EP0209554A1 - Immunoassay systems for the detection of salmonella - Google Patents

Abstract

Antiserum preparation capable of reacting with virtually all Salmonella serotypes. This preparation comprises a mixture of individual antisera developed against flagella of at least eight serotypes, and preferably all, in a group of ten selected serotypes of Salmonella. This antiserum preparation can be used in practical immunoanalytical systems for the rapid and specific detection of salmonella in samples. A method for immobilizing bacteria on a solid phase has also been developed. This process involves using titanium hydroxide as the solid phase and shaking or shaking a culture of bacteria with titanium hydroxide.

Description

"IMMUNOASSAY SYSTEMS FOR THE DETECTION OF SALMONELLA" Technical Field

The present invention relates to an antisera capable of reacting with the flagellae of substantially all Salmonella serotypes and the use of this antisera preparation in practical immunoassay systems for the rapid and specific detection of salmonellae in specimens. The present invention further relates to a method for immobilization of bacteria on a titanous hydroxyoxide (hereinafter referred to as titanous hydroxide) solid phase.

Background Art

Methods for the analysis of foods and other specimens for the presence of salmonellae have, to a large extent, followed traditional cultural methods. These methods are inherently cumbersome, costly, and require a long time to accomplish. Consequently, considerable research has been conducted to develop more practical and economic methods. Although a number of innovative methods have been proposed, this task is far from being accomplished and the use of cultural methods, in spite of their limitations, continues to dominate in testing laboratories.

The Kauffmann- hite schema classifies Salmonella serotypes on the basis of the serological properties of their 0 and H antigens. This approach has been very successful as a classification principle, and has resulted, so far, in the recognition of more than 1800 Salmonella serotypes and bioserotypes. This immunological diversity, however, has complicated the process of salmonellae detection using immunoassays. The use of 0 antigens as a basis for the specific detection of salmonellae by immunoassay is not possible because of the existence of close antigenic relationships of 0 identity between Salmonella, Arizona, Escherichia, Citrobacter, and the yeast, Candida. Limited information is available in the literature regarding the i munological relationships (cross- reactivities) of native Salmonella flagella with antigens from other microorganisms. However, it is known that, in contrast to known 0 and K antigenic relationships between Salmonella and Escherichia, there are no such relationships between flagella from these genera.

Immunoassays based upon the use of either radioactive isotopes or enzymes as labels such as radioimmunometric assays and enzyme immunometric assays, are characteristically highly specific and sensitive and have consequently been used extensively.

An important factor necessary for the success of immuno-assays which utilize radiolabelled or enzyme labelled probes is the need for effective separation of bound and non-bound antibody moieties. In liquid-phase immunoassays, separation is facilitated by the utilization of immunoadsorbents, as is the case with macromolecule antigens or by the use of charcoal as is- the case with micromolecules. The property of antibody or antigen adsorption to polymeric surfaces (disposable polystyrene or polypropylene tubes) has been used as a basis to develop a solid-phase radioimmunoassay. This adsorptive characteristic allows convenient and rapid removal of the free radioactive tracer by washing of the solid phase with water, after completion of the immune reaction. Consequently, the use of solid-phase immunoassays has increased because of the simplicity of its separation technique. Also, other substances for the immobilization of antibodies and antigens have been introduced such as commercial isothiocyanate-substituted plastic discs, polystyrene balls, nylon, activated thiol-Sepharose and microparticulate cellulose.

The immobilization of whole microbial cells on solid supports is used commercially to exploit their enzymatic activity for the production or degradation of substances. Techniques for immobilizing microbial cells include (i) entrapment in alginate gels, polyacrylamide gels, collagen membranes, metal hydroxide precipitates, agar pellets and liquid membranes, (ii) absorption onto chromatographic separation or ion exchange materials, (iii) selective binding to plant lectins and (iv) covalent binding of whole cells to a support, using bifunctional reagents, e.g. carbodiimide. However, efforts continue to find newer materials and techniques for microbial cell immobilization because of the immense industrial implication of this technology.

The immobilization of whole bacterial cells, for their diagnosis, by solid phase immunoassays has received very limited attention in the literature. The direct immobilization of cells of salmonellae and other Gram negative bacteria, for their diagnosis by solid phase immunoassays has been achieved by coating microtitre plates with bacterial suspensions. This technique requires a period of time ranging from lh to overnight incubation to accomplish cell immobilization. Other disadvantages of this approach are that (i) it is based upon non-specific absorption of antigens, which may be influenced by several factors (ii) after the completion of absorption, desorption can take place and (iii) only a small proportion of cells may be absorbed by this procedure. Disclosure of the Invention

The first aspect of the present invention consists in an antisera preparation capable of reacting with substantially all Salmonella serotypes in which said preparation comprises a mixture of individual antisera raised against the flagellae of at least eight of the Salmonella serotypes selected from the group comprising Salmonella oranienburg. Salmonella enteritidis. Salmonella kentucky. Salmonella waycross. Salmonella abortus-e ui. Salmonella tennessee. Salmonella 4, 12 : d;-. Salmonella 1, 4, 5,12;-;1, 2, Salmonella worthington and Salmonella lille. It was perceived that Salmonella flagella possess common H antigenic determinants (which are not shared by other microorganisms) at sufficiently high concentrations to enable the specific detection of salmonellae. Close examination of the H antigenic composition of Salmonella isolates which have been isolated and serotyped in

Australia during the period 1978-1980 revealed that 99% of the serotyped isolates had one or more of only ten H antigens or related antigens.

In a preferred embodiment of the first aspect of the invention the individual antisera comprising the antisera preparation were raised against flagella isolated from the following strains of Salmonella; S. oranienburg (S.R.L.* No. 1254), S. enteritidis (S.R.L. No. 1267), S. kentucky (S.R.L. No. 1285) S. waycross (S.R.L. No. 1312) S. abortus-equi (S.R.L. No. 1451) S. tennessee (S.R.L. No.

1623) S.4, 12:d:- (S.R.L. No. 1634) S. 1,4, 5, 12;-: 1, 2 (S.R.L. No. 1635) S.worthington (S.R.L. No. 1695) and S. lille (S.R.L. No. 1721). The antigenic characteristics and molecular weights of the flagellins from these ten Salmonella serotype are set out in Table 1.

*Salmonella Reference Laboratory, Adelaide, South Australia

Table 1. Antigenic characteristics and molecular weights of flagellins from ten Salmonella serotypes

Serotype ^a Reference H-antigen Antigen Molecular number b type phase weightc

S_. oranienburg 1254 m,t I 56 900 §.- enteritidis 1267 g,m I 58 400 S . kentucky 1285 ^z6 II 55 800 £5. waycross 1312 ^Z4'^z23 I 47 700 £3. abortus-equi 1451 e, n,x II 54 600 Σ5. tennessee 1623 ^z29 I 58 400 S . 4,12:d:- 1634 d I 57 100 S. 1,4,5,12:-:1,2 1635 1,2 II 57 100 S_. worthington 1695 l,w II 57 700 S. lille 1721 ^z38 I 50 700

49 700

All serotypes were monophasic except for S S.. kkeennttuucckkyy and S. worthington and in these cases, the unlisted phrases (i.e., i and z, respectively) were eliminated before flagellin isolation. Salmonella Reference Laboratory, Adelaide, South Australia. ^cDetermined by SDS-polyacrylamide gel electropheris. Two major protein bands were observed.

