~(eterina~y Microbiology, 28 ( 199 | ) 61-73 Elsevier Science Publishers B.V., A m s t e r d a m
61
Immunological properties of Actinobacillus pleuropneumoniae hemolysin I Joachim Frey and Jacques Nicolet Institute for Veterinao' Bacteriology, University of Berne, L~inggasstrasse 122, (17f-3012 Berne, Switzerland (Accepted 19 December 1990)
ABSTRACT Frey, J. and Nicolet, J., 1991. Immunological properties of Actinobacillus pleuropneumoniae hemolysin I. ~k~t.Microbiol., 28: 61-73. The 105 kDa hemolysin 1 protein from Actinobacillus pleuropneumoniae serotype 1 type strain 4074 (HlyI) was shown by immunoblot analysis to be the predominant immunogenic protein if convalescent field sera or sera from pigs experimentally infected with A. pleuropneumoniae serotypc 1 were used. SDS gel- and immunoblot-analysis using total culture, washed cells or culture supernatant showed that Hlyl is essentially secreted and is not found attached to the bacteria. Proteins in the 105 kDa range that react strongly with anti-HlyI antibody, are produced by all serotypes and are presumed to be their hemolysins. Sera from pigs experimentally infected with each of the 12 serotypes strongly reacted with Hlyl. In addition, some sera from pigs that were confirmed to be negative for A. pleuropneumoniae, also reacted with Hlyl as well as with related proteins from Actinobacillus rossii and Actinobacillus suis. These two species produce proteins in the 105 kDa range which cross-react strongly with HlyI. They could be the source of the immunological reactions of the A. pleuropneumoniaenegative sera with Hlyl. However, no cross-reactions could be found between Hlyl and the Pasteurella haemolytica leukotoxin, the Escherichia coli alpha-hemolysin or related proteins from various hemolytic E. coli strains isolated from pigs. The immunological cross-reactions of HlyI wilh related proteins from A. rossii, A. suis and possibly from other bacterial species may create uncertainty in interpretation if Hlyl is used as the antigen in serodiagnosis ofA. pleuropneurnoniae.
INTRODUCTION
Actinobacillus pleuropneumoniae, the causative agent of swine pleuropneumoniae (Shope, 1964), secretes a hemolysin which is believed to play an important role in virulence (Kume et al., 1986; Rosendal et al., 1988; Frey and Nicolet, 1988b). The hemolysin of the type strain 4074 serotype 1, designated HlyI, is a monomeric protein with an apparent molecular weight of 105 kDa and an isoelectric point pI = 4.3 (Frey and Nicolet, 1988a,b ). Gene cloning and complementation experiments with the Escherichia coli and Proteus vulgaris hemolysin activator gene hlyC and the E. coli genes for hemolysin secretion hlyB and hlyD showed that HlyI is closely related to the E. coli al0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
62
J. F R E Y A N D J. N | C O L E T
pha-hemolysin and the P. vulgaris hemolysin (Gygi et al., 1990). Therefore, it belongs to a group of cytotoxins which are widely spread among Gramnegative human and animal pathogens. The hemolysin produced by serotype 2 (HlyII) was shown to differ from HlyI by the fact that its biosynthesis is not inducible by Ca 2+ (Frey and Nicolet, 1988b). The analysis of the 12 different serotypes ofA. pleuropneurnoniae revealed a heterogeneous picture of their hemolytic activities, most of them producing hemolytic activities similar or identical to HlyI or HlyII or both of them (Frey and Nicolet, 1990). Recently, Kamp and van Leengoed (1989) detected a cytotoxic activity in culture supernatants of all serotypes except serotype 6, but only in serotypes 1, 5, 9, 10 and 11 a hemolytic activity. This latter group of serotypes corresponds to those which are strongly hemolytic and which produce HlyI-like hemolysins (Frey and Nicolet, 1990). All serotypes were shown to produce a protein of 105 (104) kDa, supposedly the respective hemolysin, that immunologically strongly reacted with HlyI (Devenish et al., 1989; Frey and Nicolet, 1990). Using immunoblot analysis and neutralization assays, Devenish et al. ( 1989 ) found serological cross-reactions between the A. pleuropneurnoniae serotype 1 hemolysin protein and hemolysins/cytolysins of A. suis, P. haemolytica, and alpha-hemolysin producing E. coli. A structural relationship between the hemolysin of a serotype 5 strain and the E. coli alpha-hemolysin and the P. haemolytica leukotoxin was shown by comparing DNA sequence-derived aminoacid sequence analysis (Chang et al., 1989). An indirect enzyme-linked immunosorbent assay (ELISA), using culture supernatant ofA. pleuropneurnoniae serotype 5 strain J45 which showed to produce large amounts of Ca 2+ inducible hemolysin protein, was described to detect specifically sera from pigs infected with any of the 12 different A. pleuropneumoniae serotypes (Ma and Inzana, 1990). Contrary to this, Devenish et al. (1990) found many sera from pigs that were confirmed to be free ofA. pleuropneumoniae to give a positive reaction in a similar hemolysin capture ELISA. Moreover, they found that serum from pigs experimentally infected with A. suis produced antibodies detected by this ELISA which might be the source of unspecific cross-reactions. However, very little is known yet about the extent to which hemolysin contributes to the i m m u n e response of swine infected with A. pleuropneumoniae. There is not much information either about immunological cross-reactions with antigens (hemolysins ofcytolysins) from other bacterial species that are isolated from pigs. A better knowledge of such serological cross-reactions is required for the use of HlyI in the serodiagnosis of A. pIeuropneumoniae as well as for studying the role of HlyI in the pathogenesis of the diseases. HlyI is a very efficiently and specifically secreted protein (Frey and Nicolet, 1988a,b; Gygi et al., 1990); it is therefore assumed to be strongly immunogenic. The aim of this work was to determine the immunogenic property of HlyI from serotype 1 and the secreted hemolysins of the other serotypes in
IMMUNOLOGICAL PROPERTIES OF ACT1NOBAUILLUS PLEUROPNEUMONIAE HEMOLYSIN I
63
naturally and experimentally infected pigs. In addition, possible immunological cross-reactions with similar proteins from other bacteria isolated from pigs are analyzed. MATERIALS
AND METHODS
Bacterial strains. The bacterial reference strains used in this study are listed in Table 1. The field strains were isolated in our institute. Chemicals, growth media and growth conditions. All strains were grown in liquid medium: Columbia broth (BBL Microbiology Systems, Cockeysville, M D ) supplemented with 1% IsoVitaleX (BBL). For the growth ofA. pleuropneumoniae, the m e d i u m was supplemented with 10/~g/ml offl-NAD (Sigma Chem. Co., St. Louis, MO) and 10 m M CaC12. Hemolysin purification and titration. The purification of the 105 kDa HlyI protein and the titration of the hemolysin activities were described earlier (Frey and Nicolet, 1988a). Polyclonal antibodies and sera. Polyclonal antibodies directed against the purified HlyI ( 105 kDa protein) were produced in rabbits as described previously (Frey et al., 1989). The polyclonal antibodies against the E. coli alpha-hemolysin were obtained from T. Chakraborty, Wiirzburg, Germany. Serotype 1-positive field sera from infected pigs were obtained from S. Larivibre, Ste. Hyacinthe, Canada. Sera from pigs experimentally infected with the various A. pleuropneumoniae serotypes were obtained from N. Nielsen, Copenhagen, Denmark. Sera from healthy pigs of different ages from primary SPF herds were obtained from M. Kobisch, Ploufragan, France. Immunoblot analysis. Bacterial cultures were grown to mid-exponential phase at an A65o=0.8. Proteins from total cultures, from supernatants and from washed cells (washed in 10 m M Tris-HC1 pH 7.5) of these cultures were separated for subsequent immunoblot analysis by sodium dodecyl sulfate-polycrylamide gel electrophoresis SDS-PAGE ( l 0% acrylamide), as described by Laemmli (1970). Samples corresponding to 10/~1 of culture or 100 ng of purified protein were loaded per slot. Analysis by immunoblot technique was made as described by Towbin et al. (1979). The dilution of the pig sera was 1:200, and of the rabbit hyperimmunsera 1 : 2000. When indicated, sera were adsorbed with equal volumes of sonicated cultures of the corresponding bacteria grown to mid-exponential growth phase o f A 6 5 o = 0.8.
64
J. FREY AND J. NICOLET
TABLE 1 Bacterial strains Bacterial Strain
Strain Nr.
Serotype
Ref.
