Introduction

Clostridium septicum, an anaerobic gram-positive rod-shaped bacteria has played a significant role as a causative agent of many acute diseases in man and animals. One of the acute fetal infections of the bacterium is malignant edema or gas gangrene in all kinds of animals and humans (Chakravorty et al., 2015, de Assis et al., 2012, Pilehchian Lan- groudi, 2015, Thachil et al.,  2012).

C. septicum produces several extracellular toxins that include alpha-toxin, beta-toxin (DNase), gamma-toxin (hyaluronidase), and delta-toxin (thiol-activated toxin), protease and neuraminidase (Ballard et al.,1992). Al- pha- toxin, the main tool in the pathogenesis of C. septicum, is a hemolysin, necrotizing and lethal factor (Ballard et al.,1991, Bo- zorgkhoo et al., 2014, Lancto et al.,  2014).

Alpha-toxin is secreted as a pro-toxin that needs to be activated by a proteolytic cut in the carboxyl-terminal, and connects the cell membrane by glycosylphosphatidylinosi- tol receptors (GPI) (Gordon et al.,1999). A protease can release a 4.1 kDa peptide. Then the active form of the alpha-toxin with seven monomers is formed which makes the pore in the membrane, disturbs the ion balance of the cell, and causes cell lysis. Some prote- ases like trypsin, chymotrypsin, C. septicum secreted proteases, and furin (on the surface of eukaryotic cells) can activate alpha-toxin (Hang’ombe et al.,2007). Some commercial Elisa kits have been developed for identifi- cation of C. septicum (Thachil et al.,2013) but none of them are specific enough for al- pha-toxin detection. It is important to devel- op a short procedure to produce alpha-toxin as the unique known lethal factor of C. sep- ticum and its specific antibody (Vazquez-Ig- lesias  et  al.,2017).    In  our  previous  work

(2006), ten strains of C. septicum were iden- tified in Razi Vaccine and Serum Research Institute of Mashhad-Iran with alpha-toxin production capability. This study was de- signed to purify alpha-toxin of C. septicum and the antibody which could be used in di- agnosis kits, potency tests of the vaccines, and other related applications.

Material and Methods:

Preparation of Bacteria: C. septicum strains NH2 (NCBI: Eu482189) were isolat- ed and characterized previously from sheep dung (Hemmaty et al., 2005, Hemmaty et al., 2006). This strain shows high alpha-toxin production in comparison to other strains in our previous work.

Alpha-toxin production: Five ml of freshly prepared C. septicum strains NH2  was inoculated in 500 ml of liver broth (liv- er powder 1 g, peptone 15 g, glucose10 g, NaCl 1.25 g, pH =7.3) and incubated for 24  h at 37 °C in an anaerobic jar with gas pack type A (Merck). Biomass of the culture was removed by centrifugation at 5000 rpm for  15 min at 4 °C. The supernatant containing toxins was stored at -20 °C for further use.

Purification and characterization of al- pha-toxin: All purification steps were con- ducted at 4°C (Jonson and Ryden1989). This method was chosen because it is one of    the

most temperature sensitive toxins. The first purification step was done by adding 25 percent ammonium sulfate and spinning at 5000 rpm for 15  min.  Ammonium  sulfate up to 60% was added to the supernatant fol- lowed by centrifugation at 5000 rpm for 15 min. The pellet was dissolved in 20 ml of 50 mM Tris buffer (pH= 7.4) (using Tris, dH2O, HCl). The solution  was dialyzed  in 50  mM

Tris buffer with pH= 9.5 overnight (dialysis bag cut off 25 KDa) for removing ammoni- um sulfate residues and to change the buffer. In the next step, DEAE-Sephadex (Di- ethylaminoethyl Sephadex, Sigma) was used for ion-exchange chromatography (Adiguzel et al., 2015, Sato et al., 2012). Three ml of the last step’s prepared protein solution was passed from the column after equilibration with 50 mM Tris buffer  (pH=9.5) for 15- 20 min. Then the proteins were eluted with different concentrations of 5 ml NaCl from 0.1-0.6 M with 0.1 M intervals. Hemolysin and protein assays were performed on each fraction by hemolysis method and Bradford method respectively. The pool fractions of alpha-toxin were used for the next step. One percent of the previous fraction with higher hemolysis activity was applied to gel filtra- tion chromatography Sephadex G-50 (100 cm ×1 cm) with a flow rate of 0.2 ml/ min. The samples were collected by a fraction collector and the optical density of the frac- tions was measured at 280 nm. Each sample was tested for hemolysin activity with  rabbit

RBC.

