- Email: [email protected]
Direct selection of monoclonal antibodies neutralising the cytotoxic activity of Aeromonas sobria
FEMS MicrobiologyLetters 71 (1990) 311-314 Published by Elsevier
311
FEMSLE 04153
Direct selection of monoclonal antibodies neutr~!ising the cytotoxic activity of Aeromonas sobria Michael R. Barer
1
and Sally E. Millership
2
i Department of Microbiology. Unicersity of Newcastle upon l~vne, NewctL~tleupon l~vne. and " Department of Bacteriology; Royal Postgraduate Medical School. London, U.K.
Received25 May lqg0 Accepted 30 May 1990 Key words: Monoclonal antibody; Aeromonas sobria: Cytotoxin
1. S U M M A R Y
2. I N T R O D U C T I O N
Monoclonal antibodies directed against the cytotoxic activity of Aeromonas sobria were raised by immunising mice with a culture supernatant concentrated by ammonium sulphate precipitation. Neutralising anlibooies were specifically selected for by exposing hybridomas to cytotoxic levels of the immunising preparation, Cultures free from cytopathic effects after three hours were selected for further investigation. Ten cytotoxin resistant hybridomas were isolated but only two of these producer, detectable neutralising activity in Vero and rabbit red blood cell assays. Different polypeptide binding patterns were observed for the neutralising antibodies compared with the other antibodies in immunoblotting studies. One of the neutralising antibodies was shown to act at an early stage in the development of cytotoxicity, probably by inhibiting binding.
Aeromonas sobria belongs to the motile group of aeromonads amongst which human enteric and systemic pathogens are recognised. Although conclusive evidence for the contribution of aeromonas cytotoxins to the pathogenesis of natural infections is lacking, several studies have shown that the presence of such activity in culture supernatants from A. hydrophila and A. sobria is linked with pathogenic potential in animal test systems and in clinical isolates [1-3]. The cytotoxic activity itself may be due to more than one toxin [4] and several other putative virulence factors, including proteases and a cytotonic enterotoxin may also be involved in the pathogenesis of enteric disease [5,61. In order to study the contribution of cytotoxic activity to the virulence of A. sobria, murine hybridomas were prepared after immunisation with a crude concentrated culture supernatant. This paper reports the successful use of a strategy to specifically select for neutralising monoclonal antibodies I:roduced by these hybridomas and the effects of these antibodies on haemolytic and cytotoxic activities in culture supernatants.
Correspondence to: M.R. Barer, Department of Microbiology. University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne NE2 4HH, U.K.
0378-1097/90/$03.50 © 1990 Federation of European MicrobiologicalSocieties
312 3. MATERIALS A N D METHODS A concentrated culture supernatant was prepared from an 18 h 3 7 ° C Tryptone soy broth culture by ammonium sulphate precipitation (70%) as described previously [7] using strain A27, a clinical isolate of A. sobria from a patient with diarrhoea. This material was stored in 100 and 500 #1 aliquots at - 7 0 ° C until use. Cytotoxic and haemolytic activities were respectively determined in Vero cell and rabbit red blood cell microtitration assays as described previously [7]. One unit of activity was defined as the amount required to produce a 50% maximal effect in the assay system concerned. Cytotoxicity was also quantitated by staining cells with 1% Naphthol yellow S after exposure and quantitating bound dye in an ELISA reader [8]. The concentrated culture supernatant was used to immunise three female B A L B / c mice at 50 #g protein in Freund's complete adjuvant per dose, two subcutaneous doses 1 month apart followed by an intravenous dose in saline 6 weeks later. Fusions with S P 2 / O - A g l 4 cells were performed 72 h after the final immunising dose according to the method of Johnstone and Thorpe [9]. Wells containing growing hybridomas were screened for
neutralising activity at the end of the second week after fusion. The cytotoxic activity of the concentrated culture supernatant was titrated (2-fold dilutions) against a hybridoma producing an irrelevant antibody. The effects were observed at 3 h and the end-point determined as the dilution at which the cytopathic effect was last visible. The concentrate was then diluted in H A T medium to a point at which the addition of 50 #1 to wells containing putative cytotoxin specific hybridomas and 200/~1 cell culture medium exceeded the minimum amount required to be cytotoxic for the irrelevant hybridoma by between 4- and 8-fold. Wells containing cells showing no cytopathic effects were then harvested and replated into 24-well cell culture plates. When these cultures were confluent, hybrids were cloned by limiting dilution in m i o o titre plates. Wells producing stable growth were then expanded up to 2 cm 2 growth area in 24-well plates and supernatants from these wells were used in the further studies described. Neutralisation assays were performed either by determining the reduction in titrated activity resulting from the addition of 10 lal of test material (hybridoma supernatants or purified antibodies) to each assay well or by adding a fixed amount of
Fig. l. Micrographsof hybridomas after exposure to cytotoxic material showingprotection (A) and cytopathic effect (B) after 3 h, Original magnification x 400,
313 cytotoxic/haemolytic activity to dilutions of the test material and pre-incubating for 30 min at 3 7 ° C prior to addition to the indicator system. For the latter, dilutions of the concentra:cd culture superoatant were adjusted so that 4 haemolytic or cytotoxic units (pre-neutralisation) were added to the indicator system. Immunoblotting was carried out as described in ref. 9.
