LPS (lipopolysaccharide), an outer membrane component of Gram-negative bacteria, plays an important role in the pathogenesis of sepsis and lipid A is known to be essential for its toxicity. Therefore it could be an effective measure to prevent sepsis by neutralizing or destroying LPS. Numerous studies have indicated that many traditional Chinese medicines are natural antagonists of LPS in vitro and in vivo. The goal of this study is to develop a rapid method to screen anti-sepsis components from Chinese herbs by use of a direct lipid A-based affinity biosensor technology based on a resonant mirror. The detergent OG (n-octyl β-D-glucopyranoside) was immobilized on a planar non-derivatized cuvette which provided an alternative surface to bind the terminal hydrophilic group of lipid A. A total of 78 herbs were screened based on the affinity biosensor with a target of lipid A. The aqueous extract of PSA (Paeonia suffruticosa Andr) was found to possess the highest capability of binding lipid A. Therefore an aqueous extraction from this plant was investigated further by our affinity biosensor, polyamide chromatography and IEC–HPLC. Finally, we obtained a component (PSA-I-3) from Paeonia suffruticosa Andr that was evaluated with the affinity biosensor. We also studied the biological activities of PSA-I-3 against sepsis in vitro and in vivo to further confirm the component we screened with the biosensor. In vitro, we found that PSA-I-3 could decrease TNFα (tumour necrosis factor α) release from RAW264.7 cells induced by LPS in a dose-dependent manner. In vivo, it increased remarkably the survival of KM (KunMing) mice by challenging both lethal-dose LPS and heat-killed Escherichia coli compared with control groups. Our results suggest that the constructed affinity biosensor can successfully screen the anti-sepsis component from Chinese herbs.

INTRODUCTION

Sepsis is a complex syndrome that is defined as the presence of infection plus SIRS (systemic inflammatory response syndrome) [1]. In the USA, sepsis syndrome is responsible for a tenth of all deaths. The mortality rate for inpatients with sepsis is approximately 30–55% and higher for the elderly [2,3]. Moreover, the conditions of sepsis are worse in developing countries owing to poor healthcare and environment. Although Gram-negative bacteria are no longer the number one cause of sepsis in recent years, as people now pay more attention [4,5], infections still result in severe outcomes such as sepsis, severe sepsis and septic shock in most of these critically ill populations who do not receive appropriate and timely therapy [6].

LPS (lipopolysaccharide) is an outer membrane component of Gram-negative bacteria and more than 2×106 LPS molecules comprise a polysaccharide part, O-antigen and core regions, and a lipid anchor, called lipid A, known to be essential for its toxicity [7]. Numerous studies have indicated that LPS triggers pathogenesis of sepsis by recognizing TLR4 (Toll-like receptor 4) and CD14 on the surface of cells; so an effective way to prevent sepsis could be by neutralizating LPS [812]. In recent years, many therapeutiic strategies were taken against LPS including McAbs, polyclonal antibodies, leucocyte proteins (BPI), PMB (polymyxin B), mimic peptides, etc. [1319]. Unfortunately, to date, there is not yet a safe and effective agent to protect victims from sepsis caused by LPS.

In China, Chinese herbs have been used as drugs for critically ill patients for thousands of years. There is evidence that also suggests that many Chinese traditional herbs possess an anti-sepsis function [20]; however, traditional herbs are so complex that their clinical applications are restricted greatly. We have thus become interested in exploring a method with advantages of being sensitive, fast, accurate, label-free, simple and quantitative. In the present study, lipid A was immobilized on the OG (n-octyl β-D-glucopyranoside)-coated surface of a non-derivatized cuvette to construct an affinity biosensor developed to screen anti-sepsis herb technology in 2004. This identified some anti-sepsis fractions or monomers from Radix Paeoniae Rubra, Scutellaria baicalensis Georgi and Terminaliachebula Retz in the following years [2123]. Herein, we found that 12 aqueous extractions from 78 Chinese herbs could bind greatly to lipid A which was immobilized on to a non-derivatized cuvette of the affinity biosensor and the non-tannin aqueous extraction from PSA (Paeonia suffruticosa Andr) possessed the highest lipid A-binding ability after consuming the indicated LPS concentration. Therefore an aqueous extraction from this plant was investigated further by an affinity biosensor, polyamide chromatography and IEC–HPLC. Finally, we obtained a fraction (PSA-I-3) from Paeonia suffruticosa Andr and investigated its anti-sepsis activities in vitro and in vivo.

MATERIALS AND METHODS

Reagents

LPS from Escherichia coli O111:B4 and lipid A from Salmonella Re 595 were obtained from Sigma Chemicals. Mouse TNFα (tumour necrosis factor α) ELISA kits were obtained from Biosource International. The kinetic turbidimetric LAL (Limulus amoebocyte lysate) kit was obtained from Zhanjiang A & C Biological Ltd., and gelatine was purchased from Sino-pharm Chemical Reagent Co., LtdS.

Chinese herbs

Twelve Chinese herbs were identified in the Chongqing Academy of the Chinese Materia Medica (Chongqing, China) which were purchased from Sichuan Province.

