CA1061252A - Antigenic compositions - Google Patents

Abstract

ABSTRACT

A vaccine comprising an antigen selectively adsorbed onto erythrocytes possessing their natural surface characteristics.
When the antigen is the K88a,b/adhesion factor of E. coli and the erythrocytes are chicken erythrocytes, the vaccine is especially suitable for reducing the incidence of neonatal diarrhoea in pigs if it is administered parenterally to a sow about three weeks prior to parturition.

Description

Tllis invention relates to nntige1lic colrlpositions.
The invention provic~es an anti~enic compositi~n comprising an antigen, in particular an enteropa-tho~enic bac-terial adhesion factor, selec-tively adsorbed onto unmodified erythrocytes (red blood cells), and arises from our discovery that certain um~lodified erythocytes possess selective receptor sites on -their surfaces and thereby are able to form complexes ~ith particular proteins. By "unmodified erythrocytes" ~e mean erythrocytes tha-t possess their natural surface characteristics, as distinct from modified erythrocytes, for e~ample erythrocytes ~hose surfaces have been chemically modified by con-tact with materials such as formaldehyde, tannic acid or glutaraldehyde, and hence have lost their natural ability to selectively form protein complexes. For any given antigen, there may be very fe-~ types of erythrocytes - perhaps only one - that are capablc of selectively adsorbing that particuLar antigen.
For any given antigen, the appropriate erythrocytes can be found by screening possible erythrocytes types to find those .
that are agglutinated by an aqueous solution of the antigen, adopting a procedure such as lS described later herein.
For simplicity, the invention will be descri~ed in detail in relation to the enteropathogenic strains of Eo coii , which are specially implicated--in-neonatal diarrhoea of pigs.`~;

2~ Infection of the newborn piglet with these stralns can lead rapidly to dehydration and dea-th of the piglet, and thus ~" .

- . , . , , : , ~, . . , . : , .. :

c0~112 represents a serious problern :in the prc)duction O f pi~S for ~`ood. Among -the enteroto~in-proclucing strains most frequently encountered in neonatal diarrhoea are the follo~ing:
.: , International_Serot~pe Classification 08:K87 (B), I~88a,b (L) 08 1~87 (B), K88a,c ~L) 0~5a,c:K 'E65', K88a,c (L) 0138:K81 (B), K88a,c (L) 0141 K85ajb (B), I~88a,b ~L) 0147:K89 (B), K88a,c (L) 014~ K91 (B), K88a,c-(L) 0157 K 'V17', ~88a,c (L) ~-In neonatal diarrhoea of the pig, the an-terior small ; intestine is colonised by one or more of such strains, and it has been postula-ted that their resistance to removal from the small intestine by peristalsis is due to adhesive substances present on the surface of the bacterium. Furthermore, it has been noted that all -these strains synthesise one or other of two ~losely rela-ted protein surface antigcns designated K~8a,b and K88a,c, which are substantially insoluble between p~ 3.5 and 5.5, are of high molecular weight (sedimentation .
coefficient about ~5S), and are heat-labile, being denatured when heated above 70C. It has been suggested that it is these substances that are responsible for the gut-adhesive properties of the enteropathogenic strains of E. coli .

:

, . : , : : ., . ~ ~ , . . . . .