4074 S 1536 S 1421 M 62 K17 L 20 Femo W F 83 405 CV113261 13039 56153 8329 192 ( + A T C C 27072 ) C D C 62-57 C D C 63-57 RVC 2907
1 2 3 4 5a 5b 6 7 8 9 10 111 12
-'.3 -" 2 -"
Reference and test strains
A. pleuropneumomae .4. pleuropneumomae A. pleuropneumomae A. pleuropneumomae ,1. pleuropneumontae A. pleuropneumontae ,1. pleuropneumontae A. pleuropneumontae A. pleuropneumontae A. pleuropneumonlae A. pleuropneumomae :t. p[{'~rop~'lo~Ollla(" A. plettropnettmotllae A. rossii E. colt E. colt E. colt E. colt ~1. suis P. haemo@lica
A T C C 15558 P1148
O138:H OI39:K12:HI O141 :H4 O149
2,4 2,4 5 > 5 7 8 '~ ~o ~., e ~3 ~3 ~3 13 ~4
1
Field isolates
Pasteurella sp. ( hemolytic ) l'asteurella sp. ( hemolytic ) 4clinobacillussp. ( h e m o l y t i c ) 4ctinobacillus suis
$258 $911 S194/84 $623
~Erroneously described previously as serotype 10 a n d tentatively classified as serotype 11. References: 2Kilian el al., 1978: 3pohl el al., 1983; 4Nielsen, 1986a; SRosendal and Boyd, 1982; 6Nielsen and O ' C o n n o r , 1984; VNielsen, 1985a; 8Nielsen, 1985b; 9 K a m p et al., 1987; mNielsen, 1986b; ~ Ross et al., 1972; t2Sneath a n d Stevens, 1990; ~31.&F. Orskov, C o p e n h a g e n , D e n m a r k ; ~4Wetmore et al., 1963: ~5our laboratory.
RESULTS
Immune response to HlyI of A. pleuropneumoniae infected pigs The extent to which the 105 kDa HlyI protein contributes to the immune response of naturally infected pigs is demonstrated in Fig. 1. Aliquots corresponding to 10/~1 cultures ofA. pleuropneumoniae serotype 1 grown to midexponential phase, either from the total culture (including cells and culture medium) or the medium supernatant or the washed cells, were analyzed on two identical SDS polyacrylamide gels. One gel was blotted onto nitrocellu-
1M M U N O L O G I C A L P R O P E R T I E S O F ACT1NOBA(~TLLUS PLEUROPNEUMONIAE H E M O L Y S I N 1
65
Fig. 1. Immunogeneicity of HlyI. Coomassie Blue stained 10% polyacrylamide gel (left part), and immunoblot of an identical gel (right part) reacted with convalescent A. pleuropneumoniae serotype 1 positive swine serum (field serum). The lanes contained: a, purified Hlyl; b, proteins from 10/tl total broth culture ofA. pleuropneumoniae strain 4074 in exponential growth phase at A 650= 0.8 (equivalent of 6 × 108 cells); c, proteins from the supernatant of 10/~1 culture; d, proteins from pelleted cells of 10/~1 culture; pelleted and 2 × washed cells from 10 #1 culture; f, pelleted and washed cells from 100/~1culture. Lane s contained prestained (no. 161-0305; BioRad) molecular mass standards with the following apparent molecular masses; phosphorylase b, 110 kDa: bovine serum albumin 84 kDa; ovalbumin, 47 kDa; carbonic anhydrase, 33 kDa; soybean trypsin inhibitor, 24 kDa. The arrow indicates the position of Hlyl. lose filters for s u b s e q u e n t i m m u n o b l o t analysis with serum f r o m a pig experimentally infected with A. pleuropneumoniae serotype 1. Purified HlyI ( 100 ng) was also a d d e d to the gels for control. A strong i m m u n o l o g i c a l reaction with the serum was f o u n d with the purified H l y I protein (lane a). The comparison o f the relative signal strength o f the various proteins f r o m the total culture (lanes b) on the C o o m a s s i e Blue stained gel a n d on the i m m u n o b l o t (Fig. 1 ) shows t h a t the 105 k D a H l y I protein gives by far the strongest reactions on the i m m u n o b l o t although other proteins seem to be present in m u c h higher concentrations. The analysis o f the culture s u p e r n a t a n t (lane c) a n d of the w a s h e d cells (lanes e a n d f ) indicates t h a t the 105 k D a protein is f o u n d to a m a j o r extent in the culture s u p e r n a t a n t a n d is h a r d l y detectable in the w a s h e d cells. Similar results were o b t a i n e d with convalescent sera f r o m pigs naturally infected with A. pleuropneumoniae serotype 1. The i d e n t i t y o f the 105 k D a protein to be H l y I was verified by a second i m m u n o b l o t which was i n c u b a t e d with rabbit anti-HlyI (blot n o t s h o w n ) . T h e specificity o f the i m m u n e reaction with H l y I was first tested by m e a n s
66
J. FREY AND J. NICOLET
o f field sera f r o m s e r o t y p e 1-infected pigs a n d sera f r o m pigs o f d i f f e r e n t age f r o m a p r i m a r y S P F herd. T h e analysis was p e r f o r m e d on nitrocellulose stripes c o n t a i n i n g e i t h e r the full A. pleuropneumoniae s e r o t y p e 1 a n t i g e n s f r o m a total liquid c u l t u r e or p u r i f i e d H l y I p r o t e i n . Figure 2 s h o w s n i n e results e a c h r e p r e s e n t a t i v e o f p o s i t i v e a n d n e g a t i v e swine sera. All sera f r o m pigs i n f e c t e d with A. pleuropneumoniae s e r o t y p e 1 s h o w e d the s a m e r e a c t i o n as i l l u s t r a t e d in Fig. 1 ( l a n e a ) w h e r e the H l y I r e a c t i o n was p r e d o m i n a n t . O n the o t h e r
Total proteins from A.pleuropneurnoniaeserotype I
÷
+
÷
4-
+
÷
÷
+
+
½M 4M 6M 7M 12M 15M 18M 24M36M
Hemol~ysin I from A.pleuropneumoniaeserotype 1 Fig. 2. Immunoblot analysis ofA. pleuropneumoniae positive and negative field sera. Total liquid culture of A. pleuropneumoniae strain 4074 in exponential growth phase at ,465o=0.8 (equivalent of 6× 108 cells) (upper part) or purified HlyI (lower part) was separated on a 10% polyacrylamide gel and subsequently transferred to nitrocellulose filter. The filter was cut in strips which were then incubated with various A. pleuropneumoniae serotype 1 positive convalescent sera ( + ), or with sera from primary SPF pigs known to be negative for A. pleuropneumoniae at different ages (months: ½, 4, 6, 7, 12, 15, 18, 24, 36). The stripes contain proteins from 10/~1culture (upper) or 100 ng of purified Hlyl. The arrow indicates the position of Hlyl.
IMMUNOLOGICALPROPERTIESOF AC77NOBACILLUSPLEUROPNEUMONIAEHEMOLYSIN1
67
side, none of the sera from the SPF pigs showed any reaction with HlyI. Some minor reactions were observed only in the blots with other proteins. All sera from pigs infected experimentally with the different A. pleuropneu-
~=~,~-
,,,,=
,.,=
m
~
== ,
A
1 23
45a5b
H123 ~
~
~
4_~a
=,=
,,.,=
w
~
~
~
~,,~
6 7 8 91012
7 8 9101112
--
B
H 1 2 3 4 5a5b 6 7 8 9101112 ~,-
.,~, m
~
C
Fig. 3. Immunological cross-reactions of HlyI with 105 kDa proteins from serotypes 1-12. (A) Immunoblot stripes containing 100 ng HlyI reacted against sera from experimentally infected swine with A. pleuropneumoniae serotypes 1-12. (B) Immunoblot reacted with serum from a pig experimentally infected with A. pleuropneumoniae serotype 1. The different lanes contain supernatants from cultures in exponential growth phase at A65o--0.8 (equivalent of 6× l0 s cells) ofA. pleuropneumoniae serotypes 1-12. Lane H contains 100 ng Hlyl. (C) Immunoblot reacted with rabbit polyclonal anti HlyI serum. The different lanes contain supernatants from cultures in exponential growth phase at A65o= 0.8 (equivalent of 6 × l0 s cells) of A. pleuropneumoniae serotypes 1-12. Lane H contains 100 ng Hlyl. The arrow indicates the position of HlyI.