In a further embodiment of this aspect of the invention the flagella are isolated by culturing the bacteria in a defined medium, subjecting the bacteria to a pH of less than 3, and preferably 2 or less, for a time period of no more than 1 hour, and preferably from 20 to 40 minutes, removing the bacterial cells, and recovering the depolymerized flagellae. In a still further embodiment of this aspect of the invention the individual antisera components of the antisera preparation are produced by immunization of suitable antibody producing animals with flagellin emulsions in an adjuvant by multiple intrader al injection. The second aspect of the present invention consists in a method for the immobilisation of bacteria on a solid phase comprising shaking or agitating a culture of bacteria with titanous hydroxide. The titanous hydroxide was prepared by neutralization of a diluted solution of titanous chloride with an ammonia solution and then washing the suspension with saline to remove ammonium ions.

The immobilization of microorganisms with titanous hydroxide is a simple, one step method, involving shaking for approximately 10 min. The degree of immobilization of microorganisms is usually improved by increasing the duration and/or the intensity of agitation, because of increasing the probability of contact between the cells and the titanous hydroxide particles. Nevertheless, a duration of only 10 min was quite sufficient for almost complete immobilization. It is also possible that agitating too violently or for too long a time can cause dissociation of immobilized mircoorganisms. This, however, did not occur with titanous hydroxide in this investigation, indicating a high degree of stability.

The third aspect of the present invention consists in a method for the detection of Salmonella serotypes in a specimen comprising immobilizing the bacteria in the specimen on a solid phase, adding a quenching agent to mask unutilized binding sites on the solid phase, adding a Salmonella H antisera preparation according to the first aspect of the invention, and detecting the presence of bound antibodies. In a preferred embodiment of the third aspect of the invention the specimen is cultured before being immobilized on a solid phase. In a further embodiment of this aspect the solid phase is either titanous hydroxide or microtitre trays. In a still further embodiment of the third aspect of the invention the presence of bound antibodies are detected by either radioimmunometric or enzyme immunometric assay.

Detection and specificity for salmonellae using individual antisera and antisera mixtures

Salmonella serotypes were cultured in brain heart infusion broth (BHI) , diluted with saline and after removing aliquots for standard plate count determinations, the minimum detectable populations were determined with radioimmunometric assay (RIMA) after the immobilization of salmonella cells on titanous hydroxide. The antisera used in these assays were (i) each of the ten individual antisera (ii) a mixture of four antisera raised against flagellins from the Salmonella serotypes S_. abortus-equi, £>. kentucky, S_. tennessee and S_. waycross and (iii) a mixture of the 10 antisera. The dilution of each individual antisera or each component in an antisera mixture was 1:1000.

The specificity of radioimmunometric assay (RIMA) for the detection of salmonellae was carried out as follows. Sixteen species of Enterobacteriaceae, other than salmonella, namely, Citrobacter freundii, Edwardsiella tarda, Enterobacter aerogenes, Erwinia herbicola, Hafnia alvei, Proteus rettqeri, Proteus vulgaris, Serratia marcescens, Shigella dysenteriae, Proteus mirabilis, Proteus morganii, Shigella flexneri, Yersinia enterocolitica, Klebsiella pneu oniae, Yersinia pseudotuberculosis, and Escherichia coli, were grown in BHI at their optimum temperatures (30 to 37°C) . All of these species were motile, except those of Klebsiella and Shigella. The cultures where then subcultured in tetratnionate broth, Oxoid (TTB) and after incubation for 18h at 37°C, they were assayed using RIMA. These assays utilized two different radiolabelled tracer reagents which were used separately to establish their effectiveness in relation to salmonellae detection. These reagents were

125 I-labelled protein A and a specific radiolabelled antibody, against rabbit antibody ( 125I-labelled donkey anti-rabbit F(ab) ₂ fragment, 12*_I-F(ab) ₂) , purchased from the Radiochemical Centre, Amersham, England. The incubation times during assays, using these reagents, were

Ih at 30°C and 2h at 37°C, respectively.

Seventy seven Salmonella serotypes were also cultured separately in BHI then subcultured in mannitol selenite cystine broth (MSCB) , Oxoid, either alone or in association with a mixture of the sixteen enterobacterial species. The inoculation volumes in MSCB of salmonellae and the enterobacterial mixture were 1:100, respectively. After incubation of MSCB cultures for 18h at 42°C, they were assayed for salmonellae with RIMA, using radio¬ labelled protein A. The antisera used in these assays consisted of the ten antisera mixture (PFA) (at 1:400) as well as the Spicer-Edwards polyvalent antisera (SEA) (at 1:200) .

Detection of salmonellae using individual antisera and antisera mixtures

Radioimmunometric assay was conducted using varying levels of immobilized Salmonella cells and homologous antisera. The log minimum detectable populations of Salmonella serotypes using individual antisera in RIMA are listed in Table 2. The log of minimum detectable populations varied with the serotypes, averaging 4.60 + 0.54 with homologous antisera (n = 48).

The correlation coefficient between the relative Table 2. Log minimum populations of Salmonella serotypes, detected by radioimmunometric assay

a

H antisera H antigens of Salmonella serotypes m,t g,m ^z6 ^z4'^z23 e,n,x ^z29 ^d 1,2 l,w ^z38

m,t 5.04 5.00 - 6.23 - 5.70 - - 5.90 6.20 g,m 6.11 4.00 - 5.48 5.90 - - 4.78 5.28 6.48

- ^z6 4.70 4.60 - 4.60 - 5.18 - - -

- - ^z4'^z23 4.85 4.00 • - 5.60 - - 4.48 4.78 e,n,x - 6.04 6.48 - 4.48 6.30 4.60 4.48 6.30 -

- 29 5.04 5.60 5.48 5.00 - 4.30 6.60 5.95 5.18 d - 5.23 - - 5.20 - 4.60 4.60 5.20 5.26

1,2 - - - 6.78 5.58 - 5.50 4.60 - 6.95

1,w - - - 6.58 7.08 - - 5.48 5.85 - ^z38 5.04 — 6.70 — — 4.48 — 6.30 4.60 4.60

^aAntisera diluition was 1 :1000.

