SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

We aimed to construct the DNA vaccine encoding Mycobacterium Tuberculosis (Mtb) dormancy antigen Rv1733c and investigate its immunogenicity in mice. The recombinant plasmid pcDNA-Rv1733c was transfected into P815 cells and its product was detected by indirect immunofluorescence. The mice were immunized once every 2 weeks by intramuscular injection of pcDNA-Rv1733c plasmid for a total of three times. The specific antibodies in the serum of the immunized mice were detected by enzyme-linked immunosorbent assay at the indicted time. Enzyme-linked immunosorbent spot was applied to determine the levels of IFN-γ, IL-2 and IL-4 secreted by splenic lymphocytes. Total cytotoxicity T lymphocyte (CTL) active of the splenic lymphocytes was detected by lactate dehydrogenase assay. Additionally, we analysed the percentages of CD4+ and CD8+ T cells in splenic lymphocytes using flow cytometry. The specific antibody was detected at 2 weeks after the first immunization, and the antibody titre was increased with time which was reached to 1:1600 at 8 weeks. The stimulation index of spleen lymphocytes and the levels of IFN-γ, IL-2 and IL-4 of pcDNA-Rv1733c-immunized mice were both higher than those of saline-immunized mice (< 0.05). However, no difference was found in the percentages of CD4+ and CD8+ T cells and the activity of CTL between the pcDNA-Rv1733c- and saline-immunized mice (> 0.05). So we got the conclusion that the plasmid pcDNA-Rv1733c DNA could induce specific humoral and cellular immunity in mice. Improving the immune effect of Rv1733c by several strategies, such as choosing appropriate immunization route and adjuvant, would be significant for Rv1733c as new tuberculosis vaccine.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Tuberculosis (TB) remains to be a global health problem, killing millions of people every year. Globally, there are 8 million new TB cases and 2 million deaths per year. Once infected, about 10% of cases will develop the active disease, usually within 1–2 years after exposure. Remaining individuals enter into a latency stage [latent tuberculosis infection (LTBI)], while the Mtb can be reactivate at a later stage, particularly if the individual becomes immunocompromised [1]. It is estimated that one-third of the world's population is latently infected with Mycobacterium Tuberculosis (Mtb) [2]. Therefore, controlling the latent Mtb infection could be important for outbreak investigation and disease control of TB [3].

Mtb mainly exists in the form of dormant bacillus during latent infection [4]. To adapt host environment and escape from host immune system, dormant tubercle bacilli transcribe and express 48 latency antigens which are regulated by the dormancy regulon (DosR) [3, 5]. These proteins, considered to be essential for bacteria survival during persistence in vivo, can be designed as therapeutic targets for humans in latent infection [6].

Specific immune response for latency antigens can be detected in patients latently infected with Mtb but not in Bacillus Calmette–Guerin (BCG)-vaccinated population. It is considered that taking the Mtb latency antigens as target antigen to design vaccine could not only gain the protective immune response for the latent Mycobacterium tuberculosis infection but could also clear the dormant bacillus and prevent the recurrence of TB [7]. Latency antigen Rv1733c is a protein that can be recognized by peripheral blood T lymphocytes in latent infection and can effectively stimulate T cells to secrete IFN-γ [8]. Riano et al. [9] also found that DosR Rv1737c, Rv2029c and Rv2628, and Rpf Rv2389c are able to induce significant levels of IFN-γ in latently infected individuals, mainly produced by T CD4+ cells with a central memory phenotype (CD45RO+CD27+), which may be associated with latency maintenance and protection against reactivation. A recent study suggested that Rv1733c, Rv0140 and Rv1009 induced significantly stronger IFN-γ responses in LTBI than in patients with TB [10]. Interestingly, IFN-γ responses to Rv2031, Rv1733c and Rv2626c antigens in BALB/c mice persistently infected with Mtb were also stronger than that in acutely infected mice. In this study, we constructed the Rv1733c eukaryotic expression vector and tested its immunology characteristic as DNA vaccine, which could lay the foundation for the research of TB candidate antigens.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

All animal studies were approved by the Animal Ethics Committee of the Fourth Military Medical University and performed in accordance with the ethical standards.