68
J. FREY AND J. NICOLET
moniae serotypes showed a similar and strong reaction with the 105 kDa HlyI protein on immunoblots (Fig. 3A). If total culture antigens from A. pleuropneumoniae serotypes 1 to 12 were analyzed on immunoblots with A. pleuropneumoniae serotype l-positive pig serum (Fig. 3B) or with rabbit anti-HlyI antibodies (Fig. 3C), a protein band comigrating with HlyI is seen to give a positive reaction. With the exception of serotype 3 that shows only a very faint band with the rabbit anti-HlyI antibodies, the intensity of the reaction varies in all serotypes. These results clearly indicate that all the A. pleuropneumoniae serotypes produce a protein in the 105 kDa range that cross-reacts immunologically with HlyI. The extent of cross-reaction seems to be similar and rather strong for all serotypes (Fig. 3A). However, the amount of the 105 kDa hemolysin protein produced seems to vary considerably among the different serotypes (Figs. 3B and C). The same result was obtained with silverstained polyacrylamide gels (results not shown). Immunological cross-reactions of HlyI with related proteins from other bacterial species In screening many sera from pigs of various geographical origin, known to be free ofA. pleuropneurnoniae infection, we detected some sera that showed
:ng
~ ~
~~
~ ~~
~
~
.
.
~oooo .
A
.
.
.
~;
. e~~~ . ~
~
.
.~ ~~
~
e ~~ ~
~N ® o . ~ '•N e~ e o o o o ~ ~
e e ~~} }~} e< ~ ~
~-
~
~~
o
B
Fig. 4. Immunological cross reactions. I m m u n o b l o t s containing proteins from total cultures at exponential growth phase at A65 o = 0.8 (equivalent of 6 × 108 cells) from various hemolytic bacterial strains as indicated and purified Hemolysin I; ( A ) reacted with a pig field serum that was confirmed to be negative for the k n o w n serotypes ofA. pleuropneumoniaebut showed positive reactions with Hlyl; ( B ) reacted with the same serum that was preadsorbed with HlyI. The arrow indicates the position of Hlyl.
69
I M M U N O L O G I C A L PROPERTIES OFAC77NOB.4CILLUSPLEUROPNEUMON1AE HEMOLYSIN 1
**
!-
-
~
~
-
~
oo
~
~
A
.
~
~
~
~
o ~ o-~a}~~. ~
~
~
o ,
~
~o
~
B
Fig. 5. Immunological differentiation of HlyI from E. colihemolysins. Immunoblots containing proteins from total cultures in exponential growth phase at A65o=0.8 (equivalent of 6× 108 cells ) from various hemolytic bacterial strains as indicated and purified Hemolysin I, (A) reacted wilh rabbit polyclonal anti-E, coli alpha-hemolysin, (B) reacted with rabbit polyclonal anti-Hlyl. The arrow indicates the position of HlyI. a strong reaction with HlyI, although they were confirmed to be negative for A. pleuropneumoniae by complement fixation assay. With total A. pleuropneumoniae antigens used, the patterns of the immunoblots of these negative sera differed clearly from the ones of the positive pig sera. They showed immune reactions mainly to the 105 k D a HlyI and marginally to the other proteins (Fig. 4A, last 2 lanes). In order to be able to find proteins possibly responsible for this unspecific reaction with HlyI, we analyzed on the same i m m u n o b l o t total culture proteins from several hemolytic bacterial species c o m m o n l y isolated from pigs (Fig. 4A). Clear reactions with proteins in the 105 kDa size range (supposed to be related hemolysins/cytotoxins) were seen with P. haemolytica serotypes 1, Actinobacillus fossil and E. coli O138, O 139 and O141, whereas the reaction with A. suis was weak (Fig. 4A). After these sera had been adsorbed with purified HlyI prior to the i m m u n o b l o t assay, the reactions with the 105 k D a proteins from A. pleuropneumoniae serotype 1 (HlyI), and from A. rossii, as well as the slight reaction with A. suis disappeared (Fig. 4B). However, the reactions remained with proteins of similar size ofP. haemolytica serotype 1 and of the different E. coli strains, indicating that these latter ones are not involved in the observed cross-reactions with HlyI. In order to confirm the supposed immunological relationship between HlyI and a protein produced by A. fossil (supposedly the hemolysin), i m m u n o b -
70
J. FREY AND J. NICOLET
lots with total culture proteins from the same bacterial species as shown in Fig. 4 were made, using polyclonal antibodies from rabbits against the E. coli alpha-hemolysin and against HlyI. By this means it was also possible to exclude any cross-reactions with proteins from the other bacterial species tested, in particular the hemolytic E. coli and P. haemolytica. The immunoblot with anti-E, coli alpha-hemolysin (Fig. 5A) shows reactions only with a protein in the 105 kDa range (supposedly the E. coli hemolysins) of the E. coli strains (O138, O139, O141, and O149). Hlyl and the related proteins of the other strains did not show any cross-reactions. On the other hand, when polyclonal antibodies against HlyI were used, reactions were seen only with HlyI and with a protein in the 105 kDa range ofA. rossii (Fig. 5B). A very faint reaction was observed with a 105 kDa A. suis protein. DISCUSSION
Some bacterial virulence factors, especially secreted toxins, are in general strongly immunogenic and may then be protective antigens. We have shown in our work that the secreted hemolysin protein HlyI ofA. pleuropneurnoniae, a supposed virulence factor, is a predominant immunogenic factor in naturally and experimentally infected pigs. Although immunoblot analysis is not able to give exact quantitative results, the use of many different convalescent sera showed that the reaction with the 105 kDa hemolysin protein is by far the strongest one. This was confirmed in spite of the fact that the portion of HlyI in the total proteins of the culture, which are used as antigens in the immunoblots, is rather low (Fig. 1, lanes b). We also showed that pelleted and washed cells of an A. pleuropneurnoniae serotype 1 culture is nearly devoid of the 105 kDa HlyI antigen, while all of this protein is found in the supernatant. Therefore, we conclude that HlyI is very efficiently secreted, and that no substantial amounts of HlyI are in the cytoplasm or cell bound. Using total culture proteins of the different serotypes and serum from pigs experimentally infected with serotype 1 strain or rabbit hyperimmune serum against HlyI, we have shown that the reference strains of all serotypes produce 105 kDa proteins that cross-react with HlyI. The extent of the production of these proteins, however, varies strongly among the different serotypes. The serotype 3 strain seems to produce very low amounts which are only seen when rabbit hyperimmune serum is used. All sera from pigs infected with the different serotypes react strongly with purified HlyI. This even happens with serotype 3 which produces low amounts of 105 kDa protein in liquid culture. Consequently, the 105 kDa proteins are strongly immunogenic and closely related to Hlyl, and are expected to be the hemolysins of the different serotypes. Biochemical evidence, however, indicates a certain heterology among these proteins (Frey and Nicolet 1990). Devenish et al. (1990) and Ma and Inzana (1990) used the hemolysin of
I M M U N O L O G I C A L PROPERTIES OF ACT1NOBACILLUS PLEUROPNEUMONIAE HEMOLYSIN I
71
A. pleuropneurnoniae for the immunological detection of all 12 serotypes of A. pleuropneumoniae in a indirect ELISA assay. Ma and Inzana (1990) described the assay to be highly specific. Devenish (1990) showed that an important portion of sera from pigs that were confirmed to be free ofA. pleuropneumoniae, gave a positive HlyI-ELISA reaction. They stated that these crossreactions could be due to the presence of antibodies against A. suis and possibly alpha-hemolysin producing E. coli. In addition, Devenish et al. ( 1989 ) noted the cross-reactions between the hemolysin from A. pleuropneumoniae strain CM-5 (serotype 1 ) and alpha-hemolysin from E. coli, the leukotoxin from P. haemolytica and an extracellular antigen produced by A. suis. Our results indicate that field sera from pigs known to be free of A. pleuropneumoniae show strong immunological reactions with HlyI as well as with proteins in the 105 kDa range (presumably the corresponding hemolysins/cytotoxins) from P. haemolytica, A. suis, A. rossii, and E. coli strains O141, O139, and O138 (Fig. 4A). Adsorption of these sera with HlyI prior to immunoblot analysis, however, shows that the observed immunological reactions of pig sera with P. haemolytica and the different E. coli strains are not due to a cross-reaction with HlyI. Moreover, in some field sera from pigs infected with A. pleuropneumoniae, there was no immunological reaction with P. haemolytica and only a weak one with E. coli (results not shown). On the other hand, A. rossii and A. suis produce an antigen in the 105 kDa range which strongly cross-reacts with HlyI (Fig. 4). These results were confirmed by the use of polyclonal antibodies in rabbits against HlyI or against E. coli alpha-hemolysin (Fig. 5 ). In addition, we did not see cross-neutralization between HlyI and alpha-hemolysin when using the polyclonal rabbit antibodies (results not shown). Contrary to the findings of Devenish et al. (1989), our results indicate that there is no substantial immunological relationship between HlyI ofA. pleuropneumoniae and the E. coli alpha-hemolysin. It is to be noted that the use of different strains ofA. pleuropneumoniae for the production of HlyI might be the source of the divergent results. However, it cannot be excluded that, due to the functional and structural relationship between the A. pleuropneumoniae hemolysins and the E. coli alpha-hemolysin (Chang et al. 1989; Gygi et al. 1990), these two hemolysins may share some common epitopes which do not appear to cause detectable cross-reactions in our conditions. We conclude that Hlyl is a predominant immunogenic protein of A. pleuropneurnoniae serotype 1. Immunologically related proteins (with similar or identical molecular weight), which are presumably the hemolysins of the other A. pleuropneumoniae serotypes, are also very strongly immunogenic. Some related bacterial organisms like A. rossii and A. suis produce proteins of 105 kDa that show strong immunological cross-reactions with HlyI. These cross-reactions have to be taken into consideration especially if HlyI is used in serodiagnosis and in the study of pathogenicity.