RIMA was <sonduicted usi:ng each antigen (i.e. Salmonella serotype) over a range of cell concentration. Each antigen was tested with its homologous antisera as well as each of the ni heterologous antisera.

cross-reactivities % of individual antisera and log minimum detectable populations of heterologous antigens (Tables 2 and 3) varied with individual antisera from 0.98 to 0.76. However, overall (i.e. with all antisera) such correlation amounted to 0.67 which was highly significant (p*^0.005) and the % of variation due to regression was 45.0.

Based upon the results in Tables 2 and 3, it appeared possible to detect the ten Salmonella serotypes using a mixture of only four antisera (e.g. z_g; e,n,x; z., z₂₃ and z_2g) • Consequently, such a mixture was prepared and used in RIMA for determining the minimum detectable populations of the ten Salmonella serotypes. The final dilution of each individual antisera in the mixture was 1:4000. The performance of this mixture, which was capable of detecting all 10 serotypes, was also compared to that of a mixture comprising the ten individual antisera at the same final dilution. The results (Table 4) show that the mixture of the ten antisera was superior to the mixture of the four antisera in relation to the detection sensitivity. While the four antisera mixture enabled the detection of a mean log population of 5.52 + 0.76, the ten antisera mixture enabled the detection of 4.66 + 0.72 (Table 4). Such a difference was highly significant using t test for paired data ( p -ζ 0.005). Moreover, there was no significant difference (p^0.4) between the log minimum detectable populations of the ten Salmonella serotypes using the ten antisera mixture and individual homologous antisera (Table 2, where the antisera dilutions used were 1:1000) . A more concentrated mixture of the ten antisera was prepared using each component at a final dilution of 1:1000. This mixture was used in RIMA for the determination of the minimum detectable populations of several Salmonella serotypes. The results (Table 5) showed that increasing the antisera concentration did not

Table 3. Relative cross-reactivities % between antisera and native flagella of ten Salmonell serotypes as determined by radioimmunometric assay.

H antiserab H antigens of Salmonella serotypes m,t g,m ^z6 ^z4^,z23 e,n,x ^z29 ^d 1,2 1,w ^z38

m,t 100.0 60.0 - 9.2 - 13.0 - - 20.0 10.0 g,m 15.7 100.0 - 7.5 4.8 - - 10.1 13.2 6.6

- 34.7 100.1 D 111.7 - 11.3 - - - ^z6

- ^z4'^z23 29.7 - 100.0 - 13.8 - - 35.5 28.2 e,n,x - 3.9 3.7 - 100.0 3.1 56.6 31.6 5.0 - ^z29 33.9 18.6 19.5 27.1 . - 100.0 - 3.4 12.7 26.3 d - 17.7 - - 31.5 - 100.0 56.7 23.3 14.8

1,2 - - - 3.7 10.2 - 8.6 100.0 - 4.0 l,w - - - 20.2 5.0 - - 121.0 100.0 - ^z38 30.0 — 5.0 — — 92.5 — 10.0 45.0 100.0

^aThe above values were obtained using populations of approximately 10 C .F.U. of each

Table 4. Log minimum detectable population of Salmonella serotypes using antisera mixtures³ in radioimmunometric assay.

H antigen of H antisera mixture, consisting of

Salmonella serotypes 4 antisera 10 antisera⁰

m,t 6.48 5.15 qg,,mm 5 5..4455 4.40 ^z6 4.52 4.38 ^z4^,z23 4.60 3.48 e,n,x 4.54 4.40 ^z29 6.26 5.20 dd 5 5..4488 4.34

1,2 5.57 3.95 l, 6.34 5.56

Z•_ o 5.95 5.70

^aThe final dilution of each component of the antisera mixtures was 1:4000. Antisera used were anti- e,n,x; anti-z_fi; anti-z₄,z₂₃ and anti- z₂g ^cA Annttiisseerraa uusseedd wweerre those prepared from the serotypes listed in Table 1.

Table 5. Log minimum detectable populations (M.D.P.) of Sal onella serotypes and other

Ente obacteriaceae speicies, using antisera mixture³ in radioimmuno etric '■ assay.

Miocroorganism H antigen M.D.P. Microorganism H antigen M.D.P.

S. oranienburg ,t: - 5.14 __.- derby f,g: (l,2) 3.48

S. kentucky -^:z6 3.60 Ui- orientalis k:e ,n, z, -. 3.85

S. abortus-equi -:e,n,x 4.58 S. l,4,5,12:e,h:- e,h: - 3.85

S. worthington -:l,w 5.41 S. senftenburg g,s,t:- 5.12

S . 1,4, 5,12:-:1,2 -:1,2 4.23 S . 3,15:y:- y:- 5.00

S. 4,12:d:- d:- 4.61 S. l,4,5,12:i:- i :- 5.18

S. lille ^z38^* 5.63 S . birkenhead c;l,6 5.38

S. waycross 3.48 ^z4' ^z23 __.• havana f,g, (s) :- 5.51

S. tennessee ^z29^:" 5.45 Escherichia coli N.A. 7.00

S. enteritidis g,m:- 4.08 Proteus vulqaris N.A. 7.00

S. infantis r:l,5 4.53 Enterobacter aerogenes N.A. 7.00

S. give l,v:l,7 5.63 Citrobacter freundii N.A. 7.00

-

a _ .

The antisera mixture consisted of ten antisera (Table 3), at a final concentration of 1:1000 of each component.

significantly improve the detection sensitivity, as compared to those levels detected in Table 4 (p>>0.35). This antisera mixture was also capable of detecting other Salmonella serotypes (Table 5) which possessed serotype- specific H antigens (according to the Kauffmann-White schema) differing from the Salmonella antigens used to raise the ten antisera (e.g. S_. infantis, S_. give and S_. birkenhead) . The log of minimum detectable populations ranged from 3.48 to 5.63. Moreover, this antisera mixture was highly specific for salmonellae detection using RIMA and did not detect populations of up to 10 colony forming units / lOOul of the species E. coli, £. vulgaris, E. aerogenes and C. freundii. Such specificity is also reflected in Table 6 where RIMA was conducted using ¹²⁵I-labelled protein A and ¹²⁵I-labelled F(ab)₂ as tracer reagents. As can be seen, none of the sixteen enterobacterial species tested were detected by either system.

Both radiolabelled reagents were also used to establish the detectability of twenty Salmonella serotypes which were grown alone and in association with a mixture of the sixteen enterobacterial species in TTB. The results in Table 7 show ι ~ - the ability of RIMA utilizing I-labelled protein A to detect all Salmonella serotypes tested. The detection of 77 different Salmonella serotypes in

MSCB with IRMA, utilizing the ten antisera mixture (1:400) , the commercially available Spicer-Edwards polyvalent

1 -s

Salmonella H antisera (1:200) and ³I-labelled protein A was possible. Most of these salmonellae possessed serotype-specific H antigens which differed from the Salmonella H antigens used to raise the ten antisera utilized in the assay. All serotypes were detectable after growing alone or in association with the enterobacterial species mixture (Table 8) . The results also show that the performance of the ten antisera mixture was superior to that Table 6. Detectability of different species of Enterobacteriacea grown in tetrathionate broth as determined by % bound of 125 'i,-labelled protein A and 125 '.I-labelled F(ab) 2' in radioimmunometric assay.