cO. 1.~
~ 5 ~
'L~lere clre in -t}ie ~it~rahlre ~c~riv~ls r~erences tv the e~perim~ntal administrntivn of v~cci~es (ie antigen composi-tiolls) containing I~88 antigen to pregnant so~s in order that aclclitional protective antibodies might be generated in the sow's circulatory immune system and, on the so~v's farrowing, might pass to the colostrum (first milh) and so be made available at the site of E. coli colonisation in the gut of the suc~ling pigle-t. Ho~e~er, it is accepted (see J M Rutter "Escherichia coli infect:ions in piglets: Pathogenesis, virulence and vaccination" in The ~eterinar~ Record, 22 February 1975, 96, 171-175) that variab:le results have been obtained ~ith such vaccines: experiments ~ith them sometimes produced positive resul-ts, and sometimes did notO
The present invention provides improved antigenic compositions for administration to pregnant sows, and arises from our discoveries that the K88a,b/adhesion factor is strongly aclsorbed from aqueous meclia onto certain unmodified erythrocytes (red blood cells), particularly those of the chicl~en; that the complex thus produced is strongly immunogenic;
and that the antibodies produced on parenteral administration - o the complex inhibit the adhesion in the p:iglet gut o~ all the E. coli strains specified earlier, ~hether the K88 antigen present on the surface of those strains is KS8a,b or K88a,c.
~ccording to one embodiment of the invention, therefore,~
- 25 there is provided an antigenic composition comprisin~
unmodified erythrocytes on ~vhich is adsorbed 1~88a,b/adhesion -~
factor.

.. ~ ~' .
,.~ ' . ' . ~ : .: , . ," . . . ..

cO. 1.1,"
: ~o~
Preferal)ly, tlle rcl~io oL adsorbe(l]C88a,b/a(lhe~sion fac-tor to tll~ vol~lme of` erytl~rocytes in -the composition is in the range of 125-800 mits per rnl; ancl a ra-tio in -the range of ~L00-800 units per rnl is particularly preferred~ Adhesion factor in excess of the ratio 800 units per ml may be present in the composition, but the excess is for the most part not adsorbed on the erythrocy-tes and is in~unogenically not utilised so efi`iciently as the adsorbed material. The units referred to (haemogglutination units) are measured by a procedure adapted from one that is conventional in immunology, anc] is illustrated in Example 1 la-ter in this specification.
The antigenic composition may be a simple aqueous one, or may contain a phase of adju~ant oil, such as mineral oil or arachis oil, ~hose presence enables an enhanced antibody response (enhanced by comparison with that from the simple aqueous composition3 to be obtained on adminis-tratiorl by injection. In preferred form, -the composition is a multiple oil-~ater emulsion, of the kind generally disclosed in British Patent Specification No. 1,080,994. In this form, the continuous phase is ~n aqueous one ~such as sterile wa-ter, physiological saline or phosphate buffer), and the antigenic material (in our case the K88a,b/adhesion factor - red blood cell complex) is present in a further (secondary) aqueous phase ~hich is itself dispersed in a primary disperse phase of oil.
To prepare the antigenic composition, a cell-free aqueous solu-tion contain:ing the K88a,b/aclhesion fac-tor is brought into contact ~ith unmodified erythrocytes, particular:Ly u~nodified ,~ . .
. .
. - . . . ._" .. . . . . . .
, ; . , :-, . , . , : . .

:'. ' : ' ,~ , :, ; c0.~2 ch:iclcen ery~hrocy-tes, uncler con(litions SUCII -tllat the I~8~a,b/
adilesion f`actor hecomes adsorbed by the erythrocytes. ~le cell-free aqueous solution can :itsel~ be prepared ~rom a K~8a,b s-train (con~eniently -the en-teroto~ic strains G7 or G68 type 1, obtainable from the Central Veterinary Laboratory, Ministry of Agriculture and F'isheries, Nes~ Haw, l~eybridge, Surrey U~) Follo~ing generally the procedure clescribed by - Stirm, Ors~ov, Ursl~ov and Mansa in J. Bacteriology, 93J
731-739~ These procedures entail the release of K88 material from the surfaces o~ the bac-teria into ~ater, as by heating the live bacteria in an aqueous medium buffered to pH 6-9 to 60-65C for 15 minu-tes or more, or by applying ric-tion to the surfaces o~ the bacter:ia in the aqueous buffer, for example in a ~aring Blendor. Bacteria are then removed, by filtration or centrifugation, from -the aqueous ~.~8 e~trac-ts thus obtained~
The K88 material present in the cell-free extracts can, if desired (as for assay purposes~, be purified by repeated soelectric precipitation (pH about 5) and re-solutionO On a large scale, a K88a,b strain o~ E. col:i can be grown on a conventional casein hydrolysate/sucrose medium supplemented with the usual additives such as vitamins, under aeration (eg 3 litres/minute/10 litres of medium), with constant stirring7 pE con-trol (about 7) and temperature control (37C).
The resulting culture is harvested at 24 hours, and centrifuged ~5 to sediment the bacteria. These are then suspended in 0.15M
saline at 3% of -the original culture volume, and are then heated at about 60C to release the adhesion factor into the aqueous medium. Cell debris is removed by centrifugation, and ,~