The purification  process  was  analyzed  by SDS-PAGE electrophoresis. Two gel concentrations, stacking,  and  separating were used in discontinuous electrophoresis method.  Separating  gel  concentration   was

12.5 percent. 10 µl of sample was heated at 100 °C for 3 min pre-loaded with 3 µl sam- ple buffer (4x) accompanied with SDS and 2-mercaptoethanol. The sample was centri- fuged for removing any insoluble part of the sample. The Voltage of electrophoresis was 50 and 110 V in thickening and separating gel respectively. Bromophenol blue of buf- fer showed the finished sample at the end of the gel. The gel was fixed and finally stained with silver staining method.

Alpha-toxin assay

Five types of red blood cell (man, mice, rat, sheep, rabbit) were evaluated for sensitivity to alpha-toxin and one substrate was chosen. The hemolytic activity of alpha-toxin was determined with some modifications as de- scribed by different researchers (Ballard et al.,1995, Fatmawati et al., 2013, Uppalapati et al., 2013). Briefly, 200 µl of alpha-toxin sample was added to 1% washed rabbit red blood cell (RBC) in a total volume of 1ml of PBS (phosphate buffer saline) and incubated for 1 h at 37 °C. The reaction mixture was centrifuged for 5 min at 1000 rpm. The su- pernatant absorbency was measured by the spectrophotometer (Mitsubishi) at 565 nm for hemoglobin content. All of the tests were performed in duplicate tubes. Deionized wa- ter for the positive control (100% hemolysis) and phosphate buffer for negative control were added on RBC suspension. One hemo- lysin unit(HU) is equal to reduction of one percent hemolysis of RBC against control positive at one minute and one ml.

Antibody production

Alpha-toxin rabbit antibody production : As a recommended method for immunization of animal(Leenaars and Hendriksen,2005), two New Zealand healthy 6 month-old fe- male rabbits with 350-450 gr body weight were chosen. Alpha-toxin (250 µl) with complete Freund’s adjuvant (500 µl) was prepared and a total 750 µl emulsion was injected laterally over the thorax of animals subcutaneously. Two weeks later, alpha-tox- in (250 µl) with incomplete Freund’s adju- vant (500 µl) was injected subcutaneously (SC). The third injection was performed just 15 days after the second injection in the same manner.

Bleeding and serum collection: To eval- uate  the  antibody  levels,  rabbits  were bled

from their marginal vein three days after the last injection. Serum was separated with cen- trifugation and stored at 4 °C after sodium azide (0.1%) addition.

Purification of Antibodies

The following steps were performed for antibody purification:

Precipitation of total immunoglobulin: Ammonium sulfate (2.95 gr) was gently added to the rabbit serum (10 ml) with con- tinuous stirring in an ice bath, after  solving of ammonium sulfate the mixture  was kept at 4 ⁰C for overnight. The precipitated pro- teins were isolated by centrifugation at 3500 rpm for 20 min at 4 oC, and the supernatant was discarded. The sediment was dissolved in PBS( 10 ml, pH 7.4), and dialyzed over- night against PBS pH 7.4 overnight (dialysis bag cut off 25KDa) for removing ammonium sulfate and buffer changing.

Ion    exchange    chromatography:   DE- AE-Sephadex column was equilibrated with 100 mM Tris-base buffer. One ml of dialysis sample was applied to DEAE-Sephadex for 20 min and washed out with 0.2 ml/min flow rate. The outed fraction was collected and used for next step(Fathi Najafi et al., 2005) . Gel filtration chromatography: Sample was poured into gel, Sephadex G-50 column (100 cm × 1 cm) equilibrated with PBS buf- fer (pH =7.4). Fractions of the chromatogra- phy with 10 ml/h flow rate were collected in 4-ml  vials  and absorbency  was read  at 280

nm (Curling 1980).

Alpha-toxin  antiserum evaluation

Double immunodiffusion (DID): In this method,  a  2%  agarose  in  PBS  buffer (pH=7.3) was prepared and the gel was poured on slide (about 2 mm height) and allowed to solidify. The agarose gel was drilled to form small holes for the antigen or antibody. The purified  alpha-toxin  (20  µl)  in  the  central

cavity and the serums (20 µl) in the round cavities were loaded. For sediment forma- tion, the plates were incubated for 24 h at 37

°C in humid condition and then for 2 days at 4 °C.