125 I protein A 125 I-F(ab)

Salmonella waycross 18.0 5.0 Citrobacter freundii -2.0 -1.1

Edwardsiella tarda -1.4 -1.1

Enterobacter aerogenes -1.6 -0.8

Erwinia herbicola -0.8 -0.9

Escherichia coli -0.9 -1.4 Hafnia alvei -1.6 -1.5

Klebsiella pneumoniae -1.4 -0.7

Proteus mirabilis -2.7 -0.5

Proteus morganii -1.8 -0.1

Proteus rettgeri -1.7 -0.9 Proteus vulgaris -0.5 -0.8

Serratia marcescens -1.6 -0.2

Shigella dysenteriae -1.9 -0.6

Shigella flexneri -1.6 -0.5

Yersinia enterocolitica -1.9 -1.3 Yersinia pseudotuberculosis -2.1 -0.3

All species were motile except for those of Klebsiella and Shigella. Table 7. Detectability of Salmonella serotypes grown in tetrathionate broth alone and in association with a mixture of sixteen other species of Enterobacteriaceae³ as determined by % bound of I-labelled protein A and F(ab)₂ in radioimmunometric assay.

Salmonella Growth ccinditions serotype Alone In association

Protein A F(ab)₂ ProteinA F(ab)₂

S. oranienburq 6.4 -0.6^b 2.1 0.1

S. kentucky 9.4 1.0 2.8 0.4

S. abortus-equi 10.4 1.1 -0.9^b -l.l^b

S. worthington 23.9 4.4 17.2 7.6

S . 1,4,5,12:-:1,2 11.6 1.5 1.2 -0.6^b

S. 4,12:d;- . 21.4 4.9 19.2 3.6 S. lille 1.9 -0.4^b 1.1 -0.5^b

S. waycross 22.8 5.0 15.0 2.2

S. tennessee 10.8 1.8 7.8 1.4

S. enteritidis 12.9 1.8 5.0 0.1

S. infantis 2.6 -0.4^b 1.3 0.2

S. give 6.2 1.2 6.9 -1.4^b

S. derby 19.5 4.2 14.3 2.5

S. orientalis 4.8 0.6 1.0 -0.9^b

S . l,4,5,12:eh:- 19.4 3.0 14.5 1.8

S. senftenburg 8.5 6.8 8.3 5.1

S. 3,15:y:- 9.2 5.9 7.8 3.3

S. l,4,5,12:i:- 5.9 1.1 3.9 0.1

S. birkenhead 2.1 0.6 1.1 -0.3^b

S. havana 6.0 1.9 4.1 1.9

These species were those listed -in Table 6. The inocula used for the tetrathionate broth consisted of salmonellae and the enterobacterial species at the ratio of 1:100, respectively.

After streaking tetrathionate broth cultures on bismuth sulphite agar, salmonellae colonies were present in all cases except for S_. abortus-equi grown in association with enterobacterial species.

Table 8. Detection of various salmonellae grown in mannitol selenite cystine broth alone and in association with a mixture of other species of Enterobacteriacea' by radioimmunometric assay, utilizing ¹²⁵I-labelled protein A.

% bound ¹²⁵ I-protein A with cultures

Salmonella Composition Alone In association of H antigen SEA^b PFA^C SEA PFA

Co]ntrol NIL -1.7 -1.8 3.2 4.5

S. aberdeen i:l,2 31.5 38.3 33.0 36.9

S. abony b:e,n,x 27.6 40.3 25.8 45.6

S. abortus-equi -:e, n,x 36.9 54.2 26.0 52.3

S. adelaide f,g:- 13.3 44.2 11.5 30.1

S. agona f,g,s:- 14.8„ 45.5 11.4 23.4

S. alsterdorf g,m,t :- 14.0 35.8 12.7 17.5

S. anatum e,h:l, 6 17.9 28.4 12.0 21.4

S. bareilly Y:l,5 38.0 33.1 34.0 32.8

S. birkenhead c :1, 6 42.4 36.2 43.2 18.7

S. blockley k:l,5 13.2 23.9 18.6 27.4

S. bornum ^z38^:" 13.5 14.8 19.4 15.5

S. bovis r:l,5 31.0 32.1 21.2 24.1 morbificans

S. braenderup e,h:e,n,z-ι _c 48.1 41.5 28.9 36.4

S. bredeney l,v:l,7 14.2 43.1 20.1 42.6

S. charity d:e,n,x 23.1 30.6 18.8 23.4

S. cholera-suis c:l,5 13.7 33.8 15.1 30.9

S. coley-park a:l,w 15.1 24.2 12.6 19.4

S. cubana ^z29^:" 13.0 28.3 10.1 16.9

S. derby f,g:(l,2) 25.1 44.1 31.3 42.7

S. dublin g,p:- 14.8 53.6 12.0 24.3

S. eastbourne e,h:l, 5 23.9 30.0 25.4 26.3

S. emek g,m,s :- 24.7 48.1 16.8 40.2 Table 8 ( cont ' d)

% bound 125I-protein A with cultures

Salmonella Composition Alone In association of H antigen SEA^b PFA^C SEA PFA

S. enteritidis g,m:- 39.7 50.6 44.4 49.1

S. give l,v:l,7 15.1 50.7 23.5 47.7 s. havana f,g, (s) :- 16.5 29.6 18.9 32.7 s. infantis r:l,5 32.2 48.0 32.4 35.1 s. indiana z:l,7 28.3 33.6 21.2 29.3 s. jangwani a:l,5 33.8 51.9 29.0 48.0 s. iava b:(l,2) 35.8 43.2 26.3 25.1 s. javiana p ϊJ-, zt 27.2 55.4 28.6 52.0 s. kentucky i:z₆ 20.3 47.8 22.3 47.9 s. kottbus e,h:l,5 25.0 36.4 27.7 35.2 s. lansing i:l,5 18.6 24.0 14.2 32.3 s. lexington z_1Q:l,5 14.6 30.1 14.4 29.5 s. lille ^z38^:' 25.0 29.6 22.8 28.2 s. livingstone d:l,w 37.0 59.3 28.5 18.2 s. london l,v:l,6 17.9 51.3 10.3 25.4 s. anila Zi -ι!lf J 19.4 28.8 12.2 21.2 s. bandaka z₁₀:e,n,z₁₅ 19.6 29.6 23.9 31.1 s. ississippi b:l,5 24.6 28.0 28.5 28.7 s. montevideo g,m,s :- 22.5 31.2 22.0 28.6 s. newbrunswick l,v:l,7 34.2 56.2 34.2 51.6 s. newington e,h:1, 6 13.7 11.0 10.5 25.2 s. neinstedten b: (l,w) 40.2 50.9 40.6 48.6 s. ohio b :l,w 28.4 52.3 29.0 41.0 s. oranienberq m,t :- 40.7 52.6 36.4 37.2 s. orientalis k :e,n, z, -. 25.2 52.1 17.2 35.7 s. orion y:l,5 21.6 46.9 14.7 39.4 s. Panama l,v:1.5 20.0 28.6 20.8 37.7 Table 8 (cont'd) .