.. . .

cO.Jl' 5J~
the s~lpe-rna-tan-t Liclui(l-is assclyed for its con~ten~ of adhes:ion factor, anc! the adl1esion ~clc-tor is tlhen adsorbed onto chlcken erythrocytes.
On parenteral adminis-trat:ion of the an-tigenic cornposition to rabbits, sheep, ra-ts, pigs and so forth in an entirely conven-tional manner, antisera are procluced which can be administered orally to newborn piglets to combat neonatal diarrhoea. However, the compositions are best made use of by parenterally aclministering them to pregnan-t sows about 3 weeks before farrowing (that is, about 95 days after service by the boar), so that at parturition -the colostrum contains an effec-tive concentra-tion of antibodies to the K88a,b/adhesion factor of enterotoxic E. coli strains.
The invent:ion is further illus-trated by the folloi~ing ~xamples.
E~IE 1 , a. Preparation and isolation of 1~88a,b/adhesion factor A pure culture of the E. coli strain G7 inter-na-tional serotype classifica-tion 08:K87 (B) -X88a,b (L) was sub-cultured into nutrient broth ancl incuhated at 37C overnight. Yutrlent agar - slopes in Roux flasks ~ere heavily inoculated with the cu7ture and-incubated at 37C for 24 hours.
The confluent surace growth was washed off and harvested aseptically using sterile O.lM phosphate ~uffer (Na2HPO~NaH2PO~) pH 7.5, and the resulting bacterial suspension ~as heated to 60C for 30 minutes to release the K88a,b/adhesion factor from - 7 _ ..

: : . . ~ : : :: :

~ 3~ ~O.IL2 the cell s~lrf'nces. C~llular IrlateriaL WclS removed l~y cen-trlf~l~ation (300()g; 10 minu-tes), and after the addi-tion of sodiurn azide (0.~%) to prevent 'bacterial ~ro~th, the supernatan-t liquicl~as st~red a-t 4C for 3 days. The impurities that had settled out were then removed 'by centrifuga-tion, and dilute '~ acetic acid was added to the continuous1y stirred supernate to reduce the plI to 5. (The isoelectric point of' the K88a,b/ adhesion factor is between 4.5 and 5.5.) The precipitated material was allowed to stand Cor 24 hours at ~C, then collected by centrifuga-tiorl, and washed twice with 0.005 McIlvaine buf~er (Na2~1P0~/citric acid) p~I 5Ø It ; was purified by repeated dissolution (using phosphate-buffered 0.15~ saline at pH 7.2) and precipitation (0.15~1 McIlvaine buffer, pH 5.0). Finally, the precipitated material ~as clissolved in plI 7.2 ` phosphate-buffered 0.15~1 saline ~and stored a-t -20C. -b. Assay of K88a,b/ac1hesiQn factor in solution of unknown concentration This assay can be used as a screening test to ' ~
identify those unmoclified erythrocytes -that are ' ~' capable of selectively adsorbing a given antigen.
A~portion (0.05 ml) of the solutiorl (Y)-to be ~' assayed is serially diluted with phosphate-buf~ered .
saline pH 7.1 in successive depressions ~wells) of a mictotitre plate to obtain a series of solutions of adhesion factor concentration 1/2, 1/4, 1/8 ...

