SEARCH

SEARCH BY CITATION

Keywords:

  • outer membrane protein;
  • immunogenicity;
  • ELISA ;
  • opsonophagocytosis assay;
  • positive selection

Abstract

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

Extraintestinal pathogenic Escherichia coli (ExPEC) is an important pathogen that can cause systemic infections in a broad spectrum of mammals and birds. To date, commercial vaccines against ExPEC infections in pigs are rare and antibiotic resistance has become a serious clinical problem. Identification of protective antigens is helpful for developing potentially effective vaccines. In this study, two outer membrane porins, OmpC and OmpF, of porcine ExPEC were cloned and expressed to investigate their immunogenicity. Intraperitoneal immunization of mice with the purified recombinant proteins OmpC and OmpF stimulated strong immunoglobulin G (IgG) antibody responses. Both IgG1 and IgG2a subclasses were induced, with a predominance of IgG1 production. After challenge with 2.5 × 107 CFU (5 × LD50) of the highly virulent ExPEC strain PCN033, 62.5% and 87.5% protection was observed in mice immunized with OmpC and OmpF, respectively. In addition, both anti-OmpC and anti-OmpF sera can mediate complement-dependent opsonophagocytosis. Phylogenetic analysis showed that the ompC gene was ubiquitously present in all E. coli strains, whereas the ompF gene was mutated in certain strains. Furthermore, the selection analysis indicated that gene ompC may be subject to strong immune pressure. Our results demonstrated that OmpC is a promising vaccine target against ExPEC infections in swine.


Introduction

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

Pathogenic Escherichia coli is an important zoonotic etiological agent that can infect a broad spectrum of mammals and birds. Pathogenic E. coli can be divided into two classes: intestinal and extraintestinal pathogenic E. coli (ExPEC) strains (Russo & Johnson, 2000). ExPEC strains possess certain specific virulence traits that enable them to invade and colonize extraintestinal sites and cause a wide range of infections, such as urinary tract infections, meningitis, pneumonia, osteomyelitis, and surgical site infections (Orskov & Orskov, 1985). Recent reports show that ExPEC has been isolated frequently from clinical samples in the pig industry in China (Tan et al., 2012). However, to date, the damage caused by ExPEC infections in swine has not been paid sufficient attention.

The two common approaches for prevention and therapy of bacterial diseases are vaccination and antibiotic therapy. Our recent study has demonstrated that antibiotic resistance is ubiquitously present in the porcine ExPEC strains isolated in China; 95.2% of which carried resistance to at least five antibiotics, and 60.3% were resistant to > 10 antimicrobials (Tang et al., 2011). Therefore, antibiotic treatment against ExPEC infections in pigs is limited. In addition, Tan et al. (2012) have reported that ExPEC infections are epidemic in China and have become a potential public health threat. It is desirable to find potential vaccine candidates to prevent this serious swine disease.

Outer membrane proteins are located at the bacterial surface and are likely to be important antigens that can induce intensive immune protection against infection. Many studies have reported certain immune features of outer membrane proteins from different bacterial species. The E. coli OmpX, which can induce a Th1/Th2 mixed humoral response, is considered to be an immunogenic protein (Maisnier-Patin et al., 2003). The OmpA of many bacteria exhibits immune properties, and it has been shown to activate dendritic cells and macrophages, produce cytokines, and elicit strong humoral responses (Torres et al., 2006; Pore et al., 2011).

Porins are a class of outer membrane proteins that are vital for Gram-negative bacteria, and often act as diffusion channels responsible for transport of certain hydrophilic nutrients (Benz, 1988). Porins are the main passage for many antibiotics and altered expression of these proteins is related to antibiotic resistance of bacteria (Pages et al., 2008). Porins are also involved in interactions with the host immune system (Massari et al., 2003). Two major porins, OmpC and OmpF from Salmonella enterica serovar Typhi, are immunogenic (Kumar et al., 2009). From the perspective of structure and function, the OmpF of E. coli has putative antigenic epitopes located on several loops (Klebba et al., 1990; Fourel et al., 1993), indicating that it may have some immune properties.