4. RESULTS A N D DISCUSSION The screening and subsequent cloning procedures yielded 10 hybridomas which were unaffected by the addition of concentrated cytotoxic supematant. Fig. 1 shows the cytopatltic effect seen after 3 h exposure compared with the appearance of unaffected (protected) cells. When supernatants from these cultures were examined for their ability to neutralise haemoiytic and cytotoxic activity against " t h i r d party" indicator cells
1
2
3
4
5
only two, designated A3 and FI2, showed detectable activity. Immunoblots prepared using A3, F12 and 5 of the other cultures are shown in Fig. 2. A3 and F12 recognised three major polypeptide bands (approx. Mrs 52000, 47000 and 43000) while the other five supernatants only produced clear reactions with one ( M r 52000) band. Minor reactions were also observed with the M r 43000 band (lanes 1 and 5). The discrepancies between neutralisation in the hybridoma cultures and third party neutralisation may be due either to different ep;~tope specificities of antibodies or to different toxin binding capacities in the supernatants examined. In the former case, antibodies expressed on the surface of the hybridoma might bind the toxin and prevent access to the site of action irrespective of whether active sites on the toxin were directly involved, while for the latter, differences in the quantity or affinity of the antibody concerned would be required. Studies on purified active cytotoxin from A. hydrophila have suggested Mrs in tbe range 48000 to 53000
6
7
8
9
10
Fig. 2. Immunoblots of concentrated A. sobria cytotoxic supernatant showing the binding pattern of "'third party" neutralising monoclonal antibodies (lanes 3 and 4), "third party" oon-neutralisinghybridoma supernatants (lanes l, 2, 5, 6 and 7), and a rabbit polyclonal serum (neutralising) (lane 8). Amido Black stained blots of the antigen and Mr standards (116000,66000. 45000 and 29000 from top to bottom) are shown in lanes 9 and I0 respectively.
314 F12 to assay wells at timed intervals after adding 3 units o f cytotoxic activity, yielded an exposure time-toxicity profile very similar to that obtained by washing cells free o f u n a b s o r b e d toxin at the s a m e times (Fig. 3). N o neutralising activity was detected after pre-incubating cells with m o n o clonal a n t i b o d y followed by washing and addition o f cytotoxin. These findings suggest that F12 acts by inhibiting the action o f high titre cytotoxic activity in A. sobria culture supernatants at an early stage, p r o b a b l y by toxin binding.
!
I_~ - ~ , ~ °0
tO
m
I"~ a0
4o
~
~o
Fig. 3. Survival of Veto cells after brief periods of exposure to
cytotoxic activity and after neutralisation of eytotoxicity by adding of F12 monoclonal antibody at timed intervals. The results were obtained in two parallel experiments. In the washing experiment 3 cytotoxic units were added to test monolayers for the times shown. All monolayers were then washed twice in cell culture medium and fresh medium added. Toxicity was assessed by staining with Naphthol yellow S after a further 24 h incubation period. In the FI2 experiment the monoclonal antibody was added to monolayer cultures previously exposed to 3 units of cytotoxic activity for the times indicated. Toxicity was again assessed after an additional 24 h incubation period. Points represent means of values obtained from four replicate monolayers. [1,2,10,11,12] with some evidence of the heterogeneity between different isolates [12]. The possible role o f proteolytic activation of a protoxin should also be considered [11]. F12 and A3 were further characterized as IgGtantibodies, and purified from large scale culture supernatants by protein A affinity chromatography. Both were shown to have neutralising activity against the haemolytic a n d cytotoxic activities in the A. sobria concentrated culture supernatant. F12 at a final concentration o f 0.014 m g / m l was able to completely neutralise the effects of diluted A. sobria supernatants containing up to 50 cytotoxic units of activity per ml. F12 showed 4-fold greater neutralising activity against haemolysis c o m p a r e d to cytotoxicity. Sub-neutralising levels of F12 were also found to retard cytotoxicity. Addition of neutralising a m o u n t s o f
ACKNOWLEDGEMENTS W e thank Dr. M.E. Devey for immunising the mice, Mr. G.F. M a n n for help with the cell fusion~, and A n i t a Singh and Clare Berry for technical assistance.