Materials

Eukaryotic expression vector pcDNA3.1(−), E. coli DH5α, P815 cell (mouse mastocytoma cell strain, H-2d) were kept in our laboratory. Recombinant plasmid pM18-T-Rv1733c was constructed and kept in our laboratory. Purified Rv1733c protein and the serum from Rv1733c protein-immunized mice were kept in our laboratory. Lipofectamine 2000 was purchased from Invitrogen, San Diego, CA, USA, G418 was from Roche, Basel, Switzerland), Plasmid extraction kit was from OMEGA, Norcross, GA, USA), CytoTox 96 Non-Radioactive Cytotoxicity Assay Kit was from Promega, Madison, WI, USA), Rat anti-mouse CD4: FITC/CD8: RPE dual colour Reagent was from AbD Serotec, Kidlington, UK) and sheep FITC-conjugated anti-mouse IgG was from Beijing Golden Bridge Biotechnology Company Ltd, Beijing, China). BALB/c male mice (6–8 weeks, 18–20 g per mouse) were provided by Laboratory Animal Center, The Fourth Military Medical University.

Methods

Construction of eukaryotic expression vector for Rv1733c. The genes encoding the full-length latency DosR-encoded antigen from Mtb Rv1733c were inserted into the pcDNA3.1 (−) eukaryotic expression vector with the restriction enzyme BamHI and HindIII. Positive pcDNA-Rv1733c clone was confirmed by sequencing. Then, the clone was propagated, and the plasmids were extracted, purified and stored at 20 °C until use.

Transfection of P815 cells with pcDNA-Rv1733c plasmid. Briefly, P815 cells were cultured in 24-well plate with the concentrations of 2 × 105 per well the day before transfection. The purified pcDNA-Rv1733c or pcDNA3.1 (−) plasmids were transfected into P815 cells using Lipofectamine 2000 (Invitrogen, Camarillo, CA, USA) according to the manufacturers' instruction when the cell growth reached up to 60–70%. G418 (400 mg/l) was added into the cells after transfection to select the pcDNA-Rv1733c-positive cells.

Detection of the expression of Rv1733c in P815 cells. The pcDNA-Rv1733c-positive cells were cultured in 24-well plate, and indirect immune fluorescence (IIF) assay was performed to determine the expression of Rv1733c after 24 h. The cells were blocked by goat serum for 30 min at room temperate and treated with the mice serum containing antibodies against Rv1733c at 4 °C overnight. After treated with FITC-IgG antibody and following stained with Evans blue, the cells were examined under a microscope.

Immunization of mice with pcDNA-Rv1733c plasmid. BALB/c mice were divided into three groups randomly (five mice per group) and were injected with 50 μl 0.25% bupivacaine hydrochloride 24 h before immunization. Mice of treatment groups were immunized by intramuscular injection with pcDNA-Rv1733c plasmid DNA (100 μg per mouse), while mice of positive control group were injected with BCG (1 × 105 CFU/0.1 ml per mouse). Mice of negative control group were treated with the same volume of normal saline. All mice were treated in the same way every 2 weeks for a total of three times. Sera were collected from the immunized mice 2 weeks post-vaccination and applied to detect the antibody levels. Mouse spleen lymphocytes were isolated at 4 weeks after the third immunization for detecting the proliferation of lymphocytes, the levels of IFN-γ, IL-2 and IL-4, the number of CD4+ and CD8+ T cells as well as the CTL (cytotoxic lymphocyte) activity.

Detection of the specific antibodies in the immunized mice. Enzyme-linked immunosorbent assay (ELISA) was utilized to detect the levels of specific antibodies in the serum of immunized mice. The ELISA plate was coated by the purified Rv1733c protein (10 μg/ml) overnight at 4 °C, which was then washed with PBS and blocked by 5% bovine serum albumin for 1 h at 37 °C. After washed with PBS, the plate was treated with the serum derived from the immunized mice as the primary antibody and horse reddish peroxidase IgG (1:2000; Zhongshan Co., Beijing, China) as the second antibody. o-Phenylenediamine as the substrate was added to each well for 10 min, and H2SO4 (2 m) was added to stop the reaction. The optical density was determined at 490 nm.