Mice

Eighty KM (KunMing) mice (4–6 weeks old), with an equal number of males and females at random, were obtained from the Experimental Animal Center of Chongqing Medical University (Chongqing, China). The weight of the mice was 20±2 g. The mice were housed in SPF (specific pathogen-free) conditions until use, and in agreement with principles stated in the Guide for the Care and Use of Laboratory Animals. National Research Council Publication, 1996 edition.

Cell line and culture

The mouse macrophage cell line RAW264.7 was obtained from American Type Culture Collection (Manassas, VA), and was grown in DMEM (Dulbecco's modified Eagle's medium) supplemented with 10% (v/v) FBS (HyClone), 2 μM glutamine, 100 units/ml penicillin and 100 μg/ml streptomycin at 37°C in a humidified 5% (v/v) CO2 incubator. In each experiment, 106 cells/well were used.

Preparation of bacterial strains

E. coli (ATCC 25922) maintained in our laboratory were subcultured on the LB (Luria–Bertani) agar plate for 18 h, and then passaged into LB broth (containing 10 g tryptone, 10 g NaCl and 5 g yeast extract per litre) and cultivated aerobically in a 50 ml volume at 37°C in a heated, shaking environmental chamber for 18 h and transferred to 1000 ml of fresh LB broth for another 18 h. They were harvested by centrifugation. Finally, the bacteria were resuspended in sterile NS (normal saline) and killed at 100°C in hot water for 30 min.

Preparation of aqueous extracts of 12 Chinese herbs

Aqueous extracts of 12 Chinese herbs were prepared as described in [22]. Briefly, 2 g of 12 herbs were added to 10 ml distilled water after pulverization and heated in a water bath at 100°C for 1 h. The 12 aqueous extractions were centrifuged at 4°C at 5000 rev/min for 3 min and each supernatant was collected after filtration for the next assays.

Removing tannins from 12 Chinese herbs

The 12 supernatants as stated above were co-incubated with excessive gelatine (4%, w/v) at 50°C for 1 h. The supernatants, after centrifugation at 4°C, 10 000 rev/min for 5 min, were collected and stored at 4°C for the following assays.

Consumption assays of the indicated LPS concentration

Approximately 50 μl supernatants, after removing tannins, were co-incubated in the presence or absence of 50 μl LPS (20, 50 ng/ml) at 30°C for 30 min. After centrifugation at 4°C, 5000 rev/min for 3 min the supernatants were collected and stored at 4°C for use in lipid A-binding assays.

Lipid A was immobilized on to the surface of a non-derivatized cuvette according to the manufacturer's instruction (Thermo Labsystem) as described in [22]. Briefly, 5 ml of 12 supernatants after consumption of the indicated LPS concentration was assayed for its binding activity to lipid A as described in [22]. Data are shown in Table 1.

Table 1
Lipid A-binding activity of aqueous extracts bound to the indicated LPS concentration after removing tannins
Response unit (arc-sec)
Extract50 ml (NS)20 ng (LPS)50 ng (LPS)
Scutellaria baicalensis Georgi 190 133 129 
Terminaliachebula Retz 473 255 211 
Fructus crataegi 580 348 227 
Biota orientalis 638 573 215 
Rhizoma coptidis 419 278 219 
Radix et Rhizoma Rhei 810 589 564 
Chinese White Olive 581 228 135 
Tripterygium Wilfordii 427 368 298 
Radix Paeoniae Rubra 511 469 244 
Radix Sanguisorbae 626 555 369 
Caulis Sargentodoxae 633 583 553 
Paeonia suffruticosa Andr 952 702 654 
Response unit (arc-sec)
Extract50 ml (NS)20 ng (LPS)50 ng (LPS)
Scutellaria baicalensis Georgi 190 133 129 
Terminaliachebula Retz 473 255 211 
Fructus crataegi 580 348 227 
Biota orientalis 638 573 215 
Rhizoma coptidis 419 278 219 
Radix et Rhizoma Rhei 810 589 564 
Chinese White Olive 581 228 135 
Tripterygium Wilfordii 427 368 298 
Radix Paeoniae Rubra 511 469 244 
Radix Sanguisorbae 626 555 369 
Caulis Sargentodoxae 633 583 553 
Paeonia suffruticosa Andr 952 702 654 

Isolation of lipid A-binding fractions from PSA

Approximately 1 kg PSA was boiled in distilled water at 80°C for 1 h via regurgitant extraction and tannins were removed by excessive gelatine (4%) at 50°C for 1 h before filtration. The filtrate was centrifuged at 4000 g for 30 min and the supernatant was collected, which was passed over column chromatography with polyamide (80–100 mesh, 45×250 mm), eluted with distilled water, absolute ethanol and NH3.H2O (1%). The elution was collected and concentrated by rotary evaporation (BUCHI Rotavapor R205) to collectively obtain four fractions, named PSAI–IV. Briefly 5 μl PSAI–IV (10 μg/ml) was assayed for its binding activity to lipid A, as described in [22]. The results indicated that the first fraction (PSA-I) bound more markedly to lipid A than the others, and at more than 1800 arc-sec (Figure 1b).