% bound 125 I-protein A with cultures

Salmonella Composition Alone In association of H antigen SEA^b PFA^C SEA PFA

S. potsdam l,v:e.n, z 15 24.8 30.9 21.1 26.7

S. reading e,h:l. 5 21.8 33.1 20.9 19.1

S. rubislaw r:e,n. X 18.0 38.5 27.7 40.5

S. san diego e,h:e.n, z 15 10.5 41.6 10.7 26.4

S. schwarzengrund d:l,7 19.2 41.0 16.9 30.2

S. senftengerg g,s,t: - 9.4 16.3 12.8 13.6

S. sofia b: (e, rι,x) 12.2 14.1 21.7 23.2

S. Stanley d:l,2 31.8 53.4 30.3 27.4 s. taksony i:z₆ 17.2 47.0 18.9 44.9 s. tennessee z₂₉:- 33.7 32.3 36.0 29.1 s. thompson k:l,5 29.4 29.8 22.5 25.8 s. urbana b :e,n. X 16.5 41.9 28.1 43.9 s. victoria l,w:l. 5 17.5 34.0 11.3 30.1 s. virchow r:l,2 35.1 42.9 31.0 •27.5 s. waycross -

^Z4^Z23^: 16.8 32.7 20.0 29.9 s. welikade l,v:l. 7 8.4 33.0 12.7 18.7 s. weltevreden r:z₆ 10.6 27.5 13.2 45.2 s. wichita d:z_3? 13.6 22.8 15.7 46.8 s. worthington z:l,w 30.5 43.0 29.6 41.3 s. Zanzibar k:l,5 14.0 33.8 24.4 34.9 s. zehlendorf a:l,5 21.7 30.1 18.5 34.1 s_. :f,g:- f,g:- 12.8 35.7 14.7 27.1 s. R:l,v: - l,v:- 11.3 39.8 15.2 27.9 s_. 1,4,5,12:-:1,2 -:1,2 26.4 32.7 18.4 21.6 s. l,4,5,12:e,h:- e,h: - 23.3 40.8 23.3 39.1 s . l,4,5,12:i:- i :- 19.8 29.3 19.9 28.7 s. 3, 15:y:- y:- 14.9 24.7 20.8 35.1 s . 4,12:d:- d:- 32.7 46.3 27.6 45.4 The mixture of enterobacterial species consisted of the sixteen species listed in Table 6. The inoculation ratio between each salmonellae and the mixture was 1:100, respectively. Spicer-Edwards Salmonella polyvalent H antisera (1:200). c Pooled anti•sera produced against ten Salmonella flagellins (Table 1), at 1:400.

of the commercially available Spicer-Edwards polyvalent antisera. This is indicated by the significantly higher values of specific binding of radiolabelled protein A by the former (p<0.001, using paired t test). A further aspect of this invention was the development of a method for the immobilization of bacteria on a non-soluble support, thereby facilitating the detection of these bacteria, by solid-phase immunoassay systems. The criteria for such a method were that it should be simple, rapid, efficient and that it should not interfere with the measurement of antigen/antibody reactions by immunoassays.

Immobilization experiments were conducted using titanous hydroxide, titanic hydroxide and zirconium hydroxide, and Salmonella cells as well as other microorganisms. These microorganisms were grown overnight in BHI broth at 37°C. The cultures were diluted to the desired cell concentrations with sterilized saline (pH 6.8) . The metal hydroxide suspensions were dispensed in polystyrene tubes at the rate of 50jul/tube. Diluted culture suspensions were added (lOOμl) and the tubes were agitated for 10 min at room temperature, in a reciprocal agitator, at a speed to maintain the metal hydroxides suspended in the tubes, except when otherwise specified. Sterilized saline (0.9ml) was then added to each tube and after vortexing, the tubes were centrifuged (1400 xg) for 5 min at 4°C. Aliquots of IOOJUI of the supernatant in the tubes were serially diluted and the bacterial populations were determined by the standard plate count method (Tables 9-18) . The standard plate count was also conducted on tubes containing no metal hydroxide suspensions (control) but which were treated in the same way as those containing the metal hydroxides. The immobilization efficiency was expressed as follows: % immobilization = (population in control tube - population in supernatant of tube containing metal hydroxide) x 100/population in control tube. Table 9. Immobilization % of salmonellae by metal hydroxides.

Salmonella Cell population Titanous Titanic Zirconium serotype x 10 hydroxide hydroxide hydroxide

S. enteritidis 4.3 96.3 93.5 92.1

43.0 97.0 91.5 93.0

S. lille 4.0 98.5 92.2 91.5

40.0 94.4 93.2 92.0 S. tennessee 3.5 91.0 94.6 92.8

35.0 94.9 94.3 94.0

S_. waycross 4.0 97.8 96.6 93.1

40.0 ^'• 94.2 . 90.1 90.0

Immobilization % using titanous hydroxide was significantly higher than that obtained with titanic and zirconium hydroxides (P= -^0.05 and -ζθ.01, respectively, using paired t test) . No significant difference existed between the immobilization by titanic and zirconium hydroxides.

Table 10. Binding % of radiolabelled antibody and protein

A A bbyy mmeettaall hydroxides in the presence of quenching materials.

Titanous hydroxide Titanic hydroxide Zirconium hydroxide

125 12

Quenchers I-IgG ϊ- Protein A 125 I-IgG 125 I-Protein A 125 I-IgG ¹²ϊ-Protein A

Nil (control) N.T. 74.8 N . T. 69.7 N.T. 71.6

Bovine serum albumin 17.5 8.6 24. . 3 28.4 14.4 15.9 (Calbiochem)

Brain heart infusion (BB ) 21.7 6.0 22 . . 3 17.6 13.3 9.7

Peptone (Oxiod) 25.4 5.6 23 . .0 11.4 13.5 8.7

Gelatin (Difco) 5.6 3.9 11 , .9 8.7 9.4 5.8

Gelatin (Sigma) 5.7 5.3 10 , .8 17.0 9.2 9.3

Normal rabbit serum 20.2 8.2 24 , .7 43.7 13.4 14.8

Volumes of metal hydroxides used were 50ul. When lOOul quantities were used the binding rates of radiolabelled reagents increased significantly (p<0.01).

Final concentration of quenchers was 2% except for normal rabbit serum (1:20). Lower final concentrations of quenchers resulted in signficantly higher binding % of the radiolabelled reagents (p<0.01).

Not tested.