When detecting the antibody titre in BCG-immunized mice, we applied the tuberculin purified protein derivative (PPD) to coat the ELISA plate. To evaluate the ratio of IgG2a/IgG1, IgG2a or IgG1 was utilized as the second antibody.

Spleen lymphocyte proliferation test. Spleen lymphocytes were isolated using lymphocyte separation medium and diluted with RPMI-1640 to the density of 5 × 105/ml. Two hundred microlitre cells was cultured in each well (in triplicate) in the 96-well plate. The Rv1733c group and normal saline control group were both treated with 50 μl purified Rv1733c proteins (50 mg/l), while the BCG group was treated with PPD (25 mg/l). All groups were cultured in an incubator with 5% CO2 at 37 °C for 68 h. At last, all groups were treated with 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) (3-4, 5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium; 5 mg/ml, 20 μl per well) for 4 h, and 10% SDS (25 μl per well) was added to stop the reaction. The cells without protein stimulation were used as negative controls. The optical density was determined at 490 nm. Results were presented with OD (treatment group)/OD (control group).

Enzyme-linked immunosorbent spot (ELISPOT) detection of spleen lymphocytes secreting IFN-γ, IL-2 or IL-4. The ELISPOT assay was conducted as previously described [11]. Briefly, nitrocellulose 96-well plates were coated overnight with 100 μl per well of IFN-γ, IL-2 or IL-4 antibody (Mabtech, Stockholm, Sweden). The plates were washed eight times with sterile PBS, and the non-specific binding sites in the plates were blocked by incubation with RPMI 1640 containing 10% foetal calf serum (200 μl per well). Spleen lymphocytes (5 × 105 per well) were added into the plates and stimulated in the presence of Rv1733c (25 μg/ml) or PPD (25 μg/ml) for 48 h at 37 °C in an incubator with 5% CO2. The cells were then washed with PBS and incubated for 24 h at 4 °C with biotinylated anti-IFN-γ, IL-2 or IL-4 antibody. The plates were washed with PBS, and streptavidin–alkaline phosphatase (ALP) (ALP, 1:1000, 100 μl per well) was added to each well and incubated for 1 h at room temperature. 5-Bromo-4-chloro-3-indolyl phosphate substrate was applied to develop the spots, and the reaction proceeded for 15–30 min. Then, the plates were rinsed with distilled water and left to dry at room temperature. Spots were counted with an automated ELISPOT reader. All assays were performed in triplicate, and the results herein were reported as the mean of triplicate values.

CD4+T and CD8+T cell counting. Spleen lymphocytes (100 μl, 1 × 105/ml) were treated with the fluorescent-labelled FITC-CD4/RPE-CD8 antibody (10 μl) for 30 min in the dark. After centrifuged and washed, the cells were resuspended with paraformaldehyde (2%, 500 μl). The results were analysed using flow cytometry.

Cytotoxicity assay. CTL activity was detected by the lactate dehydrogenase (LDH) release assay following the kit specification (Promega) using the P815 cell line stable expressing Rv1733c as the targets, while P815 cell line stable expressing Hsp65 (heat-shock protein) was applied as the target cells when the CTL activity in BCG-immunized mice was evaluated. Percentage cytotoxicity was calculated by the formula: 100% × (release in assay − spontaneous release of effector cells − spontaneous release of target cells)/(maximum release of target cells − spontaneous release of target cells). Maximum release was determined by incubating 5 × 103 target cells with lysis buffer. Spontaneous release was measured by incubating 5 × 103 target cells or effector cells (spleen lymphocytes isolated from the immunized mice) with RPMI 1640. All assays were performed in triplicate, and the results herein were reported as the mean of triplicate values.