72
J. FREY A N D J. N [ C O L E T
ACKNOWLEDGEMENTS
We are grateful to S. Larivi6re, Ste. Hyacinthe, Canada, for the gift of convalescent field sera, to R. Nielsen, Copenhagen, Denmark, for the sera from experimentally infected pigs, to T. Chakraborty, Wiirzburg, Germany, for the polyclonal antiserum against E. coli alpha-hemolysin, and to R. Ross, Ames, USA, for the A. rossii strain. We also thank M. Krawinkler and A. V6gele for their skillful technical assistance and P. Wyssenbach for valuable comments on the manuscript. This work was supported by grant numbers 3.507.087 and 31-28401.90 from the Swiss National Science Foundation.
REFERENCES Chang, Y.-F., Young, R. and Struck, D.K., 1989. Cloning and characterization of a hemolysin gene from Actinobacillus (Haemophilus) pleuorpneumoniae. DNA 8: 635-647. Devenish, J., Rosendal, S., Johnson, R. and Hubler, S., 1989. Immunoserological comparison of 104-Kilodalton proteins associated with hemolysis and cytolysis in Actinobacillus pleuropneumoniae, Actinobacillus suis, Pasteurella haemolytica, and Escherichia coli. Infect. Immun., 57: 3210-3213. Devenish, J., Rosendal, S., Boss6, J.T., Wilkie, B.N. and Johnson, R,, 1990. Prevalence of seroreactors to the 104-Kilodalton hemolysin ofActinobacillus pleuropneumoniae in swine herds. J. Clin. Microbiol. 28: 789-791. Frey, J. and Nicolet, J., 1988a. Purification and partial characterization of a hemolysin produced by Actinobacillus pleuropneumoniae type strain 4074. FEMS Microbiol. Lett. 55:4146. Frey, J. and Nicolet, J., 1988b. Regulation of hemolysin expression in Actinobacillus pleuropneumoniae serotype 1 by Ca 2+. Infect. Immun. 56: 2570-2575. Frey, J. and Nicolet, J., 1990. Hemolysin patterns ofActinobacillus pleuropneumoniae. J. Clin. Microbiol. 28: 232-236. Frey, J., Perrin, J. and Nicolet, J., 1989. Cloning and characterization of a co-hemolysin, the CAMP factor ofActinobacillus pleuropneumoniae. Infect. Immun. 57: 2050-2056. Gygi, D., Nicolet, J., Frey, J., Cross, M., Koronakis, V. and Hughes, C., 1990. Isolation of the Actinobacillus pleuropneumoniae haemolysin gene and the activation and secretion of the prohaemolysin by the HIyC, HIyB and HIyD proteins ofEscherichia coli. Molec. Microbiol. 4: 123-128. Kamp, E.M., Pompa, J.K. and van Leengoed, L.A.M.G., 1987. Serotyping of Haemophiluspleuropneumoniae in the Netherlands: with emphasis on heterogeneity within serotype 1 and (proposed) serotype 9. Vet. Microbiol. 13:249-257. Kamp, E.M. and van Leengoed, L.A.M.G., 1989. Serotype-related differences in production and type of heat-labile hemolysin and heat-labile cytotoxin ofActinobacillus (Haernophilus) pleuropneumoniae. J. Clin. Microbiol. 27:1187-1191. Kilian, M., Nicolet, J. and Biberstein, E.L., 1978. Biochemical and serological characterization of Haemophilus pleuropneumoniae (Matthews and Pattison, 1961) Shope 1964 and proposal ofa neotype strain. Int. J. Syst. Bact. 28: 20-26. Kume, J., Nakai, T. and Sawata, A., 1986. Interaction between heat-stable hemolytic substance
I M M U N O L O G I C A L P R O P E R T I E S O F ACTINOBACILLUS PLEUROPNEUMONIAE H E M O L Y S I N 1
73
from Haemophilus pleuropneumoniae and porcine pulmonary macrophages in vitro. Infect. Immun. 51: 563-570. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227: 680-685. Ma, J. and Inzana, T.J., 1990. Indirect enzyme-linked immunosorbent assay for detection of antibody to a 110,000-molecular weight hemolysin of Actinobacillus pleuropneumoniae. J. Clin. Microbiol. 28: 1356-1361. Nielsen, R., 1985a. Serological characterization ofHaemophilus pleuropneurnoniae (Actinobacillus pleuropneumoniae) strains and proposal of a new serotype: serotype 9. Acta Vet. Scand. 26: 501-512. Nielsen, R., 1985b. Serological characterization ofHaemophilus pleuropneumoniae (Actinobacillus pleuropneumoniae) strains and proposal of a new serotype: serotype 10. Acta Vet. Scand. 26: 581-586. Nielsen, R., 1986a. Serology of Haemophilus (Actinobacillus) pleuropneurnoniae serotype 5 strains: establishment of subtypes A and B. Acta Vet. Scand. 27: 49-58. Nielsen, R., 1986b. Serological characterization of Actinobacilluspleuropneumoniae strains and proposal of a new serotype: serotype 12. Acta. Vet. Scand. 27: 453-455. Nielsen, R. and O'Connor, P.J., 1984. Serological characterization of 8 Haemophilus pleuropneumoniae strains and proposal of a new serotype: serotype 8. Acta Vet. Scand. 25: 96-106. Pohl, S., Bertschinger, U., Frederiksen, W. and Mannheim, W., 1983. Transfer of Ilaemophilus pleuropneumoniae and the Pasteurella haemolytica-like organism causing porcine necrotic pleuropneumonia to the genus Actinobacillus (Actinobacillus pleuropneumoniae com. nov. ) on the basis of phenotypic and deoxyribonucleic acid relatedness. Int. J. Syst. Bacteviol. 33: 510-514. Rosendal, S. and Boyd, D.A., 1982. Haemophilus pleuropneumoniae serotyping. J. Clin. Microbiol. 16: 840-843. Rosendal, S., Devenish, J., Maclnnes, J.l., Lumsden, J.H., Watson, S. and Xun, H., 1988. Evaluation of heat sensitive, neutrophil-toxic, and hemolytic activity of Haemophilus (Actinobacillus) pleuropneutnoniae. Am. J. Vet. Res. 49:1053-1058. Ross, R.F., Hall, J.E., Orning, A.P. and Dale, S.E., 1972. Characterization of an Actinobacillus isolated from the sow vagina. Int. J. Syst. Bacteriol. 22: 39-46. Shope, R.E., 1964. Porcine contagious pleuropneumoniae. I. Experimental transmission, etiology and pathology. J. Exp. Med. 119: 357-368. Sneath, P.H.A. and Stevens, M,, 1990. Actinobacillus rossii sp. nov., Actinobacillus seminis sp. nov., nom. rev., Pasteurella bettii sp. nov., Pasteurella lymphangitidis sp. nov., Pasteurella marii sp. nov., and Pasteurella trehalosi sp. nov. Int. J. Syst. Bacteriol. 40:148-153. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications Proc. Natl. Acad. Sci. U.s.A. 76: 4350-4354. Wetmore, P.W., Thiel, F.J., Herman, Y.F. and Harr, J.R., 1963. Comparison of selected Actinobacillus species with a haemolytic variety of Actinobacillus from irradiated swine. J. Infect. Dis. 113: 168-194.