In the present study, the immunogenicity of OmpC and OmpF of porcine ExPEC was systemically evaluated and identified. Together with phylogenetic analysis, OmpC was inferred to be a novel protective antigen of porcine ExPEC and may serve as a promising vaccine candidate against ExPEC infection.

Material and methods

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

Bacterial strains

Nine ExPEC strains (listed in Table 1) were isolated from diseased pigs in Hubei province, China. All isolated strains together with E. coli strains DH5α and BL21 (DE3) were grown in Luria–Bertani (LB) medium and plated on LB medium containing 1.5% agar (w/v) at 37 °C.

Table 1. List of genes ompC and ompF present in the Escherichia coli strains used in this study
StrainOrigin ompC ompF
GenBank accession no.Length (bp)GenBank accession no.Length (bp)
  1. –, ompF gene was impaired in the strain due to mutation.

EcWH001Swine ADZ57175 1107 AFI41189 1089
EcWH009Swine ADZ57176 1107 AFI41190 1089
EcWH011Swine ADZ57177 1104 AFI41191 1089
EcWH030Swine ADZ57178 1095
EcWH044Swine ADZ57179 1095 AFI41192 1089
EcWH049Swine ADZ57180 1104
EcWH070Swine AEP03211 1107 AFI41193 1089
EcWH098Swine ADZ57181 1104 AFI41194 1089
EcWH142Swine ADZ57182 1104 AFI41195 1089
536Human ABG70254 1128 ABG68955 1089
CFT073Human AAN81212 1128 AAN79539 1089
S88Human CAR03644 1092 CAR02289 1089
UTI89Human ABE07964 1092 ABE06486 1089
UMN026Human CAR13736 1104 CAR12332 1089
IAI39Human CAR18478 1095 CAR18345 1074
HSHuman ABV06632 1107 ABV05386 1089
K12 MG1655Human AAC75275 1104 AAC74015 1089
SE11Human BAG78007 1104 BAG76512 1089
EC042Human CBG35289 1107 CBG33843 1074
55989Human CAU98339 1104 CAU96840 1089
EDL933Human AAG57350 1104 AAG55414 1089
SakaiHuman BAB36527 1104 BAB34435 1089
12009Human BAI31462 1104 BAI29823 1089
11368Human BAI26354 1089 BAI24372 1089
E2348Human YP_002329865 1116 YP_002328485 1071
CB9615Human ADD57300 1104 ADD55764 1089
E110019Human EDV88964 1104 ZP_03052581 1089
E24377AHuman ABV19634 1092 ABV20002 1089
H10407Human CBJ01853 1104 CBJ00505 1089
B7AHuman EDV64303 1089 ZP_03028823 1089
Sb227Human YP_408498 1119 ABB66790 1089
Sd197Human YP_402530 1125 ABB62405 1089
SF301Human NP_708110 1119 AAN42555 1089
SFV8401Human ZYP_689709 1119 YP_688458 1089
Ss046Human YP_311155 1104 AAZ87671 1089
K12 DH10BHuman ACB03377 1104 ACB02129 1089
UM146Human ADN70550 1092 ADN71947 1089
11128Human BAI26354 1092 BAI34952 1089
ABU83972Human ADN47057 1128 ADN45543 1089
NRG857cHuman ADR27664 1128 ADR26276 1089
IHE3034Human ADE89394 1092 ADE90312 1089
BW2952Human ACR63954 1104 ACR63953 1089
DH1 ME8569Human BAJ44007 1104 BAJ42736 1089
WHuman ADT75850 1104 ADT74541 1089
ED1aHuman CAR08862.2 1128 CAR07161 1089
APECChicken ABJ01608 1092 ABJ00346 1089
SMS-3-5EnvironmentACB16831095 ACB18257 1074

