial cells was assessed in order to assess whether Shenfu protects the intestinal mucosal barrier. Cecal ligation and puncture resulted in elevated HR, body temperature, and WBC, indicating systemic inflammation, which was not detected in the sham surgery. It also significantly elevated serum levels of ALT, AST, BUN, and Cr, markers of liver and renal func- tion, and bacteria were detected in blood cultures and the peritoneal exudate cultures after CLP. This model met the criteria of the Guidelines for the Diagnosis and Treatment of Sepsis: 2012 [15], indicating success- ful establishment of a rat model of severe sepsis.

After severe injury and sepsis, accelerated apoptosis has been observed in intestinal mucosal epithelial cells, lamina propria lympho- cytes, and eosinocyts, leading to increased permeability of intestinal mucosa and bacterial translocation [16]. In the model of severe sepsis we established, intestinal mucosal morphology was significantly al- tered 8 hours after CLP surgery, and an increased rate of intestinal epithelial cell apoptosis was observed. These observations may be at- tributed to the redistribution of blood flow that occurs as a result of sep- sis, reducing the intestinal microcirculation [17]. Hypoxia-ischemia of intestinal mucosa and epithelial cells resulted in increased mucosal

Fig. 1. Hematoxylin and eosin staining of ileal tissues. Magnification, x100. In the sham group, the ileal mucosal villi were orderly arranged. The structure of surrounding blood vessels was normal. There was no obvious bleeding. The muscle fibers were orderly arranged, and the serosa was normal. In the CLP group, the ileal mucosal villi were damaged. Swelling, ulceration, and bleeding were observed around blood vessels, and fractured basal layer was observed. In the LSF group, slight damage to the ileal mucosal villi, local necrosis, and some bleeding were observed. In the HSF group, slight damage to the ileal mucosal villi, local necrosis, and a little bleeding were observed.

Fig. 2. Immunohistochemical staining of Bax (A) and Bcl-2 (B) in ileal tissues. Magnification, x400. C, integrated option density values of Bax and Bcl-2 in immunohistochemical assay. *P b .05 vs CLP group; **P b .01 vs CLP group; ^?P b .05 vs LSF group; ^??P b .01 vs LSF group. A, In the sham group, tan regions in the base of ileal mucosal villi can be observed normally, indicating positive results for Bax. In the CLP group, a lot of tan regions in the mucosal villi can be observed, indicating positive results for Bax. In the LSF group, many tan regions in the base of ileal mucosal villi can be observed, indicating positive results for Bax. In the HSF group, much more tan regions in the base of ileal mucosal villi can be observed, indicating positive results for Bax. B, In the sham group, a lot of tan regions in the base of ileal mucosal villi can be observed, indicating positive results for Bcl-2. In the CLP group, a few tan regions in the base of ileal

mucosal villi can be observed, indicating positive results for Bcl-2. In the LSF group, some tan regions in the base of ileal mucosal villi can be observed, indicating positive results for Bcl-2. In the HSF group, many tan regions in the base of ileal mucosal villi can be observed, indicating positive results for Bcl-2.

Fig. 3. Occludin expression in ileal tissues.

vascular permeability, WBC infiltration, and impairment of the mucosal barrier function translocation [16].

In addition, TNF-? is released at the initial stage of inflammation, stimulating expression of adhesion molecules on the surface of intesti- nal microvascular Endothelial cells and neutrophils, promoting neutro- phil aggregation, and contributing to the release of reactive oxygen and proteolytic enzymes, thereby destroying the intestinal mucosal epithe- lium and accelerating apoptosis of intestinal mucosal epithelial cells [18]. Tumor necrosis factor ? also triggers an inflammatory cascade pro- moting the secretion of IL-1, platelet-activating factor, IL-6, and nuclear factor ?B and the release of inflammatory mediators, such as IL-8, aggravating intestinal mucosal injury and inducing apoptosis of intesti- nal epithelial cells, and activating the coagulation pathway [18]. Microthrombus formation may contribute to the worsening of the intes- tinal mucosal circulatory disturbance, accelerating apoptosis of intesti- nal mucosal epithelial cells; this requires further investigation.

During sepsis, intestinal injury induces overexpression of inducible nitric oxide synthase and, thus, redundant production of Nitric oxide . The high concentration of NO leads to deposition of nitrite perox- ide and nitric peroxide on the mitochondrial membrane, impairing mi- tochondrial function and reducing tissue oxygen consumption [19]. Hypoxia and reduced metabolism may accelerate apoptosis of intestinal epithelial cells, as observed in our model. Eight hours after CLP surgery, the AI of intestinal epithelial cells was significantly higher in the CLP group than in the control group. Shenfu has previously reported to

improve coronary perfusion by promoting NO release [20]. In our model, Shenfu injection ameliorated CLP-induced ileal epithelial cell apoptosis in a dose-dependent manner.