Single radial immunodiffusion (SRID): According to the Standard Operating Proce- dure (SOP), agarose (2%) was melted using PBS buffer (pH =7.3) plus 0.05% (w/v) so- dium azide and the purified alpha-toxin an- tibody was added. Twenty microliters of a series of the diluted samples (1:2, 1:4, 1:8, 1:16) added into each rounded well were allowed to diffuse into the matrix and react with the antiserum for 24 h at room tem- perature. The precipitin zones were stained with Coomassie brilliant solution (Choi et al.,2017).

Electrophoresis: Purity of antiserum was analyzed from different stages of purification by SDS-PAGE according to the Laemmli method (Costa et al., 2013). Separating gel concentration was 12.5% and staking was prepared with 5%. Four volumes of purified samples of each fraction were added in one volume of sample buffer (5x) and the mix- ture was boiled for 5 min at 100 °C. To each well 15 µl of mixtures and to last well stan- dard molecular weight marker were loaded. Gel electrophoresis was performed at 120 voltage in concentrating gel and 100 volts for separating gel. Finally, the gel was stained with silver nitrate (Noshahri et  al.,2016).

Western blot

The Western blotting technique was used for alpha-toxin purification evaluation. At first SDS-PAGE electrophoresis was done with alpha-toxin samples and standard mo- lecular weight marker. Bands were trans- ferred to nitrocellulose membranes by wet transmission at 120 volts for 45 min. After saturation of nitrocellulose paper with a solution of bovine serum albumin 1% (BSA), the IgG conjugated with horseradish peroxidase (HRP) was added before substrate hydrogen peroxide (H2O2) and was finally stained by di-amino benzidine (Dabaghian et al.) to in- vestigate non-specific bands (Noshahri, Fathi Najafi, Kakhki, Majidi, Mehrvarz,  2016).

Results

Purification and characterization of Al- pha-toxin

According  to the study,  the maximum  al-

pha-toxin production of C. septicum was acquired at the end of the logarithmic phase before the stationary phase. Results on the alpha-toxin effect on different RBCs showed that alpha-toxin is able to lyse all five types of RBC (man, mice, rat, sheep, rabbit). How- ever, the hemolysis rate (percentage of RBC lysis) in rats, sheep, humans, mice, and rab- bits is 100, 97, 81, 62, and 55,    respectively

(Fig. 1).

Ammonium sulfate concentration results showed that the maximum amount of al- pha-toxin is deposited in 60% concentration. Ammonium  sulfate  in  25%  concentration is able to precipitate the non-specific pro- teins that are almost 40% of the total protein present  in  the  supernatant  of  cell   culture.

This step helps to remove large amounts of unwanted proteins and purification process will be facilitated (Table 1). Trace  amounts of ammonium sulfate were removed by Tris buffer dialyzing.

The results of ion exchange chromatog- raphy with NaCl gradient concentrations showed that 0.4M NaCl could dissociate  most of the alpha-toxin that binds to DEAE. The results of gel filtration chromatography, hemolysin test results, and absorbency at 280 nm are shown in Fig. 1. Alpha-toxin purifi- cation steps are summarized in Table  1.

Alpha-toxin with the special activity of 394 U/mg in the first culture medium reached to the special activity of 13666 U/mg at the last stage of purification. The purification meth- ods gave 35 times purity with 47% toxin re- covery. The alpha-toxin purification analysis by SDS-PAGE showed that the toxin has a molecular weight of about 48 KDa (Fig.  2).

Antibody Preparation

By  the  method  of  antibody   preparation

Figure 1.

Figure 2.

Table1. Purification table of Clostridium septicum NH2 alpha toxin

based on standard immunization pharma- copeia, the anti-toxin that was prepared and used as an antibody detected the C. septicum alpha-toxin. The  partially  purified  serum  by ammonium sulfate precipitation, ion ex- change, and gel filtration showed that puri- fied immunoglobulins had a good percent- age of purity and results were confirmed by SDS-PAGE, and Western Blot (Fig.  3).

Antitoxin evaluation

In DID test, alpha-toxin was poured into the central hole and the primary serum with dilutions of 1:2, 1:4, and 1:8 in the holes. The

high-density sediment line visibly appeared between the antigen and the initial serum after Coomassie blue staining. Also, pre- cipitation line formed between the holes of serum dilution of 1/2 and 1/4 were weaker. Increasing the antibody dilution in well 4 and the lack of observation of precipitation line with the dilution of 1/8 indicates un-equili- bration between Ab and Ag concentrations (Fig.3-A).