Aqueous extracts from PSA and their lipid A-binding abilities

Figure 1
Aqueous extracts from PSA and their lipid A-binding abilities

(a) Isolation with polyamide chromatography. Aqueous extract of PSA was passed over column chromatography with polyamide (80–100 mesh, 45 mm×250 mm), and eluted with distilled water, absolute ethanol and NH3.H2O (1%), respectively. We obtained together four fractions, named PSAI–IV. (b) Five microlitres of PSAI–IV (10 μg/ml) were assayed for binding activity to lipid A. Lines I–IV represented their lipid A-binding abilities.

Figure 1
Aqueous extracts from PSA and their lipid A-binding abilities

(a) Isolation with polyamide chromatography. Aqueous extract of PSA was passed over column chromatography with polyamide (80–100 mesh, 45 mm×250 mm), and eluted with distilled water, absolute ethanol and NH3.H2O (1%), respectively. We obtained together four fractions, named PSAI–IV. (b) Five microlitres of PSAI–IV (10 μg/ml) were assayed for binding activity to lipid A. Lines I–IV represented their lipid A-binding abilities.

In order to investigate the anti-LPS characteristics of PSA-I, we further purified the first fraction (PSA-I) via IEC–HPLC. To this end, the fraction (PSA-I) was injected on to the IEC–HPLC system (SPFF 90 μm, GE Healthcare), eluted with acetic acid (pH 4.0), double-distilled water and NH3.H2O (1%) adding ethanol (1%, v/v), respectively, to obtain three HPLC fractions (PSA1–3). Briefly 5 μl of PSA1–3 (10 μg/ml) was assayed for its binding activity to lipid A as described previously [22]. We found that the third fraction (PSA-I-3) bound markedly to lipid A at up to 550 arc-sec (Figure 2b).

The elution from PSA-I via IEC–HPLC and their lipid A-binding abilities

Figure 2
The elution from PSA-I via IEC–HPLC and their lipid A-binding abilities

(a) In order to investigate further the anti-LPS characteristics of PSA-I, we purified further the first fraction (PSA-I) via IEC–HPLC. To this end, the fraction (PSA-I) was injected on to an IEC–HPLC system (SPFF 90 μm, GE Helathcare), eluted with acetic acid (pH 4.0), double-distilled water and NH3.H2O (1%) adding ethanol (1%), respectively. We obtained, in total, three HPLC fractions, named PSA1–3. (b) Five microlitres of PSA1–3 (10 μg/ml) were assayed for binding activity to lipid A. Lines 1–3 represented their lipid A-binding abilities.

Figure 2
The elution from PSA-I via IEC–HPLC and their lipid A-binding abilities

(a) In order to investigate further the anti-LPS characteristics of PSA-I, we purified further the first fraction (PSA-I) via IEC–HPLC. To this end, the fraction (PSA-I) was injected on to an IEC–HPLC system (SPFF 90 μm, GE Helathcare), eluted with acetic acid (pH 4.0), double-distilled water and NH3.H2O (1%) adding ethanol (1%), respectively. We obtained, in total, three HPLC fractions, named PSA1–3. (b) Five microlitres of PSA1–3 (10 μg/ml) were assayed for binding activity to lipid A. Lines 1–3 represented their lipid A-binding abilities.

LPS neutralization by PSA-I-3 in vitro

To determine whether PSA-I-3 could antagonize the bioactivity of LPS, we verified LPS neutralization by PSA-I-3 in vitro using the Limulus test, which is widely used to detect the concentration of microamounts of endotoxin. For this experiment, two concentrations of the PSA-I-3 (1 and 10 μg/ml) were co-incubated in the presence or absence of LPS (10 ng) at 37°C for 30 min; the control experiments were as follows: positive control (1 μg PMB+10 ng LPS), negative control (10 ng LPS) and internal controls (sterile NS). Then 100 μl of each mixture was added to LAL reagent of an equal volume, and the kinetic turbidity was measured using an EDS-99 Tube Reader (Zhanjiang A&C Biological Ltd.).

Inhibition of TNFα release by RAW264.7 cells induced by LPS

For this experiment, 1×105 RAW264.7 cells (200 μl) were plated on to a 96-well microplate and cultured at 37°C in a humified 5% CO2 incubator for 4 h to subconfluency. The cells were pretreated with PSA-I-3 at the indicated dose (45, 90 and 180 μg/ml) for 30 min, followed by replacement with 0.5 ml serum-free DMEM. Cells were then stimulated with and without LPS (100 μg/l) for 4 h. Positive control (10 μg PMB+100 μg/l LPS), negative controls and blank control groups were 100 μg/l LPS and an equal volume of serum-free DMEM, respectively. Supernatants were collected to test the TNFα level using mouse ELISA kits.

Protection of PSA-I-3 on mice challenged by LPS

A total of 40 mice (KM) were randomly divided into four groups (10 mice per group), mice (KM) were challenged with LPS (18 mg/kg) via the tail vein. Subsequently, PMB (2 mg/kg) and PSA-I-3 (100 mg/kg) were injected as the positive control and treatment groups using identical procedures as stated above. PSA-I-3 (100 mg/kg) alone was used as a blank control. The mice survival was assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h.