61
Immunological properties of Actinobacillus pleuropneumoniae hemolysin I Joachim Frey and Jacques Nicolet Institute for Veterinao' Bacteriology, University of Berne, L~inggasstrasse 122, (17f-3012 Berne, Switzerland (Accepted 19 December 1990)
ABSTRACT Frey, J. and Nicolet, J., 1991. Immunological properties of Actinobacillus pleuropneumoniae hemolysin I. ~k~t.Microbiol., 28: 61-73. The 105 kDa hemolysin 1 protein from Actinobacillus pleuropneumoniae serotype 1 type strain 4074 (HlyI) was shown by immunoblot analysis to be the predominant immunogenic protein if convalescent field sera or sera from pigs experimentally infected with A. pleuropneumoniae serotypc 1 were used. SDS gel- and immunoblot-analysis using total culture, washed cells or culture supernatant showed that Hlyl is essentially secreted and is not found attached to the bacteria. Proteins in the 105 kDa range that react strongly with anti-HlyI antibody, are produced by all serotypes and are presumed to be their hemolysins. Sera from pigs experimentally infected with each of the 12 serotypes strongly reacted with Hlyl. In addition, some sera from pigs that were confirmed to be negative for A. pleuropneumoniae, also reacted with Hlyl as well as with related proteins from Actinobacillus rossii and Actinobacillus suis. These two species produce proteins in the 105 kDa range which cross-react strongly with HlyI. They could be the source of the immunological reactions of the A. pleuropneumoniaenegative sera with Hlyl. However, no cross-reactions could be found between Hlyl and the Pasteurella haemolytica leukotoxin, the Escherichia coli alpha-hemolysin or related proteins from various hemolytic E. coli strains isolated from pigs. The immunological cross-reactions of HlyI wilh related proteins from A. rossii, A. suis and possibly from other bacterial species may create uncertainty in interpretation if Hlyl is used as the antigen in serodiagnosis ofA. pleuropneurnoniae.
INTRODUCTION
Actinobacillus pleuropneumoniae, the causative agent of swine pleuropneumoniae (Shope, 1964), secretes a hemolysin which is believed to play an important role in virulence (Kume et al., 1986; Rosendal et al., 1988; Frey and Nicolet, 1988b). The hemolysin of the type strain 4074 serotype 1, designated HlyI, is a monomeric protein with an apparent molecular weight of 105 kDa and an isoelectric point pI = 4.3 (Frey and Nicolet, 1988a,b ). Gene cloning and complementation experiments with the Escherichia coli and Proteus vulgaris hemolysin activator gene hlyC and the E. coli genes for hemolysin secretion hlyB and hlyD showed that HlyI is closely related to the E. coli al0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
62
J. F R E Y A N D J. N | C O L E T
pha-hemolysin and the P. vulgaris hemolysin (Gygi et al., 1990). Therefore, it belongs to a group of cytotoxins which are widely spread among Gramnegative human and animal pathogens. The hemolysin produced by serotype 2 (HlyII) was shown to differ from HlyI by the fact that its biosynthesis is not inducible by Ca 2+ (Frey and Nicolet, 1988b). The analysis of the 12 different serotypes ofA. pleuropneurnoniae revealed a heterogeneous picture of their hemolytic activities, most of them producing hemolytic activities similar or identical to HlyI or HlyII or both of them (Frey and Nicolet, 1990). Recently, Kamp and van Leengoed (1989) detected a cytotoxic activity in culture supernatants of all serotypes except serotype 6, but only in serotypes 1, 5, 9, 10 and 11 a hemolytic activity. This latter group of serotypes corresponds to those which are strongly hemolytic and which produce HlyI-like hemolysins (Frey and Nicolet, 1990). All serotypes were shown to produce a protein of 105 (104) kDa, supposedly the respective hemolysin, that immunologically strongly reacted with HlyI (Devenish et al., 1989; Frey and Nicolet, 1990). Using immunoblot analysis and neutralization assays, Devenish et al. ( 1989 ) found serological cross-reactions between the A. pleuropneurnoniae serotype 1 hemolysin protein and hemolysins/cytolysins of A. suis, P. haemolytica, and alpha-hemolysin producing E. coli. A structural relationship between the hemolysin of a serotype 5 strain and the E. coli alpha-hemolysin and the P. haemolytica leukotoxin was shown by comparing DNA sequence-derived aminoacid sequence analysis (Chang et al., 1989). An indirect enzyme-linked immunosorbent assay (ELISA), using culture supernatant ofA. pleuropneurnoniae serotype 5 strain J45 which showed to produce large amounts of Ca 2+ inducible hemolysin protein, was described to detect specifically sera from pigs infected with any of the 12 different A. pleuropneumoniae serotypes (Ma and Inzana, 1990). Contrary to this, Devenish et al. (1990) found many sera from pigs that were confirmed to be free ofA. pleuropneumoniae to give a positive reaction in a similar hemolysin capture ELISA. Moreover, they found that serum from pigs experimentally infected with A. suis produced antibodies detected by this ELISA which might be the source of unspecific cross-reactions. However, very little is known yet about the extent to which hemolysin contributes to the i m m u n e response of swine infected with A. pleuropneumoniae. There is not much information either about immunological cross-reactions with antigens (hemolysins ofcytolysins) from other bacterial species that are isolated from pigs. A better knowledge of such serological cross-reactions is required for the use of HlyI in the serodiagnosis of A. pIeuropneumoniae as well as for studying the role of HlyI in the pathogenesis of the diseases. HlyI is a very efficiently and specifically secreted protein (Frey and Nicolet, 1988a,b; Gygi et al., 1990); it is therefore assumed to be strongly immunogenic. The aim of this work was to determine the immunogenic property of HlyI from serotype 1 and the secreted hemolysins of the other serotypes in
IMMUNOLOGICAL PROPERTIES OF ACT1NOBAUILLUS PLEUROPNEUMONIAE HEMOLYSIN I
63
naturally and experimentally infected pigs. In addition, possible immunological cross-reactions with similar proteins from other bacteria isolated from pigs are analyzed. MATERIALS
AND METHODS
Bacterial strains. The bacterial reference strains used in this study are listed in Table 1. The field strains were isolated in our institute. Chemicals, growth media and growth conditions. All strains were grown in liquid medium: Columbia broth (BBL Microbiology Systems, Cockeysville, M D ) supplemented with 1% IsoVitaleX (BBL). For the growth ofA. pleuropneumoniae, the m e d i u m was supplemented with 10/~g/ml offl-NAD (Sigma Chem. Co., St. Louis, MO) and 10 m M CaC12. Hemolysin purification and titration. The purification of the 105 kDa HlyI protein and the titration of the hemolysin activities were described earlier (Frey and Nicolet, 1988a). Polyclonal antibodies and sera. Polyclonal antibodies directed against the purified HlyI ( 105 kDa protein) were produced in rabbits as described previously (Frey et al., 1989). The polyclonal antibodies against the E. coli alpha-hemolysin were obtained from T. Chakraborty, Wiirzburg, Germany. Serotype 1-positive field sera from infected pigs were obtained from S. Larivibre, Ste. Hyacinthe, Canada. Sera from pigs experimentally infected with the various A. pleuropneumoniae serotypes were obtained from N. Nielsen, Copenhagen, Denmark. Sera from healthy pigs of different ages from primary SPF herds were obtained from M. Kobisch, Ploufragan, France. Immunoblot analysis. Bacterial cultures were grown to mid-exponential phase at an A65o=0.8. Proteins from total cultures, from supernatants and from washed cells (washed in 10 m M Tris-HC1 pH 7.5) of these cultures were separated for subsequent immunoblot analysis by sodium dodecyl sulfate-polycrylamide gel electrophoresis SDS-PAGE ( l 0% acrylamide), as described by Laemmli (1970). Samples corresponding to 10/~1 of culture or 100 ng of purified protein were loaded per slot. Analysis by immunoblot technique was made as described by Towbin et al. (1979). The dilution of the pig sera was 1:200, and of the rabbit hyperimmunsera 1 : 2000. When indicated, sera were adsorbed with equal volumes of sonicated cultures of the corresponding bacteria grown to mid-exponential growth phase o f A 6 5 o = 0.8.
64
J. FREY AND J. NICOLET
TABLE 1 Bacterial strains Bacterial Strain
Strain Nr.
Serotype
Ref.