Bcl-2 and Bax act, respectively, to prevent and promote apoptosis. During apoptosis, Bax translocates and tightly binds to the mitochondrial membrane, disrupting mitochondrial structure and function, triggering release of cytochrome c and Caspase activation, and promoting apopto- sis via multiple caspase-dependent pathways. Bcl-2 is predominantly located in the mitochondria, endoplasmic reticulum, and nuclear mem- brane and inhibits cytochrome C release; thus, the intracellular ratio of Bcl-2 to Bax determines susceptibility to apoptosis [21].

In our model, the ratio of Bcl-2 to Bax in intestinal epithelial cells was disturbed 8 hours after CLP surgery. We detected an increase in Bax protein, and the AI of intestinal epithelial cells was significantly in- creased. Shenfu injection significantly increased expression or retention of Bcl-2 in a dose-dependent manner, thus increasing the Bcl-2/Bax ratio and decreasing the AI of intestinal epithelial cells. These results suggest that Shenfu injection can prevent apoptosis of intestinal epithe- lial cells by regulating the Bcl-2/Bax ratio, an observation previously reported after Shenfu injection in cardiac and lung tissues in a porcine model of cardiac arrest [22,23] and a rat model of Myocardial ischemia-reperfusion injury [24]. Shenfu was also demonstrated to pro- tect cardiac myocytes against hypoxia/reoxygenation injury-induced apoptosis by inhibiting down-regulation of Bcl-2 [25]. Interestingly, Shenfu injection was shown to attenuate postresuscitation lung injury through suppression of lung cell apoptosis and improvement of Energy metabolism and antioxidant capacity [23], indicating its possible use in emergency settings, provided that these properties are confirmed by well-designed clinical trials.

The intestinal epithelial tight junctions are composed of transmem- brane proteins including occludin, junctional adhesion molecules, zonula occludins, and claudins [26]. Occludin performs a structural role, and the absence of occludin leads to failure in tight junction structural integrity, but it is also involved in the signal regulation of tight junction formation [27]. Both TNF-? and interferon-? down-regulate the occludin promoter, thus narrowing the occludin-localized tight junction [28]. Ma et al [29] re- ported that TNF-? not only reduced zonula occludin-1 expression but also reduced the phosphorylation of tight junction protein claudin-1, leading to dissociation of claudin-1 from the tight junction, disrupting the junction. In inflammation, stress, ischemia, or hypoxia, the occludin protein was reduced, deficient, or abnormally distributed, leading to structural abnormalities and dysfunction of tight junctions, thus broaden- ing intercellular spaces and increased intestinal mucosal epithelium per- meability [30], allowing bacteria, endotoxin, and toxic macromolecules pass through the tight junction and enter the bloodstream.

Eight hours after CLP surgery, we found significantly increased ex- pression or retention of occludin in the ileal mucosa. Shenfu injections significantly increased the expression of occludin in ileal mucosa in a dose-dependent manner, suggesting that Shenfu injection can protect the structural integrity and function of transmembrane proteins in the ileal mucosa, thus maintaining the integrity of intestinal intercellular space and reducing the permeability of intestinal mucosa.

The results obtained in this study indicate that in a rat model of severe sepsis, expression of systemic inflammatory factor TNF-? is increased, expression of anti-inflammatory factor IL-10 is decreased, ileal MDA levels increase, and SOD activity is decreased. The increased production of inflammatory factors and oxidative stress resulted in de- creased expression of antiapoptotic protein Bcl-2 and increased expres- sion of proapoptotic protein Bax in ileal mucosal epithelial cells, which caused apoptosis of epithelial cells, and thus significant pathological changes in the ileal mucosa. Furthermore, the level of occludin in epi- thelial cells was reduced, impairing the mechanical barrier of the intestinal mucosa. Intravenous administration of 5 or 10 mL/kg Shenfu injection significantly improved these parameters in rats; ameliorated changes in TNF-? and IL-10 production, ileal MDA level, and SOD activity; balanced the Bax/bcl-2 ratio; and reduced ileal mucosal

epithelial cell apoptosis. These effects ameliorated the degree of pathological change observed in the ileal mucosa and increased occludin expression, thereby significantly improving the intestinal mucosa mechanical barrier function in rats with severe sepsis. These effects were proportional to the dose of Shenfu injection. However, the underlying signaling pathways involved in the mechanism of Shenfu action remain to be determined. It should be noted that Shenfu injection, even at a high dose, did not completely abolish the sepsis- induced elevated WBC count, likely because the time point assessed (8 hours) was relatively early to allow microbial clearance. These issues will be addressed in future work. Finally, because the sample size used here was relatively small, further studies with more animals are needed to confirm our findings.

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ajem.2015.01.056.

References

AJEM