Protection of PSA-I-3 on mice challenged by heat-killed E. coli

A total of 40 mice (KM) were randomly divided into four groups (10 mice per group) and were injected via the tail vein as follows: heat-inactivated E. coli [6.0×109 CFU (colony-forming units)/kg] alone as the negative control, heat-inactivated E. coli (6.0×109 CFU/kg) adding dexamethasome (5 mg/kg) as the positive control, and heat-inactivated E. coli (6.0×109 CFU/kg) and immediate subsequent injection of PSA-I-3 (100 mg/kg) as the treatment group. PSA-I-3 (100 mg/kg) alone was the blank control. The general conditions and mice mortalities were assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h.

Statistical analysis

Statistical comparisons of the survival between the experimental and control groups were made using the χ2 test, and others were conducted via independent samples using Student's t test with the SPSS 11.0 software package. P<0.05 (double-sided) was considered as significant, and P<0.01 was considered as very significant.

RESULTS

Aqueous extracts of 12 Chinese herbs (after removal of tannins) bound to the indicated LPS concentration

In previous experiments, we found that 12 out of the 78 herbs possess high lipid A-binding activities (more than 200 arc-sec) and the tannins have a great impact on binding between lipid A and Chinese herbs. Therefore we carried out an experiment to remove tannins from Chinese herbs, and then the 12 supernatants, after removal of tannins, were co-incubated in the presence or absence of the indicated LPS concentration. The lipid A-binding assays showed that Radix et Rhizoma Rhei, Caulis Sargentodoxae and PSA still possessed higher response units than that of others, and at more than 500 arc-sec. Furthermore, PSA showed the highest lipid A-binding ability among the 12 herbs (Table 1). So, we further investigated the anti-LPS ability of PSA using an affinity biosensor, polyamide chromatography and IEC–HPLC.

Lipid A-binding fractions from PSA

To determine which components of PSA could inhibit LPS, we isolated the aqueous extracts and tested their lipid A-binding abilities via polyamide chromatography and an affinity biosensor. As a result, four fractions were obtained from the aqueous extract of PSA (Figure 1a) and bound to lipid A which was immobilized on the surface of a non-derivatized cuvette. We found that the first fraction possessed the higher lipid A-binding activity compared with the other fractions, at more than 1800 RU (arc-sec) (Figure 1b).

To further study the bioactivities of the PSA-I fraction, PSA-I was passed on an IEC–HPLC system (SPFF 90 μm). We collected three HPLC fractions (PSA1–3) (Figure 2a) and measured the lipid A-binding activity. Our results showed that the third fraction (PSA-I-3) bound more significantly to lipid A than other fractions, and at up to 550 arc-sec (Figure 2b). These results suggested that PSA-I-3 perhaps could be an inhibitor of LPS in vitro or in vivo.

LPS neutralization and inhibition of TNFα release induced by LPS in vitro in RAW264.7 cells

To further evaluate the biological activities of PSA-I-3 against LPS in vitro, we measured its abilities to neutralize LPS and to inhibit TNFα release (by LPS challenge) in RAW264.7 cells. In agreement with our hypotheses, the results showed that PSA-I-3 significantly inhibited the LPS-induced RAW264.7 cells to release TNFα in a concentration-dependent manner compared with the negative control (10 ng LPS) (P<0.001) (Figure 3). Beyond our expectation, however, PSA-I-3 could not inhibit the lysis of Limulus amoebocytes by LPS, it enhanced LAL instead (Figure 4), and the underlying mechanisms remain unclear.

Inhibition of TNFα release from LPS-stimulated RAW264.7 cells

Figure 3
Inhibition of TNFα release from LPS-stimulated RAW264.7 cells

For this experiment, 1×105 RAW264.7 cells (200 μl) were plated on to a 96-well microplate and cultured at 37°C in a humified 5% CO2 incubator for 4 h to subconfluency. The cells were pretreated with PSA-I-3 at the indicated dose (45, 90 and 180 μg/ml) for 30 min followed by replacement with 0.5 ml serum-free DMEM, and then stimulated with and without LPS (100 μg/l) for 4 h. RAW264.7 cells were also treated with a positive control (10 μg PMB+100 μg/l LPS), and negative and blank control groups were 100 μg/l LPS and an equal volume of serum-free DMEM respectively. Supernatants were collected to test the TNFα level using mouse ELISA kits. The value of TNFα was expressed as mean(pg/ml)±S.D. A P-value <0.05 (double-sided) was considered as significant, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

Figure 3
Inhibition of TNFα release from LPS-stimulated RAW264.7 cells

For this experiment, 1×105 RAW264.7 cells (200 μl) were plated on to a 96-well microplate and cultured at 37°C in a humified 5% CO2 incubator for 4 h to subconfluency. The cells were pretreated with PSA-I-3 at the indicated dose (45, 90 and 180 μg/ml) for 30 min followed by replacement with 0.5 ml serum-free DMEM, and then stimulated with and without LPS (100 μg/l) for 4 h. RAW264.7 cells were also treated with a positive control (10 μg PMB+100 μg/l LPS), and negative and blank control groups were 100 μg/l LPS and an equal volume of serum-free DMEM respectively. Supernatants were collected to test the TNFα level using mouse ELISA kits. The value of TNFα was expressed as mean(pg/ml)±S.D. A P-value <0.05 (double-sided) was considered as significant, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