4074 S 1536 S 1421 M 62 K17 L 20 Femo W F 83 405 CV113261 13039 56153 8329 192 ( + A T C C 27072 ) C D C 62-57 C D C 63-57 RVC 2907
1 2 3 4 5a 5b 6 7 8 9 10 111 12
-'.3 -" 2 -"
Reference and test strains
A. pleuropneumomae .4. pleuropneumomae A. pleuropneumomae A. pleuropneumomae ,1. pleuropneumontae A. pleuropneumontae ,1. pleuropneumontae A. pleuropneumontae A. pleuropneumontae A. pleuropneumonlae A. pleuropneumomae :t. p[{'~rop~'lo~Ollla(" A. plettropnettmotllae A. rossii E. colt E. colt E. colt E. colt ~1. suis P. haemo@lica
A T C C 15558 P1148
O138:H OI39:K12:HI O141 :H4 O149
2,4 2,4 5 > 5 7 8 '~ ~o ~., e ~3 ~3 ~3 13 ~4
1
Field isolates
Pasteurella sp. ( hemolytic ) l'asteurella sp. ( hemolytic ) 4clinobacillussp. ( h e m o l y t i c ) 4ctinobacillus suis
$258 $911 S194/84 $623
~Erroneously described previously as serotype 10 a n d tentatively classified as serotype 11. References: 2Kilian el al., 1978: 3pohl el al., 1983; 4Nielsen, 1986a; SRosendal and Boyd, 1982; 6Nielsen and O ' C o n n o r , 1984; VNielsen, 1985a; 8Nielsen, 1985b; 9 K a m p et al., 1987; mNielsen, 1986b; ~ Ross et al., 1972; t2Sneath a n d Stevens, 1990; ~31.&F. Orskov, C o p e n h a g e n , D e n m a r k ; ~4Wetmore et al., 1963: ~5our laboratory.
RESULTS
Immune response to HlyI of A. pleuropneumoniae infected pigs The extent to which the 105 kDa HlyI protein contributes to the immune response of naturally infected pigs is demonstrated in Fig. 1. Aliquots corresponding to 10/~1 cultures ofA. pleuropneumoniae serotype 1 grown to midexponential phase, either from the total culture (including cells and culture medium) or the medium supernatant or the washed cells, were analyzed on two identical SDS polyacrylamide gels. One gel was blotted onto nitrocellu-
1M M U N O L O G I C A L P R O P E R T I E S O F ACT1NOBA(~TLLUS PLEUROPNEUMONIAE H E M O L Y S I N 1
65
Fig. 1. Immunogeneicity of HlyI. Coomassie Blue stained 10% polyacrylamide gel (left part), and immunoblot of an identical gel (right part) reacted with convalescent A. pleuropneumoniae serotype 1 positive swine serum (field serum). The lanes contained: a, purified Hlyl; b, proteins from 10/tl total broth culture ofA. pleuropneumoniae strain 4074 in exponential growth phase at A 650= 0.8 (equivalent of 6 × 108 cells); c, proteins from the supernatant of 10/~1 culture; d, proteins from pelleted cells of 10/~1 culture; pelleted and 2 × washed cells from 10 #1 culture; f, pelleted and washed cells from 100/~1culture. Lane s contained prestained (no. 161-0305; BioRad) molecular mass standards with the following apparent molecular masses; phosphorylase b, 110 kDa: bovine serum albumin 84 kDa; ovalbumin, 47 kDa; carbonic anhydrase, 33 kDa; soybean trypsin inhibitor, 24 kDa. The arrow indicates the position of Hlyl. lose filters for s u b s e q u e n t i m m u n o b l o t analysis with serum f r o m a pig experimentally infected with A. pleuropneumoniae serotype 1. Purified HlyI ( 100 ng) was also a d d e d to the gels for control. A strong i m m u n o l o g i c a l reaction with the serum was f o u n d with the purified H l y I protein (lane a). The comparison o f the relative signal strength o f the various proteins f r o m the total culture (lanes b) on the C o o m a s s i e Blue stained gel a n d on the i m m u n o b l o t (Fig. 1 ) shows t h a t the 105 k D a H l y I protein gives by far the strongest reactions on the i m m u n o b l o t although other proteins seem to be present in m u c h higher concentrations. The analysis o f the culture s u p e r n a t a n t (lane c) a n d of the w a s h e d cells (lanes e a n d f ) indicates t h a t the 105 k D a protein is f o u n d to a m a j o r extent in the culture s u p e r n a t a n t a n d is h a r d l y detectable in the w a s h e d cells. Similar results were o b t a i n e d with convalescent sera f r o m pigs naturally infected with A. pleuropneumoniae serotype 1. The i d e n t i t y o f the 105 k D a protein to be H l y I was verified by a second i m m u n o b l o t which was i n c u b a t e d with rabbit anti-HlyI (blot n o t s h o w n ) . T h e specificity o f the i m m u n e reaction with H l y I was first tested by m e a n s
66
J. FREY AND J. NICOLET
o f field sera f r o m s e r o t y p e 1-infected pigs a n d sera f r o m pigs o f d i f f e r e n t age f r o m a p r i m a r y S P F herd. T h e analysis was p e r f o r m e d on nitrocellulose stripes c o n t a i n i n g e i t h e r the full A. pleuropneumoniae s e r o t y p e 1 a n t i g e n s f r o m a total liquid c u l t u r e or p u r i f i e d H l y I p r o t e i n . Figure 2 s h o w s n i n e results e a c h r e p r e s e n t a t i v e o f p o s i t i v e a n d n e g a t i v e swine sera. All sera f r o m pigs i n f e c t e d with A. pleuropneumoniae s e r o t y p e 1 s h o w e d the s a m e r e a c t i o n as i l l u s t r a t e d in Fig. 1 ( l a n e a ) w h e r e the H l y I r e a c t i o n was p r e d o m i n a n t . O n the o t h e r
Total proteins from A.pleuropneurnoniaeserotype I
÷
+
÷
4-
+
÷
÷
+
+
½M 4M 6M 7M 12M 15M 18M 24M36M
Hemol~ysin I from A.pleuropneumoniaeserotype 1 Fig. 2. Immunoblot analysis ofA. pleuropneumoniae positive and negative field sera. Total liquid culture of A. pleuropneumoniae strain 4074 in exponential growth phase at ,465o=0.8 (equivalent of 6× 108 cells) (upper part) or purified HlyI (lower part) was separated on a 10% polyacrylamide gel and subsequently transferred to nitrocellulose filter. The filter was cut in strips which were then incubated with various A. pleuropneumoniae serotype 1 positive convalescent sera ( + ), or with sera from primary SPF pigs known to be negative for A. pleuropneumoniae at different ages (months: ½, 4, 6, 7, 12, 15, 18, 24, 36). The stripes contain proteins from 10/~1culture (upper) or 100 ng of purified Hlyl. The arrow indicates the position of Hlyl.
IMMUNOLOGICALPROPERTIESOF AC77NOBACILLUSPLEUROPNEUMONIAEHEMOLYSIN1
67
side, none of the sera from the SPF pigs showed any reaction with HlyI. Some minor reactions were observed only in the blots with other proteins. All sera from pigs infected experimentally with the different A. pleuropneu-
~=~,~-
,,,,=
,.,=
m
~
== ,
A
1 23
45a5b
H123 ~
~
~
4_~a
=,=
,,.,=
w
~
~
~
~,,~
6 7 8 91012
7 8 9101112
--
B
H 1 2 3 4 5a5b 6 7 8 9101112 ~,-
.,~, m
~
C
Fig. 3. Immunological cross-reactions of HlyI with 105 kDa proteins from serotypes 1-12. (A) Immunoblot stripes containing 100 ng HlyI reacted against sera from experimentally infected swine with A. pleuropneumoniae serotypes 1-12. (B) Immunoblot reacted with serum from a pig experimentally infected with A. pleuropneumoniae serotype 1. The different lanes contain supernatants from cultures in exponential growth phase at A65o--0.8 (equivalent of 6× l0 s cells) ofA. pleuropneumoniae serotypes 1-12. Lane H contains 100 ng Hlyl. (C) Immunoblot reacted with rabbit polyclonal anti HlyI serum. The different lanes contain supernatants from cultures in exponential growth phase at A65o= 0.8 (equivalent of 6 × l0 s cells) of A. pleuropneumoniae serotypes 1-12. Lane H contains 100 ng Hlyl. The arrow indicates the position of HlyI.