REFERENCES [11 Asao, T., Kinoshita, Y.. Uemera, T. and SakaguchL G. (1984) Infect. lmmun. 46, 122-127. [2] Grace),, M., Burke, V. and Robinson, J. (1982) Lancet 2. 1304-1306. [3l Millership, S.E., Barer, M.R. and Tabaqchali, S. (1986) J. Mud. Microbiol., 22, 311-314. [4] Ljungh, A., Wretlind, B. and Molby, R. (1981) Acta Pathol. Microbiol. Scand. Sect. B. 89, 387-397. [5] Wretlind, B., Heden, L, and Wadtstrom, T. (1973) J. Gen. MicrobioL 78, 57-65, [6l Chakraborty, T., Montenegro, M.A., Sanyal, S.C., Helmuth, R., Bulling, E. and Timmis, K.N. (1984) Infect. lmmun, 46, 435-441. [7] Barer, M.R., Millership, S.E. and Tabaqchali, S. (1986) J. Mud, Microbiol., 22, 303-309. [81 Barer, M.R,, Mann, G.F. and Drasar, B.S. (1986) Dee. Biol. Stand. 64, 251-259. [9] Johnstone, A. and Thorpe, R. (1987) lmmunochemistry in Practice, pp. 35-39, Blackwell, Oxford. [1(3] Bemheimer, A, Avigad, L.S..and Avigad, (3. (1975) Infect. lmmun, il, 1312-1319. [11] Howard, S.P. and Buckley, J.T. (1985) J. Bacteriol. 163, 336-340. [12] Rose, ,LM., Houston, C.W., and Kurosky, A. (1989) Infect.Immun, 57,1170-1176.
311
FEMSLE 04153
Direct selection of monoclonal antibodies neutr~!ising the cytotoxic activity of Aeromonas sobria Michael R. Barer
1
and Sally E. Millership
2
i Department of Microbiology. Unicersity of Newcastle upon l~vne, NewctL~tleupon l~vne. and " Department of Bacteriology; Royal Postgraduate Medical School. London, U.K.
Received25 May lqg0 Accepted 30 May 1990 Key words: Monoclonal antibody; Aeromonas sobria: Cytotoxin
1. S U M M A R Y
2. I N T R O D U C T I O N
Monoclonal antibodies directed against the cytotoxic activity of Aeromonas sobria were raised by immunising mice with a culture supernatant concentrated by ammonium sulphate precipitation. Neutralising anlibooies were specifically selected for by exposing hybridomas to cytotoxic levels of the immunising preparation, Cultures free from cytopathic effects after three hours were selected for further investigation. Ten cytotoxin resistant hybridomas were isolated but only two of these producer, detectable neutralising activity in Vero and rabbit red blood cell assays. Different polypeptide binding patterns were observed for the neutralising antibodies compared with the other antibodies in immunoblotting studies. One of the neutralising antibodies was shown to act at an early stage in the development of cytotoxicity, probably by inhibiting binding.
Aeromonas sobria belongs to the motile group of aeromonads amongst which human enteric and systemic pathogens are recognised. Although conclusive evidence for the contribution of aeromonas cytotoxins to the pathogenesis of natural infections is lacking, several studies have shown that the presence of such activity in culture supernatants from A. hydrophila and A. sobria is linked with pathogenic potential in animal test systems and in clinical isolates [1-3]. The cytotoxic activity itself may be due to more than one toxin [4] and several other putative virulence factors, including proteases and a cytotonic enterotoxin may also be involved in the pathogenesis of enteric disease [5,61. In order to study the contribution of cytotoxic activity to the virulence of A. sobria, murine hybridomas were prepared after immunisation with a crude concentrated culture supernatant. This paper reports the successful use of a strategy to specifically select for neutralising monoclonal antibodies I:roduced by these hybridomas and the effects of these antibodies on haemolytic and cytotoxic activities in culture supernatants.