LPS neutralization by PSA-I-3 in vitro

Figure 4
LPS neutralization by PSA-I-3 in vitro

To determine whether PSA-I-3 could antagonize the bioactivity of LPS, we verified LPS neutralization of PSA-I-3 in vitro via a Limulus test. To this end, two concentrations of the PSA-I-3 (1 and 10 μg/ml) were co-incubated in the presence or absence of LPS (10 ng) at 37°C for 30 min. Control experiments were as follows: positive control (1 μg PMB+10 ng LPS); negative control, 10 ng LPS; and the internal control was sterile NS. Then 100 μl of each mixture was added to LAL reagent of an equal volume, The kinetic turbidity was measured using an EDS-99 Tube Reader. The value of LPS was expressed as mean (EU/ml)±S.D. A P-value <0.05 (double-sided) was considered significant, and a P-value <0.01 was considered as very significant. The test was carried out more than three times.

Figure 4
LPS neutralization by PSA-I-3 in vitro

To determine whether PSA-I-3 could antagonize the bioactivity of LPS, we verified LPS neutralization of PSA-I-3 in vitro via a Limulus test. To this end, two concentrations of the PSA-I-3 (1 and 10 μg/ml) were co-incubated in the presence or absence of LPS (10 ng) at 37°C for 30 min. Control experiments were as follows: positive control (1 μg PMB+10 ng LPS); negative control, 10 ng LPS; and the internal control was sterile NS. Then 100 μl of each mixture was added to LAL reagent of an equal volume, The kinetic turbidity was measured using an EDS-99 Tube Reader. The value of LPS was expressed as mean (EU/ml)±S.D. A P-value <0.05 (double-sided) was considered significant, and a P-value <0.01 was considered as very significant. The test was carried out more than three times.

Protection of PSA-I-3 on mice challenged by LPS and heat-killed E. coli

To further investigate the biological activities of PSA-I-3 against LPS in vivo, 40 mice (KM) were injected with PSA-I-3 (100 mg/kg) prior to a lethal challenge with LPS [18 mg/kg, i.v. (intravenously)]. The survival of mice was assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h, respectively. As a result, mice began to die 6 h after challenge with LPS, and the survival of mice with PMB and PSA-I-3 treatment were 80 and 50% after 6 days, respectively. However, nine out of ten mice were dead by challenging with LPS alone approximately 1 week later. Our results suggest that PSA-I-3 protects mice significantly from challenge by a lethal-dose of LPS (P<0.005) (Figure 5).

Protection of PSA-I-3 on mice challenged by LPS

Figure 5
Protection of PSA-I-3 on mice challenged by LPS

Forty mice (KM) were randomly divided into four groups (10 mice per group), mice (KM) were challenged with LPS (18 mg/kg) via the tail vein. Subsequently, PMB (2 mg/kg) and PSA-I-3 (100 mg/kg) were injected as the positive control and treatment groups by identical manners as stated above. PSA-I-3 (100 mg/kg) alone was the blank control. The survival of mice was assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h. Statistical comparisons of the survival between the experimental and control groups were made using the χ2 test, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

Figure 5
Protection of PSA-I-3 on mice challenged by LPS

Forty mice (KM) were randomly divided into four groups (10 mice per group), mice (KM) were challenged with LPS (18 mg/kg) via the tail vein. Subsequently, PMB (2 mg/kg) and PSA-I-3 (100 mg/kg) were injected as the positive control and treatment groups by identical manners as stated above. PSA-I-3 (100 mg/kg) alone was the blank control. The survival of mice was assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h. Statistical comparisons of the survival between the experimental and control groups were made using the χ2 test, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

To determine whether PSA-I-3 also protected mice from challenge by natural LPS from E. coli, 100 mg/kg PSA-I-3 was administered followed by challenge with heat-killed E. coli (6.0×109 CFU/kg) via the tail vein. Post-challenge, the number of dead mice recorded was highest within 12 h and more than 60% of mice were killed in the non-treatment group, and none of the mice that survived died after 24 h in all of the groups. Compared with the non-treatment group, PSA-I-3 improved the survival of experimental animals but this did not show a significant difference (Figure 6).

Protection of PSA-I-3 on mice challenged by heat-killed E coli

Figure 6
Protection of PSA-I-3 on mice challenged by heat-killed E coli

Forty mice (KM) were randomly divided into four groups (10 mice per group) and were injected via the tail vein as follows: heat-inactivated E. coli (6.0×109 CFU/kg) alone as a negative control, and heat-inactivated E. coli (6.0×109 CFU/kg) adding dexamethasone (5 mg/kg) as the positive control. Heat-inactivated E. coli (6.0×109 CFU/kg) and immediate subsequent injection of PSA-I-3 (100 mg/kg) was the treatment group. PSA-I-3 (100 mg/kg) alone was a blank control. The general conditions and mice mortalities were assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h. Statistical comparisons of the survival between the experimental and control groups were made using the χ2 test, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

Figure 6
Protection of PSA-I-3 on mice challenged by heat-killed E coli

Forty mice (KM) were randomly divided into four groups (10 mice per group) and were injected via the tail vein as follows: heat-inactivated E. coli (6.0×109 CFU/kg) alone as a negative control, and heat-inactivated E. coli (6.0×109 CFU/kg) adding dexamethasone (5 mg/kg) as the positive control. Heat-inactivated E. coli (6.0×109 CFU/kg) and immediate subsequent injection of PSA-I-3 (100 mg/kg) was the treatment group. PSA-I-3 (100 mg/kg) alone was a blank control. The general conditions and mice mortalities were assessed at 6, 12, 24, 36, 48, 60, 72, 96 and 108 h. Statistical comparisons of the survival between the experimental and control groups were made using the χ2 test, and a P-value <0.01 was considered as very significant. The test was carried out no less than three times.