PCR amplification, cloning and sequencing

The following primer pairs were designed based on the consensus sequences of genes ompC and ompF from previously sequenced E. coli genomes: ompC-fw (5′-ATCGGGATCCATGAAAGTTAAAGTACTGTCCCTCC-3′), ompC-rev (5′-GGCGCTCGAGTTAGAACTGGTAAACCAGRCCMA-3′); ompF-fw (5′-ATCGGGATCCATGATGAAGCGCAATATTCT-3′), ompF-rev (5′-GGCGCTCGAGTTAGAACTGGTAAACGATAC-3′). The complete open reading frames of genes ompC and ompF were PCR-amplified using the boiled ExPEC strains as a template. PCR was carried out at 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 54 °C for 30 s and 72 °C for 1 min, and 72 °C for 7 min. The positive products were purified and cloned into an expression vector pET-28a with His-tag (Novagen, Shanghai, China) using the BamHI and XhoI sites. After transformation, the positive clone was sequenced with an ABI 3730 DNA sequencer (Applied Biosystems, Foster City, CA).

Expression and purification of recombinant proteins OmpC and OmpF

Two recombinant proteins, OmpC and OmpF, from ExPEC strain EcWH001 were overexpressed in BL21. Bacterial cells were grown to mid-log phase at OD600 nm= 0.8, then induced by adding 0.8 mM IPTG (Sigma, St. Louis, MO). After 4 h shaking at 37 °C, cells were harvested by centrifugation at 9300 g for 2 min at 4 °C. The precipitations were resuspended in 80 mL ice-cold buffer A containing 50 mM Tris base, 50 mM EDTA, 50 mM NaCl, 0.5 mM dithiothreitol and 5% glycerol, and then disrupted using a high-pressure cracker (JNBIO, Guangzhou, China). Protein purification was performed according to the method of Sambrook & Russell (2006). Protein concentration was measured with the Bicinchoninic Acid Protein Assay Kit (Beijing CellChip Biotechnology, China). The band position, molecular weight, and distribution of initial expression products and purified proteins were estimated by SDS-PAGE.

Western blot analysis

Western blotting was performed as described by Li et al. (2011). After the proteins were separated by SDS-PAGE and electrotransferred to polyvinylidene fluoride membranes, the membranes were blocked by 5% (w/v) nonfat dry milk in phosphate-buffered saline (PBS) overnight at 4 °C. The membranes were washed three times with TBST buffer (20 mM Tris–HCl, 150 mM NaCl, 0.05% Tween-20), and incubated with mouse anti-His antibody for 1 h, followed by Horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Southern Biotech, Birmingham, AL) diluted 1 : 5000 for 1 h. The membranes were developed using the DAB Horseradish Peroxidase Color Development Kit (Beyotime, Shanghai, China).

Mouse immunization and challenge

The immunization and challenge assay was performed in mice, based on the International Guiding Principles for Biomedical Research Involving Animals – 1985. A highly virulent ExPEC strain PCN033 (Tan et al., 2012) was chosen for challenge. Twenty-four female BALB/c mice (Hubei Center for Disease Control and Prevention, China) were evenly assigned to three groups. Mice in Groups 1 and 2 were injected intraperitoneally twice at 1-week intervals with 200 μL 50 μg purified OmpC and OmpF, respectively, mixed with 50% (v/v) Imject Alum adjuvant. Mice in Group 3 were injected with 50% (v/v) Imject Alum adjuvant in PBS as a control. Two weeks after the second injection, the immunized and control mice were challenged by intraperitoneal inoculation with 200 μL PBS containing 2.5 × 107 CFU of log-phase ExPEC PCN033. To determine antibody responses, sera were obtained by tail vein bleeding prior to each injection and challenge. The mortality in each group of mice was monitored daily for 7 days after challenge.

ELISA

Titers of recombinant protein-specific total IgG and two IgG subclasses (IgG1 and IgG2a) in mouse sera were examined by ELISA as described by Zhang et al. (2009). A 96-well plate was coated with purified products of 500 ng 100 μL−1 per well in sodium carbonate buffer overnight at 4 °C. The plate was washed three times with PBST (PBS supplemented with 0.05% Tween-20). After saturation with 0.5% nonfat dry milk in PBST for 2 h at 37 °C, the plate was washed three times with PBST and subsequently incubated with serially diluted mouse serum (initially in 1 : 100) for 30 min. The plate was washed three times with PBST and bound antibodies were detected by incubation with HRP-conjugated goat anti-mouse IgG antisera (Southern Biotech) for 1 h at 37 °C. After washing, the plate was developed using 3,3′,5,5′-tetramethylbenzidine as the HRP substrate. The reaction was terminated with the addition of 0.25% (w/v) hydrofluoric acid. Absorbance was measured at 630 nm in an ELISA reader (BioTek spectrophotometer, Winooski, VT). End-point titers were calculated as the reciprocal of the last serum dilution that was two-fold higher than the control.