Figure 3-B show results of SRID test. The white halo is clearly displayed, indicating an antibody  production  against  the alpha-toxin in the rabbit.

Figure 3.

The results showed that the 48 KDa pu- rified alpha-toxin has high specificity to anti-toxin on Western Blot. The purified anti- body was of the best quality and the heavy and light chains of Ig displayed on SDS- PAGE (Fig. 2).

Discussion

In previous studies 9 local C. septicum strains were isolated and confirmed by bio- chemical tests, indirect immunofluorescence and PCR (Hemmati et al. 2006). The bacteri- al growth process and the level of alpha-tox- in secretion were investigated and then the hemolysin gene sequence  was  determined  in all strains and recorded in the Gene Bank (Fathi Najafi and Hemmaty  2007).

Based on present and previous studies (Ballard et al. 1991) the highest amount of toxins release is in the logarithmic phase which increases during growth progression before the stationary phase.

The effect of the alpha-toxin on the differ- ent RBCs showed that the toxin activity can be related to the type of animal; therefore,  the bacterium pathogenicity could be de- pendent on the host. The same results were reported by other researchers (Gordon et al. 1999, Hang’ombe et al. 2005, Hang’ombe et al. 2004). According to reports, GPI receptor and its level, as well as the amount of toxin tendency to the cellular receptor, can be ef- fective in alpha-toxin activity (Gordon et al. 1999, Mukamoto et al. 2013). In this study, the results showed that the most hemolytic activity of alpha-toxin was on RBCs of rat, sheep, man, mouse, and rabbits respectively. Hang’ombe et al showed that the alpha-tox- in adheres to a protein larger  than  100 kDa in the rat’s mucus membrane while on the mice, horses, cows, pigs, and poultry it    ad-

heres to proteins weighing 45-30 kDa. In Hang’ombe’s study, alpha- toxin  hemoly-  sis on rat’s blood was 86 times higher than sheep’s blood (Hang’ombe et al.  2005).

Alpha-toxin is produced as a pro-toxin weighting 46.5 KDa and needs to be sub- sequently cut to yield the active form (41.3 KDa)(Ballard et al. 1991). Research by Hang’ombe showed that  if the alpha-  toxin is treated by proteinase K, it will decrease activity against RBC hemolysis and other proteases treatment such as trypsin and chy- motrypsin will not significantly alter the he- molytic activity of alpha- toxin (Hang’ombe et al.,2005). The best activity of the toxin will be observed when the pro-toxin converts to the toxin near the cell surface. If alpha-toxin is activated earlier, its activity will be lost. It is oligomerized and cytolytically inactivated. In our study, no proteases’ treatment on al- pha-toxin was used prior to hemolysin test, similar to Ballard (Ballard et al. 1991).

Although alpha-toxin has the main role in the pathogenesis of the C. septicum, some other factors effect on the lethality by the bacterium (Kennedy et al. 2005). Ballard suggested that alpha-toxin has the role up to 70% of the C. septicum lethality. Therefore, the use of LD50 or MLD tests cannot be a standard assay for detection of the direct al- pha-toxin effect. In this study, the quantitative hemolysin test was used to assay alpha-tox- in during the purification steps according to Hang et al (Hang’ombe et al. 2005).

The immunization course and the type of the adjuvant were equal in the present and Hang’s study. Inactivation of  alpha  toxin was done by formalin in Hang’s study but in Ballard’s and our study, it wasn’t (Ballard et al. 1992, Hang’ombe et al. 2005).

Anion exchange and cation exchange chromatography, and gel filtration were  performed for alpha-toxin purification by Bal- lard et al (1992) in the first complete report procedure. Ballard found that during the al- pha-toxin purification some other factors in- fluence the hemolysin test such as a protein with 44 KDa weight and a larger 230 kDa protein. The 44-KDa protein is formed by removal of a part of the toxin due to the pro- teolytic effect of some secreted proteases by the bacterium. The 230 KDa protein is less when the alpha-toxin concentration is near to 2 mg/L and increases when the concentration reaches to 8 mg/ml. Therefore, it is a concen- tration-dependent protein that has much less hemolytic activity than in monomeric form. A phenomenon that probably consists of the aggregation of 5-6 monomer units  (Ballard et al. 1992). In our study, these proteins were not observed, which could be the result of faster purification process  and  differences  in purification steps and conditions, or the higher stability of the  alpha-toxin  secreted by the local C. septicum NH2 strain.