DISCUSSION

Previously, we have reported the successful establishment of a platform to screen the active ingredients targeting CpG DNA from traditional Chinese herbs, and found an active component to bind with the platform. In the present study, we also used affinity biosensor technology to establish the platform to screen active ingredients from traditional Chinese herbs targeting lipid-A. It is a different target of LPS, and it was also confirmed that affinity biosensor technology is a rapid and effective method to screen the active fractions from traditional Chinese herbs.

The present study suggested that a lipid A-binding activity peptide possesses endotoxin-neutralizing molecules that could have important clinical applications, which means that a high-binding affinity substance could neutralize endotoxin. However, we have to be sure that the substance is not toxic because it will serve as a treatment for sepsis. For example, PMB possesses high-affinity endotoxin binding, however, it also has severe side effects in humans [24]. Therefore we need to find a new compound with a high affinity to neutralize LPS, but with no toxicity.

PSA (Mudan Pi), belonging to the Paeoniaceae family, is an important traditional Chinese medicine used in many traditional prescriptions. An increasing number of studies showed that extracts of PSA possessed a large cohort of pharmacological functions such as anti-inflammation, anti-oxidation, anti-tumour and bacteriocidal functions [2528]. However, the mechanisms behind the anti-inflammation effects are largely unknown owing to the absence of an effective tracking approach.

Tannins are water-soluble polyphenols, which present in many Chinese herbs and plant foods, which possess many bioactivities, such as antimicrobial, anticarcinogenic, antimutagenic, etc. In our previous experiments, we found that tannins from Radix Paeoniae Rubra bound greatly to lipid A and were anti-LPS in vivo and in vitro. However, this was restricted in clinical applications owing to a property of protein precipitation by intravenous administration. Herein, we first removed tannins of 12 Chinese herbs by adding excessive gelatine, and then co-incubated with the indicated LPS concentration. Finally, we tested the lipid A-binding abilities of each aqueous extraction. Our results demonstrated that the non-tannic aqueous extract from PSA bound greatly to lipid A after consumption with LPS. These findings suggest that PSA may possess abundant anti-LPS components among 12 Chinese herbs.

In order to investigate further the anti-LPS characteristics of PSA, we isolated the non-tannic aqueous extraction by affinity biosensor, polyamide chromatography and IEC–HPLC. Finally, we got a high-binding lipid A fraction from PSA (namely PSA-I-3) and tested the anti-sepsis activities in vivo and in vitro. Our results showed that PSA-I-3 could increase the survival of mice by lethal-dose LPS and heat-killed E. coli in vivo and inhibit significantly the TNFα release from LPS-stimulated RAW264.7 cells in vitro. However, PSA-I-3 could not arrest the lysis of Limulus amoebocytes by LPS. Instead, it enhanced LAL in spite of binding highly to lipid A The underlying mechanism is unclear, but we think that it is perhaps related to the purity of PSA-3. In conclusion, our results showed that PSA-3 possessed a potential anti-sepsis ability, and it is necessary to further investigate its bioactivities in the future.

Abbreviations

     
  • CFU

    colony-forming units

  •  
  • DMEM

    Dulbecco’s modified Eagle’s medium

  •  
  • KM

    KunMing

  •  
  • LAL

    Limulus amoebocyte lysate

  •  
  • LB

    Luria–Bertani

  •  
  • LPS

    lipopolysaccharide

  •  
  • NS

    normal saline

  •  
  • OG

    n-octyl β-D-glucopyranoside

  •  
  • PMB

    polymyxin B

  •  
  • PSA

    Paeonia suffruticosa Andr

  •  
  • SIRS

    systemic inflammatory response syndrome

  •  
  • SPF

    specific pathogen free

  •  
  • TNFα

    tumour necrosis factor α

AUTHOR CONTRIBUTION

Yiguo Chen, Jie Yao, Nin Wang and Dongneng Jiang performed the research. Jiang Zheng designed the research. Jie Yao and Yiguo Chen wrote the paper.

FUNDING

This work was supported by the National Nature Scientific Foundation of China [grant numbers 81000225 and 81071537].

References

References
1
Levy
 
M. M.
Fink
 
M. P.
Marshall
 
J. C.
Abraham
 
E.
Angus
 
D.
Cook
 
D.
Cohen
 
J.
Opal
 
S. M.
Vincent
 
J. L.
Ramsay
 
G.
 
2001 SCCM/ESICM/ACCP/ATS/SIS Int. Sepsis Definitions Conf
Crit. Care Med.
2003
, vol. 
31
 (pg. 
1250
-
1256
)
[PubMed]
2
Minino
 
A.
Heron
 
M.
Murphy
 
S.
Kochanedk
 
K.
 