Opsonophagocytosis assay

We isolated porcine neutrophils and investigated opsonophagocytosis based on the studies of Zhang et al. (2009). The number of viable cells was counted by trypan blue staining and adjusted to 4 × 106 cells mL−1 in Dulbecco's Modified Eagle's Medium. ExPEC PCN033 was grown to log phase and adjusted to 4 × 104 CFU mL−1. To facilitate interactions between bacteria and antibodies, bacterial cells were preincubated in 10% mouse serum at 37 °C for 30 min. The reaction mixture consisted of aliquots of 500 μL bacteria, 500 μL neutrophils and 100 μL healthy piglet serum as a complement source. The mixture was incubated at 37 °C for 1 h with rotation. After phagocytosis, the neutrophils were lysed with sterile water and serially diluted 10-fold. Dilutions were plated on LB plates and incubated at 37 °C overnight to determine viable counts. The sera from mice immunized only with adjuvant were used as a control. The efficiency of bacterial killing was estimated by the following formula: [1 − (number of bacteria recovered in presence of phagocytes/number of bacteria recovered in absence of phagocytes)] × 100% (Zhang et al., 2009).

Statistical analysis

Data of the opsonophagocytosis assay are summarized as mean ± standard deviation. The differences in antibody titers from the ELISA and the percentage of bacteria killed in the opsonophagocytosis assay were determined using the Mann–Whitney–Wilcoxon test. The significance cutoff was set to 0.01.

Phylogenetic analysis

The complete coding regions of E. coli porin genes ompC and ompF sequenced in the present and previous studies were collected to detect evidence of recombination and selective pressure (Table 1). Multiple sequence alignments were carried out for the translated protein sequences using the program t-coffee (Notredame et al., 2000). The aligned amino acid sequences were then mapped onto the corresponding codon sequences. A maximum likelihood phylogenetic tree was reconstructed using phyml (Guindon & Gascuel, 2003). Recombination events in our datasets were tested using the Single Break-Point (SBP) and the Genetic Algorithms for Recombination Detection (GARD) methods in the hyphy package (Pond et al., 2006). To infer selective pressure on the coding genes, the ratio of nonsynonymous substitutions (dN) to synonymous substitutions (dS) was estimated using the fixed effects likelihood (FEL) method via the Datamonkey webserver (http://www.datamonkey.org/) (Pond & Frost, 2005). A likelihood ratio test was conducted to infer whether a codon site was under selection. The significance threshold was set at 0.1. The sites subject to positive selection were mapped onto the 3D structural model predicted by the phyre server (Kelley & Sternberg, 2009). The topology of beta barrel outer membrane proteins was predicted by pred-tmbb (Bagos et al., 2004).

Results

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

Sequencing and expression of ompC and ompF

The complete coding sequences of outer membrane porin genes ompC and ompF were amplified from nine and seven porcine ExPEC strains, respectively. The GenBank accession numbers of these sequences are listed in Table 1. The length of genes ompC and ompF present in the porcine ExPEC strains were 1095–1107 bp and 1074–1089 bp, respectively. For ompF, a nonsense mutation was discovered in strain EcWH030 and a frameshift mutation in EcWH049. To express targeting proteins, two recombinant plasmids were constructed and designated as pOmpC and pOmpF, respectively. After induced expression, both recombinant proteins were present in the inclusion body. The results of SDS-PAGE and Western blotting showed that both the purified OmpC and OmpF proteins with a His-tag had a single band of approximately 40 kDa (Fig. 1), which was consistent with their theoretical molecular weight.

image

Figure 1. SDS-PAGE (a) and Western blotting (b) of purified recombinant proteins OmpC and OmpF.