Statistical analysis. Statistical comparisons were conducted using the Student's t-test by spss12.0 software. A < 0.05 was considered as significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Establishment of P815 cell line

A total of 3 pcDNA-Rv1733c-positive P815 cell clones were selected by G418. One cell strain was selected randomly for determining the expression of Rv1733c by indirect immunofluorescence assay using Rv1733c-immunized mouse serum as the primary antibody. Specific immune fluorescence could be observed in pcDNA-Rv1733c-positive P815 cells, while no specific green fluorescence was detected in P815 cells without transfection (Fig. 1). The result indicated that the transfected P815 cells screened by G418 could express Rv1733c protein stably.

image

Figure 1. IIF assay of the expression of Rv1733c protein in pcDNA-Rv1733c stably transfected P815 cells. (A) The uninfected P815 cells; (B) P815 cells stably transfected with pcDNA-Rv1733c plasmid.

Download figure to PowerPoint

The level of specific antibody in the serum of immunized mouse

Specific antibody level was detected by the method of ELISA after immunization. Results showed that the antibody titre in pcDNA-Rv1733c-immunized group was increased with time and reached the highest level (1:1600) at 8 weeks after the first immunization. However, the antibody level (OD490 = 0.67 ± 0.13) was lower than that in BCG-immunized group (OD490 = 0.91 ± 0.05; t = 3.184, < 0.05) (Fig. 2A).

image

Figure 2. Analysis of the antibodies in the serum of immunized mice. (A) The levels of antibody detected by the method of ELISA after immunization. (B) Antibody subtype profiles of IgG2a and IgG1 in immunized mice. NS refers to normal saline control group.

Download figure to PowerPoint

The proportion of IgG2a/IgG1 was detected by ELISA. IgG2a was the main type in both pcDNA-Rv1733c-immunized group and BCG-immunized group at the earlier stage of immunization (Fig. 2B). With time going, the level of IgG2a was decreased, while the level of IgG1 was increased, and consequently, the ratio of IgG2a/IgG1 gradually reached a dynamic balance. The IgG2a/IgG1 ratio in Rv1733c-immunized mice was always slightly higher than that in BCG-immunized mice from the initiation to the end of immunization.

The proliferation of spleen lymphocyte in the immunized mice

At 4 weeks after the third immunization, the spleen lymphocytes were isolated and the cell viability was estimated by MTS proliferation assay. The result showed that the splenic lymphocyte of pcDNA-Rv1733c-immunized mice proliferated specifically when stimulated by Rv1733c protein. The stimulation index of pcDNA-Rv1733c-immunized group (2.00 ± 0.36) was higher than those of BCG-immunized group stimulated by Rv1733c protein (1.1 ± 0.06; t = 3.096, < 0.05), but lower than those of BCG-immunized group stimulated by PPD protein (2.6 ± 0.17; t = 4.154, < 0.05) (Fig. 3).

image

Figure 3. Specific proliferation of spleen lymphocyte of immunized mice. The pcDNA-1733 group refers to the spleen lymphocyte that was isolated from the pcDNA-Rv1733c-immunized mice and stimulated by purified Rv1733c protein. BCG/PPD refers to the spleen lymphocyte that was from the BCG-immunized mice and stimulated by PPD protein, while BCG/Rv1733 protein refers to the spleen lymphocyte that was from the BCG-immunized mice and stimulated by purified Rv1733c protein. NS refers to the spleen lymphocyte that was from normal saline control mice and stimulated by purified Rv1733c protein. *, VS. BCG/PPD group, < 0.05; #, VS. BCG/Rv1733 group, < 0.05.

Download figure to PowerPoint

The frequency of spleen lymphocyte secreting IFN-γ, IL-2 or IL-4 in the immunized mice

From Fig. 4A, we found that there were 41.48 ± 5.30 spot forming cells (SFC) in every 106 spleen lymphocytes in pcDNA-Rv1733c-immunized group, which was higher than that in normal saline control group (2.75 ± 1.37; t = 4.752, < 0.05) as well as that in BCG-immunized group stimulated by Rv1733c protein (7.25 ± 2.82; t = 3.151, < 0.05). Notably, the frequency of IFN-γ secreting cells in BCG-immunized mice stimulated by PPD protein was highest (104.25 ± 6.72) SFC/106), and clear difference was found when compared to that in pcDNA-Rv1733c-immunized mice (t = 5.841, < 0.01).