68
J. FREY AND J. NICOLET
moniae serotypes showed a similar and strong reaction with the 105 kDa HlyI protein on immunoblots (Fig. 3A). If total culture antigens from A. pleuropneumoniae serotypes 1 to 12 were analyzed on immunoblots with A. pleuropneumoniae serotype l-positive pig serum (Fig. 3B) or with rabbit anti-HlyI antibodies (Fig. 3C), a protein band comigrating with HlyI is seen to give a positive reaction. With the exception of serotype 3 that shows only a very faint band with the rabbit anti-HlyI antibodies, the intensity of the reaction varies in all serotypes. These results clearly indicate that all the A. pleuropneumoniae serotypes produce a protein in the 105 kDa range that cross-reacts immunologically with HlyI. The extent of cross-reaction seems to be similar and rather strong for all serotypes (Fig. 3A). However, the amount of the 105 kDa hemolysin protein produced seems to vary considerably among the different serotypes (Figs. 3B and C). The same result was obtained with silverstained polyacrylamide gels (results not shown). Immunological cross-reactions of HlyI with related proteins from other bacterial species In screening many sera from pigs of various geographical origin, known to be free ofA. pleuropneurnoniae infection, we detected some sera that showed
:ng
~ ~
~~
~ ~~
~
~
.
.
~oooo .
A
.
.
.
~;
. e~~~ . ~
~
.
.~ ~~
~
e ~~ ~
~N ® o . ~ '•N e~ e o o o o ~ ~
e e ~~} }~} e< ~ ~
~-
~
~~
o
B
Fig. 4. Immunological cross reactions. I m m u n o b l o t s containing proteins from total cultures at exponential growth phase at A65 o = 0.8 (equivalent of 6 × 108 cells) from various hemolytic bacterial strains as indicated and purified Hemolysin I; ( A ) reacted with a pig field serum that was confirmed to be negative for the k n o w n serotypes ofA. pleuropneumoniaebut showed positive reactions with Hlyl; ( B ) reacted with the same serum that was preadsorbed with HlyI. The arrow indicates the position of Hlyl.
69
I M M U N O L O G I C A L PROPERTIES OFAC77NOB.4CILLUSPLEUROPNEUMON1AE HEMOLYSIN 1
**
!-
-
~
~
-
~
oo
~
~
A
.
~
~
~
~
o ~ o-~a}~~. ~
~
~
o ,
~
~o
~
B
Fig. 5. Immunological differentiation of HlyI from E. colihemolysins. Immunoblots containing proteins from total cultures in exponential growth phase at A65o=0.8 (equivalent of 6× 108 cells ) from various hemolytic bacterial strains as indicated and purified Hemolysin I, (A) reacted wilh rabbit polyclonal anti-E, coli alpha-hemolysin, (B) reacted with rabbit polyclonal anti-Hlyl. The arrow indicates the position of HlyI. a strong reaction with HlyI, although they were confirmed to be negative for A. pleuropneumoniae by complement fixation assay. With total A. pleuropneumoniae antigens used, the patterns of the immunoblots of these negative sera differed clearly from the ones of the positive pig sera. They showed immune reactions mainly to the 105 k D a HlyI and marginally to the other proteins (Fig. 4A, last 2 lanes). In order to be able to find proteins possibly responsible for this unspecific reaction with HlyI, we analyzed on the same i m m u n o b l o t total culture proteins from several hemolytic bacterial species c o m m o n l y isolated from pigs (Fig. 4A). Clear reactions with proteins in the 105 kDa size range (supposed to be related hemolysins/cytotoxins) were seen with P. haemolytica serotypes 1, Actinobacillus fossil and E. coli O138, O 139 and O141, whereas the reaction with A. suis was weak (Fig. 4A). After these sera had been adsorbed with purified HlyI prior to the i m m u n o b l o t assay, the reactions with the 105 k D a proteins from A. pleuropneumoniae serotype 1 (HlyI), and from A. rossii, as well as the slight reaction with A. suis disappeared (Fig. 4B). However, the reactions remained with proteins of similar size ofP. haemolytica serotype 1 and of the different E. coli strains, indicating that these latter ones are not involved in the observed cross-reactions with HlyI. In order to confirm the supposed immunological relationship between HlyI and a protein produced by A. fossil (supposedly the hemolysin), i m m u n o b -
70
J. FREY AND J. NICOLET
lots with total culture proteins from the same bacterial species as shown in Fig. 4 were made, using polyclonal antibodies from rabbits against the E. coli alpha-hemolysin and against HlyI. By this means it was also possible to exclude any cross-reactions with proteins from the other bacterial species tested, in particular the hemolytic E. coli and P. haemolytica. The immunoblot with anti-E, coli alpha-hemolysin (Fig. 5A) shows reactions only with a protein in the 105 kDa range (supposedly the E. coli hemolysins) of the E. coli strains (O138, O139, O141, and O149). Hlyl and the related proteins of the other strains did not show any cross-reactions. On the other hand, when polyclonal antibodies against HlyI were used, reactions were seen only with HlyI and with a protein in the 105 kDa range ofA. rossii (Fig. 5B). A very faint reaction was observed with a 105 kDa A. suis protein. DISCUSSION
Some bacterial virulence factors, especially secreted toxins, are in general strongly immunogenic and may then be protective antigens. We have shown in our work that the secreted hemolysin protein HlyI ofA. pleuropneurnoniae, a supposed virulence factor, is a predominant immunogenic factor in naturally and experimentally infected pigs. Although immunoblot analysis is not able to give exact quantitative results, the use of many different convalescent sera showed that the reaction with the 105 kDa hemolysin protein is by far the strongest one. This was confirmed in spite of the fact that the portion of HlyI in the total proteins of the culture, which are used as antigens in the immunoblots, is rather low (Fig. 1, lanes b). We also showed that pelleted and washed cells of an A. pleuropneurnoniae serotype 1 culture is nearly devoid of the 105 kDa HlyI antigen, while all of this protein is found in the supernatant. Therefore, we conclude that HlyI is very efficiently secreted, and that no substantial amounts of HlyI are in the cytoplasm or cell bound. Using total culture proteins of the different serotypes and serum from pigs experimentally infected with serotype 1 strain or rabbit hyperimmune serum against HlyI, we have shown that the reference strains of all serotypes produce 105 kDa proteins that cross-react with HlyI. The extent of the production of these proteins, however, varies strongly among the different serotypes. The serotype 3 strain seems to produce very low amounts which are only seen when rabbit hyperimmune serum is used. All sera from pigs infected with the different serotypes react strongly with purified HlyI. This even happens with serotype 3 which produces low amounts of 105 kDa protein in liquid culture. Consequently, the 105 kDa proteins are strongly immunogenic and closely related to Hlyl, and are expected to be the hemolysins of the different serotypes. Biochemical evidence, however, indicates a certain heterology among these proteins (Frey and Nicolet 1990). Devenish et al. (1990) and Ma and Inzana (1990) used the hemolysin of
I M M U N O L O G I C A L PROPERTIES OF ACT1NOBACILLUS PLEUROPNEUMONIAE HEMOLYSIN I
71
A. pleuropneurnoniae for the immunological detection of all 12 serotypes of A. pleuropneumoniae in a indirect ELISA assay. Ma and Inzana (1990) described the assay to be highly specific. Devenish (1990) showed that an important portion of sera from pigs that were confirmed to be free ofA. pleuropneumoniae, gave a positive HlyI-ELISA reaction. They stated that these crossreactions could be due to the presence of antibodies against A. suis and possibly alpha-hemolysin producing E. coli. In addition, Devenish et al. ( 1989 ) noted the cross-reactions between the hemolysin from A. pleuropneumoniae strain CM-5 (serotype 1 ) and alpha-hemolysin from E. coli, the leukotoxin from P. haemolytica and an extracellular antigen produced by A. suis. Our results indicate that field sera from pigs known to be free of A. pleuropneumoniae show strong immunological reactions with HlyI as well as with proteins in the 105 kDa range (presumably the corresponding hemolysins/cytotoxins) from P. haemolytica, A. suis, A. rossii, and E. coli strains O141, O139, and O138 (Fig. 4A). Adsorption of these sera with HlyI prior to immunoblot analysis, however, shows that the observed immunological reactions of pig sera with P. haemolytica and the different E. coli strains are not due to a cross-reaction with HlyI. Moreover, in some field sera from pigs infected with A. pleuropneumoniae, there was no immunological reaction with P. haemolytica and only a weak one with E. coli (results not shown). On the other hand, A. rossii and A. suis produce an antigen in the 105 kDa range which strongly cross-reacts with HlyI (Fig. 4). These results were confirmed by the use of polyclonal antibodies in rabbits against HlyI or against E. coli alpha-hemolysin (Fig. 5 ). In addition, we did not see cross-neutralization between HlyI and alpha-hemolysin when using the polyclonal rabbit antibodies (results not shown). Contrary to the findings of Devenish et al. (1989), our results indicate that there is no substantial immunological relationship between HlyI ofA. pleuropneumoniae and the E. coli alpha-hemolysin. It is to be noted that the use of different strains ofA. pleuropneumoniae for the production of HlyI might be the source of the divergent results. However, it cannot be excluded that, due to the functional and structural relationship between the A. pleuropneumoniae hemolysins and the E. coli alpha-hemolysin (Chang et al. 1989; Gygi et al. 1990), these two hemolysins may share some common epitopes which do not appear to cause detectable cross-reactions in our conditions. We conclude that Hlyl is a predominant immunogenic protein of A. pleuropneurnoniae serotype 1. Immunologically related proteins (with similar or identical molecular weight), which are presumably the hemolysins of the other A. pleuropneumoniae serotypes, are also very strongly immunogenic. Some related bacterial organisms like A. rossii and A. suis produce proteins of 105 kDa that show strong immunological cross-reactions with HlyI. These cross-reactions have to be taken into consideration especially if HlyI is used in serodiagnosis and in the study of pathogenicity.