Download figure to PowerPoint

Recombinant protein-specific IgG and IgG subclasses

The antibody titers against each recombinant protein in mouse sera were determined by ELISA. After the first immunization, the average specific IgG titer against recombinant OmpC was significantly higher in the vaccinated group than in the adjuvant control group (< 0.001). After the second immunization, the OmpC-specific IgG response was clearly enhanced (Fig. 2a). A similar result was observed in the OmpF-immunized group (Fig. 2c). Furthermore, high titers of IgG1 and IgG2a were induced in the OmpC-immunized mice, with the IgG1 titers higher than those of IgG2a (< 0.001) (Fig. 2b). In comparison with OmpC, higher titers of IgG1 and IgG2a were obtained with OmpF (Fig. 2d). For measurements of IgG titers against OmpC and OmpF, similar results have been observed for S. enterica serovar Typhi (Kumar et al., 2009).

image

Figure 2. Serum antibody responses in mice immunized with OmpC/OmpF plus adjuvant (solid circles) or adjuvant alone (open circles). (a) OmpC-specific total IgG. (b) OmpC-specific IgG subclasses (IgG1 and IgG2a) at 2 weeks after the second immunization. (c) OmpF-specific total IgG. (d) OmpF-specific IgG subclasses (IgG1 and IgG2a) at 2 weeks after the second immunization. The median of antibody titers for individual mice is shown as a line. *< 0.01; **< 0.001.

Download figure to PowerPoint

Protection of mice against ExPEC PCN033

The mice in the Group 3 adjuvant control group all died on the first day after challenge with the highly virulent ExPEC strain PCN033. Two of eight (25%) mice in Group 1 immunized with OmpC died on the first day after challenge and one died on the second day. The remaining mice in Group 1 survived for the following 5 days. One of eight (12.5%) mice in Group 2 immunized with OmpF died on the first day after challenge and the remaining mice survived (Fig. 3). This demonstrated that OmpC and OmpF provided 62.5% and 87.5% protection, respectively, against challenge with 2.5 × 107 CFU (5 × LD50) of ExPEC PCN033.

image

Figure 3. Survival of mice immunized with adjuvant alone (negative control), OmpC plus adjuvant, or OmpF plus adjuvant following challenge with 2.5 × 107 CFU (5 × LD50) of the highly virulent porcine ExPEC strain PCN033. Each group consisted of eight mice.

Download figure to PowerPoint

Opsonophagocytosis assay

Sera obtained from mice immunized with recombinant protein plus adjuvant or adjuvant alone were analyzed for their ability to promote opsonophagocytic killing of ExPEC strain PCN033 by porcine neutrophils. As shown in Fig. 4, 11.3 ± 2.6% of ExPEC strain PCN033 were killed in the absence of specific humoral response, whereas 70.5 ± 6.8% and 58.7 ± 4.9% of ExPEC strain PCN033 were killed with hyperimmune mouse sera against OmpC and OmpF, respectively. The results indicated that sera from both OmpC- and OmpF-immunized mice could mediate a significantly higher level of opsonophagocytic killing of ExPEC than sera from mice that received adjuvant alone.

image

Figure 4. Effect of antibodies on opsonophagocytic killing of Escherichia coli by porcine neutrophils. Anti-OmpC and anti-OmpF sera could mediate complement-dependent opsonophagocytosis, whereas only a few E. coli could be killed in the absence of specific humoral response. Data are expressed as percentage (mean ± standard deviation) of killed bacteria and are representative of five independent experiments. *< 0.01.