Correspondence to: M.R. Barer, Department of Microbiology. University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne NE2 4HH, U.K.
0378-1097/90/$03.50 © 1990 Federation of European MicrobiologicalSocieties
312 3. MATERIALS A N D METHODS A concentrated culture supernatant was prepared from an 18 h 3 7 ° C Tryptone soy broth culture by ammonium sulphate precipitation (70%) as described previously [7] using strain A27, a clinical isolate of A. sobria from a patient with diarrhoea. This material was stored in 100 and 500 #1 aliquots at - 7 0 ° C until use. Cytotoxic and haemolytic activities were respectively determined in Vero cell and rabbit red blood cell microtitration assays as described previously [7]. One unit of activity was defined as the amount required to produce a 50% maximal effect in the assay system concerned. Cytotoxicity was also quantitated by staining cells with 1% Naphthol yellow S after exposure and quantitating bound dye in an ELISA reader [8]. The concentrated culture supernatant was used to immunise three female B A L B / c mice at 50 #g protein in Freund's complete adjuvant per dose, two subcutaneous doses 1 month apart followed by an intravenous dose in saline 6 weeks later. Fusions with S P 2 / O - A g l 4 cells were performed 72 h after the final immunising dose according to the method of Johnstone and Thorpe [9]. Wells containing growing hybridomas were screened for
neutralising activity at the end of the second week after fusion. The cytotoxic activity of the concentrated culture supernatant was titrated (2-fold dilutions) against a hybridoma producing an irrelevant antibody. The effects were observed at 3 h and the end-point determined as the dilution at which the cytopathic effect was last visible. The concentrate was then diluted in H A T medium to a point at which the addition of 50 #1 to wells containing putative cytotoxin specific hybridomas and 200/~1 cell culture medium exceeded the minimum amount required to be cytotoxic for the irrelevant hybridoma by between 4- and 8-fold. Wells containing cells showing no cytopathic effects were then harvested and replated into 24-well cell culture plates. When these cultures were confluent, hybrids were cloned by limiting dilution in m i o o titre plates. Wells producing stable growth were then expanded up to 2 cm 2 growth area in 24-well plates and supernatants from these wells were used in the further studies described. Neutralisation assays were performed either by determining the reduction in titrated activity resulting from the addition of 10 lal of test material (hybridoma supernatants or purified antibodies) to each assay well or by adding a fixed amount of
Fig. l. Micrographsof hybridomas after exposure to cytotoxic material showingprotection (A) and cytopathic effect (B) after 3 h, Original magnification x 400,
313 cytotoxic/haemolytic activity to dilutions of the test material and pre-incubating for 30 min at 3 7 ° C prior to addition to the indicator system. For the latter, dilutions of the concentra:cd culture superoatant were adjusted so that 4 haemolytic or cytotoxic units (pre-neutralisation) were added to the indicator system. Immunoblotting was carried out as described in ref. 9.
4. RESULTS A N D DISCUSSION The screening and subsequent cloning procedures yielded 10 hybridomas which were unaffected by the addition of concentrated cytotoxic supematant. Fig. 1 shows the cytopatltic effect seen after 3 h exposure compared with the appearance of unaffected (protected) cells. When supernatants from these cultures were examined for their ability to neutralise haemoiytic and cytotoxic activity against " t h i r d party" indicator cells
1
2
3
4
5
only two, designated A3 and FI2, showed detectable activity. Immunoblots prepared using A3, F12 and 5 of the other cultures are shown in Fig. 2. A3 and F12 recognised three major polypeptide bands (approx. Mrs 52000, 47000 and 43000) while the other five supernatants only produced clear reactions with one ( M r 52000) band. Minor reactions were also observed with the M r 43000 band (lanes 1 and 5). The discrepancies between neutralisation in the hybridoma cultures and third party neutralisation may be due either to different ep;~tope specificities of antibodies or to different toxin binding capacities in the supernatants examined. In the former case, antibodies expressed on the surface of the hybridoma might bind the toxin and prevent access to the site of action irrespective of whether active sites on the toxin were directly involved, while for the latter, differences in the quantity or affinity of the antibody concerned would be required. Studies on purified active cytotoxin from A. hydrophila have suggested Mrs in tbe range 48000 to 53000
6
7
8
9
10
Fig. 2. Immunoblots of concentrated A. sobria cytotoxic supernatant showing the binding pattern of "'third party" neutralising monoclonal antibodies (lanes 3 and 4), "third party" oon-neutralisinghybridoma supernatants (lanes l, 2, 5, 6 and 7), and a rabbit polyclonal serum (neutralising) (lane 8). Amido Black stained blots of the antigen and Mr standards (116000,66000. 45000 and 29000 from top to bottom) are shown in lanes 9 and I0 respectively.