Deaths: final data for 2004
Natl Vital Stat. Rep.
2007
, vol. 
55
 (pg. 
1
-
120
)
[PubMed]
3
Angus
 
D.
Linde-Zwirble
 
W.
Lidicker
 
J.
Clermont
 
G.
Carcillo
 
J.
Pinsky
 
M.
 
Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care
Crit. Care Med.
2001
, vol. 
29
 (pg. 
1303
-
1310
)
[PubMed]
4
Annane
 
D.
Bellissant
 
E.
Cavaillon
 
J.-M.
 
Septic shock
Lancet
2005
, vol. 
365
 (pg. 
63
-
78
)
[PubMed]
5
Chapmana
 
S. J. R.
Iredell
 
J. R.
 
Gram-negative sepsis in the intensive care unit:avoiding therapeutic failure
Curr. Opin. Infect. Dis.
2008
, vol. 
21
 (pg. 
604
-
609
)
[PubMed]
6
Kang
 
C. I.
Kim
 
S. H.
Park
 
W. B.
Lee
 
K. D.
Kim
 
H. B.
Kim
 
E. C.
Oh
 
M. D.
Choe
 
K. W.
 
Bloodstream infections caused by antibioticresistant Gram-negative bacilli: risk factors for mortality and impact of inappropriate initial antimicrobial therapy on outcome
Antimicrob. Agents Chemother.
2005
, vol. 
49
 (pg. 
760
-
766
)
[PubMed]
7
Rietschel
 
E. T.
Kirikae
 
T.
Schade
 
F. U.
Mamat
 
U.
Schmidt
 
G.
Loppnow
 
H.
Ulmer
 
A. J.
Zǎhringer
 
U.
Seydel
 
U.
DiPadova
 
F.
 
Bacterial endotoxin: molecular relationships of structure to activity and function
FASEB J.
1994
, vol. 
8
 (pg. 
217
-
225
)
[PubMed]
8
Iwagaki
 
A.
Porro
 
M.
Pollack
 
M.
 
Influence of synthetic antiendotoxin peptides on lipopolysaccharide (LPS) recognition and LPS-induced proinflammatory cytokine responses by cells expressing membrane-bound CD14
Infect. Immun.
2000
, vol. 
68
 (pg. 
1655
-
1663
)
[PubMed]
9
Vallespi
 
M. G.
Alvarez-Obregón
 
J. C.
Rodriguez-Alonso
 
I.
Montero
 
T.
Garay
 
H.
Reyes
 
O.
Araña
 
M. J.
 
A Limulus anti-LPS factor-derived peptide modμlates cytokine gene expression and promotes resolution of bacterial acute infection in mice
Int. Immunopharmacol.
2003
, vol. 
3
 (pg. 
247
-
256
)
[PubMed]
10
Ochia
 
M.
Tamura
 
H.
Yamamoto
 
A.
Aizawa
 
M.
Kataoka
 
M.
Toyoizumi
 
H.
Horiuchi
 
Y.
 
A limμlus amoebocyte lysate activating activity (LAL activity) that lacks biological activities of endotoxin found in biological products
Microbiol. -Immunol.
2002
, vol. 
46
 (pg. 
527
-
533
)
[PubMed]
11
Tsuzuki
 
H.
Tani
 
T.
Ueyama
 
H.
Kodama
 
M.
 
Lipopolysaccharide: neutralization by polymyxin B shuts down the signaling pathway of nuclear factor kB in peripheral blood mononuclear cell, even during activation
J. Surg. Res.
2001
, vol. 
100
 (pg. 
127
-
134
)
[PubMed]
12
Derhaschnig
 
U.
Reiter
 
R.
Knöbl
 
P.
Baumgartner
 
M.
Keen
 
P.
Jilma
 
B.
 
Recombinant human activated protein C (rhAPC; drotrecogin alfa [activated]) has minimal effect on markers of coagulation, fibrinolysis, and inflammation in acute human endotoxemia
Blood
2003
, vol. 
102
 (pg. 
2093
-
2098
)
[PubMed]
13
Dunn
 
D. L.
Priest
 
B. P.
Condie
 
R. M.
 
Protective capacity of polyclonal and monoclonal antibodies directed against endotoxin during experimental sepsis
Arch. Surg.
1988
, vol. 
123
 (pg. 
1389
-
1393
)
[PubMed]
14
Klaerner
 
H. G.
Dahlberg
 
P. S.
Acton
 
R. D.
Battafarano
 
R. J.
Uknis
 
M. E.
Johnston
 
J. W.
Dunn
 
D. L.
 
Immunization with antibodies that mimic LPS protects against gram negative bacterial sepsis
J. Surg. Res.
1997
, vol. 
69
 (pg. 
249
-
254
)
[PubMed]
15
Jiang
 
Z.
Hong
 
Z.
Guo
 
W.
Xiaoyun
 
G.
Gengfa
 
L.
Yongning
 
L.
Guangxia
 
X.
 
A synthetic peptide derived from bactericidal/permeability-increasing protein neutralizes endotoxin in vitro and in vivo
Int. Immunopharmacol.
2004
, vol. 
4
 (pg. 
527
-
537
)
[PubMed]
16
Cross
 
A. S.
Opal
 
S.
Cooka
 
P.
Drabick
 
J.
Bhattacharjee
 
A.
 