Download figure to PowerPoint

Recombination and selection analysis of ompC and ompF

The evidence for recombination signals was found in the E. coli ompC alignment by two programs, SBP and GARD, used for testing recombination. Three potential recombination breakpoints with significant phylogenetic incongruence were identified at the nucleotide positions 492, 744 and 981 in the alignment of ompC. Four non-recombinant alignments together with the relevant trees of topological congruity were generated for the subsequent selection analysis. Based on the FEL inference, the selection profile of the ompC coding region is illustrated in Fig. 5a. The porin showed significant evidence for positive selection, with seven codon sites (47, 189, 223, 237, 322, 324 and 325) under positively selected force. Structural mapping of these sites is shown in Fig. 5b. According to the predicted topology of OmpC by pred-tmbb, all positively selected sites were located in the extracellular space and outer membrane surface. In addition, 48 negatively selected sites were detected at the 0.1 significance level. For ompF, two recombination breakpoints were identified at the nucleotide positions 234 and 809 in the gene alignment. Notably, none of the positively selected sites was detected in the ompF gene through FEL inference.

image

Figure 5. Selection on OmpC-coding genes of Escherichia coli. (a) Selection profile of OmpC-coding genes of E. coli. The horizontal axis indicates the codon sites in the alignment. The ordinate stands for the value of (1 − P) for each site, which is indicated by the vertical line above and below the abscissa when dN/dS > 1 and dN/dS < 1, respectively. The dashed lines represent the significance level of 0.1. (b) Structural prediction of E. coli beta barrel porin OmpC. Orange circles indicate amino acid residues under positive selection.

Download figure to PowerPoint

Discussion

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

The E. coli genes ompC and ompF each encode an outer membrane protein. OmpC has a narrower pore and is preferentially expressed under higher osmolar pressure compared with OmpF (Nikaido, 2003). OmpC and OmpF have both been functionally confirmed to be beta barrel porins (Basle et al., 2006), which are important for dynamic interactions with the host immune system (Massari et al., 2003). Additionally, OmpC and OmpF are involved in antibiotic resistance and bacterial virulence (Negm & Pistole, 1999; De et al., 2001; Kumar et al., 2010). In this study, the immunogenic properties of porcine ExPEC OmpC and OmpF were investigated using a mouse model.

Both porins OmpC and OmpF of ExPEC can provide high protection against lethal infection with the highly virulent strain PCN033. In addition, OmpC and OmpF both could induce high titers of IgG antibodies, indicating that these two proteins have good immunogenic properties. The type of immune responses was reflected by the two IgG subclasses produced through immunization, IgG1 and IgG2a. In mice, serum IgG1 is associated with a Th2-type response, whereas serum IgG2a is associated with a Th1-type response, which is particularly effective at mediating bacterial opsonophagocytosis (Unkeless et al., 1988). Our study showed that OmpC and OmpF elicited high titers of IgG2a, although less than IgG1, which indicated that OmpC and OmpF could induce significant Th1/Th2 immune responses. To evaluate further the contribution of antibodies against OmpC and OmpF to bacterial killing, an opsonophagocytosis killing test was conducted. The results of this assay suggested that the antibodies against OmpC or OmpF were effective for mediating opsonophagocytosis of ExPEC. This may to some extent account for the high protection against challenge with highly virulent ExPEC in the immunized mice. To gain more insight into the mechanisms of immunogenicity and protective efficacy, the roles of OmpC and OmpF in macrophage adherence and cytokine production should be evaluated.

Based on further phylogenetic analysis, the ompC gene was found to be present in all E. coli strains, but ompF was mutated in certain strains. In addition, we found significant recombination signals in both alignments of ompC and ompF. Furthermore, the porin gene ompC showed significant evidence for positive selection in seven sites, whereas no positively selected sites were detected in ompF. The previous study on the genome-wide positive selection has reported that the E. coli ompC gene shows evidence for selective pressures exerted by phage infectivity (Petersen et al., 2007). Based on more publicly available sequences, we confirmed that E. coli ompC is undergoing strongly positive selection with an enlarged spectrum of positively selected sites identified. This might provide a genetic basis for further uncovering the interactions of the important outer membrane antigen OmpC with phage binding and/or with the host immune system.

In conclusion, we characterized the immunogenicity of OmpC and OmpF from porcine ExPEC. Our results indicated that surface-exposed outer membrane protein OmpC could be a promising candidate for vaccine development against ExPEC infection. Phylogenetic analysis further showed genetic evidence for positive selection acting on the porin gene ompC under host immunological pressure.

Acknowledgements

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

This study was supported by Grants from the National Natural Science Foundation of China (NSFC no. 31030065), and the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (31121004).

References

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