Table 11. Immobilization of salmonellae by titanous

ization

n

50 95.7 99.6

Table 13. Effect of shaking time on salmonellae immobilization (%) by titanous hydroxide.

Shaking time S. enteritidis S. lille

o^ 40.3 45.4

5 min >75.3 92.0 1 100 nmin 98.17 97.0 4 h^u 80.7 99.4

24 h^L 96.8 96.8

^aOrbital shaker (200 r.p.m.) at 22°C. Population of S_. enteritidis and S>. lille were 3.3 x 10 and 4.9 x 10 colony forming units/100ιl. respectively, c^cTube contents were vortexed for only few seconds.

Vigorous shaking applied, 200 r.p.πu.

Table 14. Effect of buffer systems on immobilization (%) of salmonellae³ by titanous hydroxide.

Immobilization medium S>. waycross S_. tennessee

Saline 94.0 99.8

Imidazol-s aline 71.7 90.4

Tris -saline 69.1 74.7

NaOH-citric-saline 62.9 79.0

Phosphate-saline 40.9 30.7

Populations of S_. wayc ross and S_. tennessee used were 2.47 x and 2.93 x 10 colony forming units/100^1, respectively.

I.. pH of saline was 6.8 and the pH of other buffers was 7.2 + 0.1. Table 15. Effect of pH change in a non-buffered saline solution on the immobilization (%) of salmonellae by titanous hydroxide.

Immobilization pH S_. lille S_. enteritidis

3.00 8 888..7 7 82.6

4.05 9 900..6 6 91.1

5.10 9 944..6 6 96.2 6.05 9 911.. ,55 92.0 7.25 9 966.. ,55 96.0 8.02 8 866..0 0 82.7

Populations of S_. lille and S_. enteritidis used were 3.65 x 10 6 and 2.9 x 10fi colony forming units/lOQul, respectively.

Table 16. Immobilization of salmonellae present in saline and several broth media by titanous hydroxide,

__• liHe S. enteritid

Cells present Population Immobilization Population Immobil in: x 10⁶ % x 10⁶ %

Saline 8.3 99.9 3.1 93.

Brain heart infusion broth 340.0 99.9 4.0 96. Nutrient broth 3.4 94.1 3.2 83. Tetrathionate broth 1.5 80.2 5.3 91. Mannitol selenite broth 5.0 98.6 2.1 96.

Salmonellae were first grown in brain heart infusion broth then dilua

Salmonellae were first grown in each broth then diluted in fresh brot

Table 17. Immobilization of some Gram-negative bacteria by titanous hydroxide.

Microorganism Cell population Immbolization x 10⁶ %

Citrobacter freundii 1.21 89.9

Edwardsiella tarda 0.56 91.6

Enterbacter aerogenes 4.30 99.3

E Errwwiinniiaa h.'heerrbbiiccoollaa 4.50 96.3

Escheri hia coli 4.90 95.6

Hafnia alvei 23.00 99.3

Klebsiella pneumoniae 0.90 98.4

Proteus :mirabilis 23.20 99.9

PPrrootteeuuss :mmiirrggaanniiii 2.00 92.8

Proteus rettgeri 5.00 89.0

Proteus vulgaris 0.38 94.2

Serratia marcescens .5.90 96.4

Shigella dysenteriae 4.80 98.4

SShhiiggeellllaa ^•fflleexxnneerrii 1.59 88.7

Shigella sonnei 0.74 95.0

Yersinia enterocolitica 3.80 82.0

Yersinia pseudo- 1.78 91.7 tuberculosis

Table 18. Immobilization of some Gram-positive bacteria by titanous hydroxide

Cell

Microorganism Reference³ Population Immobilizatio

X 10^b %

Bacillus cereus ATCC 11778 0.60 97.0

Bacillus subtilis H.A.C., 106 0.77 98.4

Micrococcus luteus H.A.C., 85 1.70 93.1

Microcuccus varians H.A.C., 20 1.34 88.1

Staphylococcus ATCC 27664 0.88 91.0 aureus

Staphylococcus ATCC 14458 1.00 91.0 aureus

Staphylococcus H.A.C., 82 0.50 -- 93.4 epidermidis

Streptococcus H.A.C., 38 17.1 _r 99.6 faecalis

³ATCC, American Type Culture Collection, U.S.A.; H.A.C., Hawkesbury Agricultural College, Richmond, N.S.W., Australia

The immobilization data for four Salmonella serotypes, each studied at two population levels, are shown in Table 9. All three metal hydroxides were capable of giving a high degree of immobilization, amounting to ^90%. However, the immobilization percentages obtained with titanous hydroxide were significantly higher than those obtained with titanic and zirconium hydroxides (p =<0.05 and <0.01, respectively, using paired t-test) . The immobilization data were obtained by comparing viable cell counts before and after immobilization. While small numbers of cells were found in the supernatants of suspensions after immobilization, large numbers of cells were recovered by plating of the metallic suspensions. This indicates that the cells were immobilized rather than being killed by these hydroxides. Immobilization of microorganisms other than salmonellae by titanous hydroxide This was carried out using cultures grown in BHI which were diluted with saline. The immobilization efficiency was determined using lOOjαl diluted culture, 50μl of titanous hydroxide suspension and 10 min shaking time. The results (Tables 17 and 18) showed the ability of titanous hydroxide to immobilize a wide range of Gram- negative and Gram-positive bacteria. The immobilization %, ranged from 82.0 to 99.9 and 88.1 to 99.6, respectively. The immobilization of salmonellae and other Gram- negative and Gram-positive bacteria by titanous hydroxide was shown in this investigation to be far superior to other conventional methods. Detection of Salmonellae by Immunoassays

Salmonellae were cultured in brain heart infusion broth (BHI), Baltimore Biological Laboratory, for 18h at 37°C then inoculated into Oxoid mannitol selenite cystine (MSCB) and tetrathionate (TTB) broths. The selective media were incubated for 18h at 42 and 37°C, respectively. In some experiments, salmonellae were also inoculated into MSCB in association with a mixture of 16 cultures of other enterobacterial species, at an inoculation volume ratio of 1:100, respectively. The mixed cultures were incubated at 42°C for 18h. The growth of organisms in pure culture was determined using the standard plate count method.

Also, food homogenates were prepared (25 g food sample and 225ml of buffered peptone water) , using a Sorvall Omni-Mixer with solid foods. The homogenates were inoculated with salmonella cultures in BHI, using a straight inoculation wire, then pre-enriched for 18-24h at 37°C. The pre-enrichment cultures were inoculated (1.0ml) into MSCB and TTB, for selective enrichment, and incubated for 18-24h at 42°C and 37°C, respectively.