314 F12 to assay wells at timed intervals after adding 3 units o f cytotoxic activity, yielded an exposure time-toxicity profile very similar to that obtained by washing cells free o f u n a b s o r b e d toxin at the s a m e times (Fig. 3). N o neutralising activity was detected after pre-incubating cells with m o n o clonal a n t i b o d y followed by washing and addition o f cytotoxin. These findings suggest that F12 acts by inhibiting the action o f high titre cytotoxic activity in A. sobria culture supernatants at an early stage, p r o b a b l y by toxin binding.
!
I_~ - ~ , ~ °0
tO
m
I"~ a0
4o
~
~o
Fig. 3. Survival of Veto cells after brief periods of exposure to
cytotoxic activity and after neutralisation of eytotoxicity by adding of F12 monoclonal antibody at timed intervals. The results were obtained in two parallel experiments. In the washing experiment 3 cytotoxic units were added to test monolayers for the times shown. All monolayers were then washed twice in cell culture medium and fresh medium added. Toxicity was assessed by staining with Naphthol yellow S after a further 24 h incubation period. In the FI2 experiment the monoclonal antibody was added to monolayer cultures previously exposed to 3 units of cytotoxic activity for the times indicated. Toxicity was again assessed after an additional 24 h incubation period. Points represent means of values obtained from four replicate monolayers. [1,2,10,11,12] with some evidence of the heterogeneity between different isolates [12]. The possible role o f proteolytic activation of a protoxin should also be considered [11]. F12 and A3 were further characterized as IgGtantibodies, and purified from large scale culture supernatants by protein A affinity chromatography. Both were shown to have neutralising activity against the haemolytic a n d cytotoxic activities in the A. sobria concentrated culture supernatant. F12 at a final concentration o f 0.014 m g / m l was able to completely neutralise the effects of diluted A. sobria supernatants containing up to 50 cytotoxic units of activity per ml. F12 showed 4-fold greater neutralising activity against haemolysis c o m p a r e d to cytotoxicity. Sub-neutralising levels of F12 were also found to retard cytotoxicity. Addition of neutralising a m o u n t s o f
ACKNOWLEDGEMENTS W e thank Dr. M.E. Devey for immunising the mice, Mr. G.F. M a n n for help with the cell fusion~, and A n i t a Singh and Clare Berry for technical assistance.
REFERENCES [11 Asao, T., Kinoshita, Y.. Uemera, T. and SakaguchL G. (1984) Infect. lmmun. 46, 122-127. [2] Grace),, M., Burke, V. and Robinson, J. (1982) Lancet 2. 1304-1306. [3l Millership, S.E., Barer, M.R. and Tabaqchali, S. (1986) J. Mud. Microbiol., 22, 311-314. [4] Ljungh, A., Wretlind, B. and Molby, R. (1981) Acta Pathol. Microbiol. Scand. Sect. B. 89, 387-397. [5] Wretlind, B., Heden, L, and Wadtstrom, T. (1973) J. Gen. MicrobioL 78, 57-65, [6l Chakraborty, T., Montenegro, M.A., Sanyal, S.C., Helmuth, R., Bulling, E. and Timmis, K.N. (1984) Infect. lmmun, 46, 435-441. [7] Barer, M.R., Millership, S.E. and Tabaqchali, S. (1986) J. Mud, Microbiol., 22, 303-309. [81 Barer, M.R,, Mann, G.F. and Drasar, B.S. (1986) Dee. Biol. Stand. 64, 251-259. [9] Johnstone, A. and Thorpe, R. (1987) lmmunochemistry in Practice, pp. 35-39, Blackwell, Oxford. [1(3] Bemheimer, A, Avigad, L.S..and Avigad, (3. (1975) Infect. lmmun, il, 1312-1319. [11] Howard, S.P. and Buckley, J.T. (1985) J. Bacteriol. 163, 336-340. [12] Rose, ,LM., Houston, C.W., and Kurosky, A. (1989) Infect.Immun, 57,1170-1176.