image

Figure 4. Numbers of cells secreting IFN-γ (A), IL-2 (B) or IL-4 (C) detected by ELISPOT in mice. The pcDNA-1733 group refers to the spleen lymphocyte that was isolated from the pcDNA-Rv1733-immunized mice and stimulated by purified Rv1733c protein. BCG/PPD refers to the spleen lymphocyte that was from the BCG-immunized mice and stimulated by PPD protein, while BCG/Rv1733 protein refers to the spleen lymphocyte that was from the BCG-immunized mice and stimulated by purified Rv1733c protein. NS refers to the spleen lymphocyte that was from normal saline control mice and stimulated by purified Rv1733c protein. *, VS. BCG/PPD group, < 0.01; #, VS. BCG/Rv1733 group, < 0.05.

Download figure to PowerPoint

Similar results were observed in the frequency of IL-2 and IL-4 secreting cells. As shown in Fig. 4B, C, the frequency of IL-2 and IL-4 secreting cells in BCG-immunized mice stimulated by PPD protein was highest [(73.25 ± 4.53) SFC/106 and (66.25 ± 4.06) SFC/106, respectively], followed by the pcDNA-Rv1733c-immunized mice [(11.25 ± 4.31) SFC/106 and (10.75 ± 6.12) SFC/106, respectively]. Significant difference was observed both in the frequency of IL-2 and IL-4 secreting cells between these two groups (< 0.01). The frequency of IL-2 and IL-4 secreting cells in normal saline-immunized mice [(2.25 ± 0.06) SFC/106 and (2.82 ± 0.14) SFC/106, respectively] and BCG-immunized mice stimulated by Rv1733c protein [(3.52 ± 2.16) SFC/106 and (3.82 ± 2.14) SFC/106, respectively] was lower than those in pcDNA-Rv1733c-immunized mice (< 0.05).

The percentage of CD4+ T and CD8+ T cell in the spleen lymphocyte of the immunized mice

The percentages of CD4+ T and CD8+ T cell in mouse spleen lymphocyte of pcDNA-Rv1733c-immunized group [(18.15 ± 2.30)% and (7.68 ± 1.34)%, respectively] were lower than those of BCG-immunized group [(35.82 ± 1.79)%; t = 3.571, < 0.05 and (18.82 ± 0.79)%; t = 2.864, < 0.05, respectively], but no obvious difference was found when compared with those of normal saline control group [(16.43 ± 2.02)%; t = 0.571, > 0.05 and (7.32 ± 0.42)%; t = 0.234, > 0.05, respectively] (Fig. 5).

image

Figure 5. The percentages of CD4+ T cells and CD8+ T cells in spleen lymphocyte of immunized mice. *, VS. BCG group, < 0.05; #, VS. saline control group, > 0.05. NS refers to normal saline control group.

Download figure to PowerPoint

Specific CTL activity of the spleen lymphocyte in immunized mice

Lactate dehydrogenase release assay was applied to detect the specific CTL activity in immunized mice. As clearly exhibited in Fig. 6, the CTL activity of the spleen lymphocyte in pcDNA-Rv1733c-immunized group was (29.52 ± 1.96)% when the ratio of effective cells to target cells was 100:1, which was lower than that of BCG-immunized group (51.56 ± 2.93)%; t = 4.234, < 0.05). There was no obvious difference in the CTL activity between pcDNA-Rv1733c-immunized group and normal saline control group (25.28 ± 2.51)%; t = 0.726, > 0.05) (Fig. 6).

image

Figure 6. CTL activity of the spleen lymphocyte in immunized mice. E/T ratio refers to the ratio of effect cells to target cells.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

Cellular immune response is the main protective immune response of organism against the infection with Mtb [12]. CD4+ T lymphocytes are the most studied immunocyte and also considered to be the most important cells that mediate anti-infection activity. As the study indicated IFN-γ and IL-2 could be secreted by Th1-type CD4+ T cell to activate mononuclear macrophages and promote its phagocytic activity [13]. By specific recognizing and killing the target cells, CD8+ T cells also play an important role in the immune response against Mtb infection [12].