Selective enrichment cultures were assayed for salmonellae using either radioimmunometric and/or enzyme immunometric assays. In the radioimmunometric assay- system lOOul samples of the enrichment cultures were added to 50ul aliquots of titanous hydroxide suspension in polystyrene tubes. After immobilization of the cultured^' bacteria by agitation for 10 min. , lOOμl quantities of the PFA and SEA antisera diluted in 0.05M phosphate buffer containing 0.85% NaCl, 0.1% NaN₃ and 2% gelatin (PSAG) to 1:400 and 1:200 respectively were added. The mixtures were incubated for 2h at 37 C in an orbital shaker incubator at 180 r.p.m. The unbound antibodies were eliminated from the tubes by dilution with 3.0ml of saline, centrifugation at 1700 x g for 5 min and aspiration. Next, lOQul of

125 I-labelled protein A, diluted in PSAG to give approximately 25,000 c.p.m./lOQul were added. The tubes were placed in the shaker incubator for lh at 30°C. The unbound radioactive reagent was removed by dilution, centrifugation and aspiration and the residual radioactivities were counted using a Packard Selectronic Autogamma Spectrometer. Control tests were conducted as above for each sample, whereby quantities of lOQμl PSAG were added in place of salmonella H antisera. The net radioactivities bound were determined by deducting the residual radioactivities of the tubes containing no H antisera from those containing the antisera. The enzyme immunometric assays were also performed using titanous hydroxide suspension in polystyrene tubes, as the solid-phase using the same method used in the radioimmunometric assay except that the use of the antibacterial agent NaN., was omitted and in place of ¹²⁵I-labelled protein A, 200jul of anti- rabbit IgG (whole molecule) - alkaline phosphatase conjugate antibody raised in goat, was used. The optimum concentration of the enzyme conjugate was found to be 1:250 as determined by chequerboard titration. After incubation for lh at 37 C and the elimination of the unbound enzyme conjugate fraction, aliquots of 200,ul of alkaline phosphatase substrate were added. The enzyme substrate was prepared on the same day of use by adding one tablet (5 mg) of

Sigma phosphatase substrate 104 to 5 ml of 0.1 M glycine buffer containing 1.0 mM zinc and 1.0 noM magnesium chlorides (pH 10.4) . After incubation for lh at 37°C, 200μl quantities were removed after the centrifugation of titanous hydroxide, and transferred to a polystyrene microtitre plate with flat bottom. The absorbance was then measured using a Titertek Uniskan spectrophotometer at 405nm. The net absorbance was then calculated for each sample by deducting the control absorbance value (where lOOjul PSAG was added in place of Salmonella H antisera) . At the same time selective enrichment cultures were streaked onto XLD agar medium (Oxoid) to ascertain presence or absence of salmonella. Presumptive Salmonella colonies were confirmed using conventional culture methods. Examination of food samples

A total of 235 food samples were investigated for the presence of salmonellae. These comprised 20 raw milk, 28 pasteurized milk, 30 soft cheese (mozzarella, fetta, ricotta, cottage and cacciotta) , 26 cheddar cheese, 70 milk powder, 20 chicken carcasses, 13 chicken livers, 10 egg products (egg powder and egg noodles) and 18 minced beef samples.

The samples were tested for the presence of salmonellae using the cultural method of the Standards Association of Australia (1983) . Also after pre-enrichment and selective enrichment of food samples, only MSCB cultures were tested (in duplicate) for salmonella with immunoassays, using PFA and SEA antisera diluted to 1:400 and 1:200, respectively. The immunoassays were performed using titanous hydroxide as solid-phase.

After growth, the cultures were used for the detection of salmonellae by RIMA or EIMA, using both the ten antisera mixture (PFA) and the Spicer-Edwards polyvalent antisera (SEA) preparations. The data presented in Table 8, (RIMA) and Table 19 (EIMA) show the radioactivity and absorbance responses for each assay. High response values were obtained for all salmonellae whether grown alone or in association with other entero¬ bacterial species. The data also indicate the superior performance of PFA in comparison with SEA. Overall, the specifically bound radioactivity (RIMA) and the specific absorbance (EIMA) obtained with PFA were significantly higher than those obtained with SEA (p<0.001 and <0.05, respectively, using paired t test) . Table 19. Detection of various salmonellae grown in mannitol selenite cystine broth alone and in association with a mixture of other species of Enterobacteriacea³, by enzyme immunometric assay, utilizing anti-rabbit IgG phosphatase conjugate.