Development of an anti-core lipopolysaccharide vaccine for the prevention and treatment of sepsis
Vaccine
2004
, vol. 
22
 (pg. 
812
-
817
)
[PubMed]
17
Surolia
 
N.
Surolia
 
A.
 
Polymyxin B: an ode to an old antidote for endotoxic shock
Mol. Biosyst.
2005
, vol. 
1
 (pg. 
213
-
222
)
[PubMed]
18
Yibin
 
G.
Jiang
 
Z.
Hong
 
Z.
Gengfa
 
L.
Liangxi
 
W.
Guo
 
W.
Yongling
 
L.
 
A synthesized cationic tetradecapeptide from hornet venom kills bacteria and neutralizes lipopolysaccharide in vivo and in vitro
Biochem. Pharmacol.
2005
, vol. 
70
 (pg. 
209
-
219
)
[PubMed]
19
Jiang
 
X. T.
Liu
 
B. Y.
Zhu
 
P.
Fu
 
N.
 
Production and characterization of a cross-reactive monoclonal antibody to lipopolysaccharide
Hybridoma (Larchmt)
2009
, vol. 
28
 (pg. 
93
-
99
)
[PubMed]
20
Wang
 
H.
Xu
 
T.
Lewin
 
M. R.
 
Future possibilities for the treatment of septic shock with herbal components
Am. J. Emergency Med.
2009
, vol. 
27
 (pg. 
107
-
111
)
21
Genfa
 
L.
Jiang
 
Z.
Hong
 
Z.
Yimin
 
Z.
Liangxi
 
W.
Guo
 
W.
Ming
 
H.
Donglen
 
J.
Lizhao
 
W.
 
The screening and isolation of an effective anti-endotoxin monomer from Radix Paeoniae Rubra using affinity biosensor technology
Int. Immunopharmacol.
2005
, vol. 
5
 (pg. 
1007
-
1017
)
[PubMed]
22
Fu
 
J.
Cao
 
H.
Wang
 
N.
Zheng
 
X.
Lu
 
Y.
Liu
 
X.
Yang
 
D.
Li
 
B.
Zheng
 
J.
Zhou
 
H.
 
An anti-sepsis monomer, 2′,5,6′,7-tetrahydroxyflavanonol (THF), identified from Scutellaria baicalensis Georgi neutralizes lipopolysaccharide in vitro and in vivo
Int. Immunopharmacol.
2008
, vol. 
8
 (pg. 
1652
-
1657
)
[PubMed]
23
Yao
 
J.
Zheng
 
J.
Fu
 
W.
Zheng
 
J.
 
Extraction of anti-sepsis component from Terminaliachebula Retz and evaluation of their biological activities
Biosci. Rep.
2009
, vol. 
30
 (pg. 
27
-
32
)
[PubMed]
24
Ried
 
C.
Wahl
 
C.
Miethke
 
T.
Wellnhofer
 
G.
 
High affinity endotoxin-binding and neutralizing peptides based on the crystal structure of recombinant limulus anti-lipopolysaccharide factor
Biol. Chem.
1996
, vol. 
271
 (pg. 
28120
-
28127
)
25
Kim
 
S. H.
Kim
 
S.-A.
Park
 
M.-K.
Kim
 
S.-H.
Park
 
Y.-D.
Na
 
H.-J.
Kim
 
H.-M.
Shin
 
M.-K.
Ahn
 
K.-S.
 
Paeonol inhibits anaphylactic reaction by regulating histamine and TNF-a
Int. Immunopharmacol.
2004
, vol. 
4
 (pg. 
279
-
287
)
[PubMed]
26
Oh
 
G. S.
Pae
 
H. O.
Choi
 
B. M.
Jeong
 
S.
Oh
 
H.
Oh
 
C. S.
Rho
 
Y. D.
Kim
 
D. H.
Shin
 
M. K.
Chung
 
H.-T.
 
Inhibitory effects of the root cortex of Paeonia suffruticosa on interleukin-8 and macrophage chemoattractant protein-1 secretions in U937 cells
J. Ethnopharmacol.
2003
, vol. 
84
 (pg. 
85
-
89
)
[PubMed]
27
Kang
 
D. G.
Moon
 
M. K.
Choi
 
D. H.
Lee
 
J. K.
Kwon
 
T. O.
Lee
 
H. S.
 
Vasodilatory and anti-inflammatory effects of the 1,2,3,4,6- penta-O-galloyl- β-D-glucose (PGG) via a nitric oxide-cGMP pathway
Eur. J. Pharmacol.
2005
, vol. 
524
 (pg. 
111
-
119
)
[PubMed]
28
Liu
 
F.
Ng
 
T. B.
 
Antioxidative and free radical scavenging activities of selected medicinal herbs
Life Sci.
2000
, vol. 
66
 (pg. 
725
-
735
)
[PubMed]

Author notes

1

These authors contributed equally to this work.

2

Present address: The Clinical Medical Laboratories of People's Hospital of Jiangxi Province, Nanchang, Jiangxi, People's Republic of China.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC-BY) (http://creativecommons.org/licenses/by/3.0/) which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly cited.