Due to the advantage of easy producing, storing and transporting, DNA vaccines are widely used in the vaccine research field [14]. Immunization of mice with plasmid DNA encoding one of the secreted components of Mycobacterium tuberculosis, antigen 85 (Ag85), induced substantial humoral and cell-mediated immune responses and conferred significant protection against challenge with live Mtb and Mycobacterium bovis BCG [15]. It was found that dormant Mycobacterium tuberculosis was the main form during the incubation period. Dormant Mycobacterium tuberculosis can express specifically antigen related to its dormancy, which is important for the latent infection and the survival of Mtb [5, 16]. Rv1733c is a dormancy-related antigen that could be distinguished specifically by the peripheral blood T cells of patients with TB and can also stimulate T lymphocytes to release IFN-γ [8]. Thus, we construct the eukaryotic expression vector of Rv1733c and investigate its immunogenicity as DNA vaccine. The results showed that P815 cells transfected with recombinant plasmid pcDNA-Rv1733c could express Rv1733c protein which was demonstrated by the reaction between it and the specific sera containing Rv1733c antibody. It is suggested that the eukaryotic expression vector of pcDNA-Rv1733c was successfully constructed.

Humoral and cellular immune response was detected in the mice after immunized by the pcDNA-Rv1733c recombinant plasmid. The results indicated that antibodies against Rv1733c could be detected at 2 weeks after the first immunization, and the antibody titre of pcDNA-Rv1733c-immunized group reached their highest (1:1600) 8 weeks after immunization. However, the antibody level of pcDNA-Rv1733c-immunized group was lower than that of BCG-immunized group (t = 3.184, < 0.05). IgG2a which is one of the specific marks of Th1-type immune response was the main antibody type in the primary immune response both in DNA vaccine-immunized group and BCG-immunized group. But with the passage of time, IgG2a died down and reach an equilibrium state with IgG1. All above results showed that the specific humoral immunity could be stimulated and generated when mice was immunized with pcDNA-Rv1733c. However, the immunogenicity of pcDNA-Rv1733c was weaker than BCG, and the duration of Th1-type immune was shorter.

>The spleen lymphocyte of Rv1733c-immunized mice could proliferate specifically (< 0.05) when compared with the normal saline control mice. The levels of IFN-γ, IL-2 and IL-4 in Rv1733c-immunized mice were higher than those in normal saline controls (< 0.05). But no distinct difference was observed in the percentages of CD4+ T and CD8+ T cell and the CTL activity between mice immunized by Rv1733c and normal saline (> 0.05). The result suggested that weak cellular immune responses could be produced in Rv1733c-immunized mice. It is worth reminding that each parameter of cellular immune response of BCG-immunized group was higher than Rv1733c-immunized group (< 0.05). However, there was no specific reaction in mouse spleen lymphocyte of BCG-immunized group when stimulated by Rv1733c antigen (> 0.05). The above result indicated that BCG could not induce the specific immunity response to dormancy-related antigen Rv1733c, in agreement with previous report that BCG vaccination failed to induce the significant responses to latency antigens of Mtb [17]. Poor immune response against Mtb dormancy-related antigen was considered as the main reason for why BCG-immunized group cannot protect well against Mtb infection. Rv1733c is the Mtb dormancy-related antigen which is most easily recognized and can induce strong T cell immune response in mouse model and people persistently infected with Mtb [6, 18]. But in our study, the cellular immune response induced by Rv1733c DNA vaccine was weaker, which may be related to the immunization route, adjuvant, experimental animal and so on. In recent report, Bivas-Benita et al. [8] improved the cellular immunological response against DNA vaccine by alternating the immunization route and adjuvant. It also indicated that the selection of immunization route and adjuvant are important for immunity effect of vaccine.

In short, it is important to induce the organism to generate protective immune response against Mtb dormancy-related antigens for controlling the latent infection of Mtb. New DNA vaccine based on the Mtb dormancy-related antigen will make a significant sense in anti-infection with Mtb.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References

This project was supported by National Science and Technology Major Project (2012ZX10003008-007) and National Natural Science Foundation of China (30801055).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References