Absorbance with cu Itures

Salmonella Composition Alone In association of H antigen SEA PFA SEA PFA

Control Nil 0.02 0.05 -0.10 0.13

S. aberdeen i:l,2 0.46 0.34 0.33 1.30

S . abony b :e,n,x 0.74 1.51 0.45 0.62

S. abortus-equi -:e,n,x 0.26 0.47 0.64 0.81

S. adelaide f,g:- 0.57 0.48 0.43 0.48

S. agona f,g,s:- 1.57 0.67 0.43 0.43

S. alsterdorf g,m,t:- 0.67 0.44 0.91 0.80

S. anaturn e,h:l,6 1.16 0.32 0.74 0.65

S. bareilly y:l,5 0.83 0.27 1.20 0.76

S. birkenhead c:l, 6 0.34 0.78 0.71 0.98

S. blockley k:l,5 0.24 0.47 0.40 0.43

S. bornum ^z38^{: "} 0.24 0.42 0.80 1.09

S. bovis r:l,5 0.48 0.69 0.57 0.90 morbificans

S. braenderup e,h:e ,n, z-. -- 0.89 0.82 1.44 1.08

S. bredeney l,v:l,7 0.34 1.10 0.71 1.03

S. charity d:e,n,x 0.44 0.55 0.51 0.70

S. cholera-suis c:l,5 0.38 0.79 0.94 1.06

S. coley-park a:l,w 0.45 0.23 0.47 0.42

S. cubana 0.29 0.44 0.32 ^z29^:" 0.45

S. derby f,g:(l,2) 0.90 1.00 1.23 0.88

S. dublin g,p:- 0.79 1.19 0.70 1.15

S. eastbourne e,h:l, 5 0.37 0.53 0.13 0.78 Table 19 ( cont ' d)

Absorbance with cultures

Salmonella Composition Alone In association of H antigen SEA PFA SEA PFA

s. emek g,m,s:- 0.30 1.74 0.32 0.50

S ienteritidis g,m:- 0.31 0.58 0.45 0.56

S. give l,v:l,7 0.57 0.92 0.44 1.18 s. havana f,g(5) :- 0.62 1.08 0.41 1.17 s. infantis r:l,5 1.06 0.39 0.25 0.70 s. indiana z:l,7 0.81 - 0.89 0.92 0.98 s. iangwani a:l,5 0.63 0.50 0.65 0.74 s. java b: (l,2) 0.36 0.25 0.52 0.59 s. javiana l,z₂₈.:l,5 0.34 0.46 0.35 0.65 s. kentucky i:z₆ 0.43 2.10 0.84 1.81 s. kottbus e,h:l, 5 0.52 0.37 0.68 0.65 s. lansing i:l,5 0.56 0.80 0.47 0.99 s. lexington z, q:1, o 1.22 1.07 0.46 1.19 s. lille ^z38^{: "} 0.53 0.77 0.45 0.50 s. livingstone d:l,w 0.93 0.63 1.07 1.57 s. london 1, v:l, 6 0.79 0.81 0.28 0.39 s. manila Z-i n ϊ -L , D 0.88 0.45 0.69 0.76 s. mbandaka ^z10^"®'^^{, z}15 0.79 2.25 0.86 1.55 s. mississippi b:l,5 0.98 1.10 0.61 0.74 s. montevideo g,m,s:- 0.38 0.23 0.25 1.19 s. newbrunswick l,v:l,7 0.78 1.22 1.49 1.32 s. newington e,h:l, 6 0.43 0.57 0.42 0.49 s. neinstedten b: (l,w) 0.51 0.32 0.40 0.70 s. ohio b:l,w 0.45 0.74 0.56 0.54 s. oranienberg m,t :- 1.16 1.39 0.35 1.23 s. orientalis K 2€ _/ n Z c 0.75 0.41 0.42 0.51 s. orion Y:l,5 1.41 1.52 0.86 1.18 s. Panama l,v:1.5 0.59 0.88 0.31 0.83 Table 19 ( cont ' d)

Absorbance with cultures

Salmonella Composition Alone In association of H antigen SEA PFA SEA PFA

s. potsdam l,v:e. n,z₁₅ 0.61 0.36 0.75 1.02 s. reading e,h:l. 5 0.83 0.60 1.07 0.86 s. rubislaw r:e,n. X 0.74 0.49 0.66 0.80 s. sandiego e,h:e. n,z₁₅ 1.13 1.06 0.56 0.85 s. schwarzen rund d:l,7 1,15 3.86 0.83 4.16 s. senftengerg g,s,t: - 0.51 0.40 0.88 1.11 s. sofia b: (e,rι,x) 0.49 0.46 0.43 0.46 s. Stanley d:l,2 0.87 1.70 0.88 1.57 s. taksony i:z₆ 0.49 0.60 0.62 0.47 s. tennessee z₂₉:- 0.93 1.22 0.49 0.96 s. thompson k:l,5 1.12 0.88 0.50 0.41 s. urbana ^• b:e,n. X 0.25 0.71 0.46 0.83 s. victoria l,w:l. 5 1.20 2.85 0.86 1.49 s. virchow r:l,2 2.38 1.38 1.26 1.01 s. waycross - 2.10 ^z4^z23^: 2.20 0.85 1.50 s. welikade l,v:l. 7 0.50 0.44 0.21 0.73 s. weltevreden r:z₆ 0.47 0.70 1.10 1.21 s. wichita d:z_3? 0.72 0.69 0.27 0.45 s. worthington z:l,w 1.34 2.40 0.75 1.68 s. Zanzibar k:l,5 1.82 1.89 0.73 0.72 s. zehlendorf a:l,5 0.31 0.52 0.33 0.92

S_. R:f,g:- f,g:- 0.60 0.84 0.44 0.72

S . R:l,v:- l,v:- 0.63 0.37 0.65 0.64

S . 1,4,5,12:-:1,2 -:1,2 0.76 2.18 0.74 1.53

S_. l,4,5,12:e,h:- e,h: - 0.89 0.88 0.61 1.30

S_. l,4,5,12:i:- i :- 1.17 0.89 0.46 0.71

S_. 3,15:y:- y:- 0.78 0.94 0.26 0.89

S. 4,12:d:- d:- 1.61 3.09 0.73 2.56 ³The mixture of enterobacterial species consisted of the sixteen species listed in Table 6. The inoculation ratio between salmonellae and the mixture was 1:100, respectively.

Claims

CLAIMS : -

1. An antisera preparation capable of reacting with substantially all Salmonella serotypes in which said preparation comprises a mixture of individual antisera raised against the flagellae of at least eight of the Salmonella serotypes selected from the group comprising Salmonella oranienburg. Salmonella enteritidis. Salmonella kentucky. Salmonella waycross. Salmonella abortus-equi. Salmonella tennessee. Salmonella 4, 12 : d:-. Salmonella

1. 4, 5, :-:!, 2, Salmonella worthington and Salmonella lille.

2. An antisera preparation as claimed in claim 1 in which the preparation contains individual antisera raised against the flagellae of all the said group of Salmonella- serotypes.

3. An antisera preparation as claimed in claim 1 or claim 2 in which the Salmonella serotypes comprised within the said group are S. oranienburg (S.R.L. No. 1254), S. enteridis (S.R.L. No. 1267), S. kentucky (S.R.L. No. 1285) S. waycross (S.R.L. No. 1312) S. abortus-equi (S.R.L. No. 1451) S. tennesse (S.R.L. No. 1623) S.4, 12:d:- (S.R.L. No. 1634) S. 1,4, 5, 12:-: 1, 2 (S.R.L. No. 1635)

S.worthington (S.R.L. No. 1695) and S. lille (S.R.L. No. 1721) .

4. An antisera preparation as claimed in any one of claims 1 to 3 in which the flagellae used to raise the individual antisera are isolated by culturing the bacteria in a defined medium subjecting the bacteria to a pH of less than 3, for a time period of less than 1 hour, removing the bacterial cells and recovering the depoly erized flagellae.

5. An antisera preparation as claimed in claim 4 in which the bacteria are subjected to a pH of less than 2.

6. An antisera preparation as claimed in claim 4 in which the bacteria are subjected to a pH of less than 3 for a period of 20 to 40 minutes.

7. An antisera preparation as claimed in any one of claims 1 to 3 in which the individual antisera components of said preparation are produced by immunization of suitable antibody producing animals with flagella preparations together with an adjuvant, by intradermal injections.

8. A method for the immobilization of bacteria on a solid phase comprising shaking or otherwise agitating a culture of the bacteria with titanous hydroxide.

9. A process as claimed in claim 8 in which the bacteria is of the genus Salmonella.

10. A process as claimed in claim 8 in which the culture of the bacteria is agitated with titanous hydroxide for at least 5 minutes.

11. A process as claimed in claim 10 in which the culture of the bacteria is agitated with titanous hydroxide for at least 8 minutes.

12. A method for the detection of Salmonella serotypes in a specimen comprising immobilizing bacteria in the specimen in a solid phase, adding a quenching agent to mask unutilized binding sites on the solid phase, adding a Salmonella H antisera preparation according to any one of claims 1 to 3, and detecting the presence of bound antibodies.

13. A method as claimed in claim 12 in which the specimen is cultured prior to immobilization.

14. A method as claimed in claim 12 or claim 13 in which the bacteria is immobilized on titanous hydroxide.

15. A method as claimed in claim 12 in which the presence of bound antibodies is detected by radioimmunometric assay.

16. A method as claimed in claim 12 in which the presence of bound antibodies is detected by enzyme immunometric assay.