72
J. FREY A N D J. N [ C O L E T
ACKNOWLEDGEMENTS
We are grateful to S. Larivi6re, Ste. Hyacinthe, Canada, for the gift of convalescent field sera, to R. Nielsen, Copenhagen, Denmark, for the sera from experimentally infected pigs, to T. Chakraborty, Wiirzburg, Germany, for the polyclonal antiserum against E. coli alpha-hemolysin, and to R. Ross, Ames, USA, for the A. rossii strain. We also thank M. Krawinkler and A. V6gele for their skillful technical assistance and P. Wyssenbach for valuable comments on the manuscript. This work was supported by grant numbers 3.507.087 and 31-28401.90 from the Swiss National Science Foundation.
REFERENCES Chang, Y.-F., Young, R. and Struck, D.K., 1989. Cloning and characterization of a hemolysin gene from Actinobacillus (Haemophilus) pleuorpneumoniae. DNA 8: 635-647. Devenish, J., Rosendal, S., Johnson, R. and Hubler, S., 1989. Immunoserological comparison of 104-Kilodalton proteins associated with hemolysis and cytolysis in Actinobacillus pleuropneumoniae, Actinobacillus suis, Pasteurella haemolytica, and Escherichia coli. Infect. Immun., 57: 3210-3213. Devenish, J., Rosendal, S., Boss6, J.T., Wilkie, B.N. and Johnson, R,, 1990. Prevalence of seroreactors to the 104-Kilodalton hemolysin ofActinobacillus pleuropneumoniae in swine herds. J. Clin. Microbiol. 28: 789-791. Frey, J. and Nicolet, J., 1988a. Purification and partial characterization of a hemolysin produced by Actinobacillus pleuropneumoniae type strain 4074. FEMS Microbiol. Lett. 55:4146. Frey, J. and Nicolet, J., 1988b. Regulation of hemolysin expression in Actinobacillus pleuropneumoniae serotype 1 by Ca 2+. Infect. Immun. 56: 2570-2575. Frey, J. and Nicolet, J., 1990. Hemolysin patterns ofActinobacillus pleuropneumoniae. J. Clin. Microbiol. 28: 232-236. Frey, J., Perrin, J. and Nicolet, J., 1989. Cloning and characterization of a co-hemolysin, the CAMP factor ofActinobacillus pleuropneumoniae. Infect. Immun. 57: 2050-2056. Gygi, D., Nicolet, J., Frey, J., Cross, M., Koronakis, V. and Hughes, C., 1990. Isolation of the Actinobacillus pleuropneumoniae haemolysin gene and the activation and secretion of the prohaemolysin by the HIyC, HIyB and HIyD proteins ofEscherichia coli. Molec. Microbiol. 4: 123-128. Kamp, E.M., Pompa, J.K. and van Leengoed, L.A.M.G., 1987. Serotyping of Haemophiluspleuropneumoniae in the Netherlands: with emphasis on heterogeneity within serotype 1 and (proposed) serotype 9. Vet. Microbiol. 13:249-257. Kamp, E.M. and van Leengoed, L.A.M.G., 1989. Serotype-related differences in production and type of heat-labile hemolysin and heat-labile cytotoxin ofActinobacillus (Haernophilus) pleuropneumoniae. J. Clin. Microbiol. 27:1187-1191. Kilian, M., Nicolet, J. and Biberstein, E.L., 1978. Biochemical and serological characterization of Haemophilus pleuropneumoniae (Matthews and Pattison, 1961) Shope 1964 and proposal ofa neotype strain. Int. J. Syst. Bact. 28: 20-26. Kume, J., Nakai, T. and Sawata, A., 1986. Interaction between heat-stable hemolytic substance
I M M U N O L O G I C A L P R O P E R T I E S O F ACTINOBACILLUS PLEUROPNEUMONIAE H E M O L Y S I N 1
73
from Haemophilus pleuropneumoniae and porcine pulmonary macrophages in vitro. Infect. Immun. 51: 563-570. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227: 680-685. Ma, J. and Inzana, T.J., 1990. Indirect enzyme-linked immunosorbent assay for detection of antibody to a 110,000-molecular weight hemolysin of Actinobacillus pleuropneumoniae. J. Clin. Microbiol. 28: 1356-1361. Nielsen, R., 1985a. Serological characterization ofHaemophilus pleuropneurnoniae (Actinobacillus pleuropneumoniae) strains and proposal of a new serotype: serotype 9. Acta Vet. Scand. 26: 501-512. Nielsen, R., 1985b. Serological characterization ofHaemophilus pleuropneumoniae (Actinobacillus pleuropneumoniae) strains and proposal of a new serotype: serotype 10. Acta Vet. Scand. 26: 581-586. Nielsen, R., 1986a. Serology of Haemophilus (Actinobacillus) pleuropneurnoniae serotype 5 strains: establishment of subtypes A and B. Acta Vet. Scand. 27: 49-58. Nielsen, R., 1986b. Serological characterization of Actinobacilluspleuropneumoniae strains and proposal of a new serotype: serotype 12. Acta. Vet. Scand. 27: 453-455. Nielsen, R. and O'Connor, P.J., 1984. Serological characterization of 8 Haemophilus pleuropneumoniae strains and proposal of a new serotype: serotype 8. Acta Vet. Scand. 25: 96-106. Pohl, S., Bertschinger, U., Frederiksen, W. and Mannheim, W., 1983. Transfer of Ilaemophilus pleuropneumoniae and the Pasteurella haemolytica-like organism causing porcine necrotic pleuropneumonia to the genus Actinobacillus (Actinobacillus pleuropneumoniae com. nov. ) on the basis of phenotypic and deoxyribonucleic acid relatedness. Int. J. Syst. Bacteviol. 33: 510-514. Rosendal, S. and Boyd, D.A., 1982. Haemophilus pleuropneumoniae serotyping. J. Clin. Microbiol. 16: 840-843. Rosendal, S., Devenish, J., Maclnnes, J.l., Lumsden, J.H., Watson, S. and Xun, H., 1988. Evaluation of heat sensitive, neutrophil-toxic, and hemolytic activity of Haemophilus (Actinobacillus) pleuropneutnoniae. Am. J. Vet. Res. 49:1053-1058. Ross, R.F., Hall, J.E., Orning, A.P. and Dale, S.E., 1972. Characterization of an Actinobacillus isolated from the sow vagina. Int. J. Syst. Bacteriol. 22: 39-46. Shope, R.E., 1964. Porcine contagious pleuropneumoniae. I. Experimental transmission, etiology and pathology. J. Exp. Med. 119: 357-368. Sneath, P.H.A. and Stevens, M,, 1990. Actinobacillus rossii sp. nov., Actinobacillus seminis sp. nov., nom. rev., Pasteurella bettii sp. nov., Pasteurella lymphangitidis sp. nov., Pasteurella marii sp. nov., and Pasteurella trehalosi sp. nov. Int. J. Syst. Bacteriol. 40:148-153. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications Proc. Natl. Acad. Sci. U.s.A. 76: 4350-4354. Wetmore, P.W., Thiel, F.J., Herman, Y.F. and Harr, J.R., 1963. Comparison of selected Actinobacillus species with a haemolytic variety of Actinobacillus from irradiated swine. J. Infect. Dis. 113: 168-194.