- 8 ~

~ rj~ c~.lL~
, 1/2n th(lt of tlle so:lution ~'. Lo each of` these solut:ions is acldecl 0.025 ml of a 2% tllr:ice-waslle~
itll phosphate-buf~ered saline pfl 7.1) suspension of chicken ery-throcytes in the same buffer. The plate is mechanically shaken and the ~ell contents are allo-~ed to settle. .~fter 1/2 hour the plate is inspected. Agglutination of the chicken erythrocytes occurs in the stronger solutlons (-tllose in -the earlier filled ~ells? bu-t no-t in the weal~er ones, and the - 10 end point of the titration (assessed at room temperature) is taken as that solu-tion in ~hich haemagglutination only just occurs. In a typical procedure, the end-point might be in the thirteenth ~ell, and the ti$re o~ the solution Y wou:ld then be 213 (=8192, say 8000, corresponding to 8000 haemagglutination (HA) units per ml).

Preparation of simple aqueous chicken erythrocyte vaccine -Following the procedure of lb above, a solution prepared according to a was assayed and found to have a K~8a,b/adhesion ~actor titre oL 8000. ~rom this solution, a vaccine was prepared by first diluting 1 ml o~ the solution to 10 ml wi-th :
phosphate-buffered saline pH 7.1 (PBS) and then stirring the diluted solution (Z) for 1/2 hour at 37C with 10 ml of ~hrice-~
2~ washed (PBS) packed chicken erythrocytes obtained b~ centri-fwgation (lOOOg; LO minutes). This procedure led -to substantial saturation of the chiclcen erythrocyte surfaces wi~th the adhesion ' - 9 - /...
:

CO. ~
s~
fclctor, ie t3-1ere ~!ere at satllr~ltion 8000 1-l1~ units of ~dtlesior factor per 10 ml ol' chicl;erl recl blood cells.
Sodium azicle (0.2% by ~eight) ~as adcled -to the resulting vaccine, which was storecl at ~C un-til required for use.
The vaccine thus obtained is suitable administered to pregnant so~s at a dose level of 2 ml. ~t -this level is has about lO times the antibody-generating effect of the (non-erythrocytes-con-taining) solution Z from which it was made.
Instead of using socliuin azide as preservative as descri'bed above, the vaccine can be treated with formalin (4~/o aqueous formaldehyde solution), suitable in the proportion l volume of forrnalin:100-50 volumes of the vaccine.
E,Y~LE 3 .
This Example illustra-tes the preparation oL vaccines containing an oil acljuvant.
i) l ml of the mannide mono-oleate emulsifier solcl under -the Trade Mark "~xlacel A" was dissolve~
in 9 ml of a pharmaceutical grade mineral oil. The -oil solution ~as then added to an equal volume of the simple aqueous chicken erythrocyte vaccine obtained accorcling to Example 2. Mixing was carried out in a homogeniser, to form a-creamy wa-ter-in-oil - emulsion.
ii) l Volume of the water-in-oil vaccine of i) was added to an equal volume of a 2% solution in 0.15M
saline of the polyoxyethylene sorbitan mono-oleate emulsifier soLd under the Trade Mark "'n~een 80"~
Mixing was carried out in a homogeniser.

' ' ' - 10 - /.,, ;:. ~ .. . - .. , . . - . . , . . .: . .. -- . . , : . , ~:' .' '' , . ~

cO.L:l2 The low-vlscosity ~ater-in-oil-in-~va-ter vaccine thus :~ormed ~vas injected :in-tramuscularly at a dose :Ievel o 8 ml illtO SOWS 3 weeks before farrowing (ie about 95 days a~ter service by the boar). The an-ti-adhesive act:ivity thus generated was associated witll three antibody classes (IgG, IgPI and IgA) providing a wide spectrum of ant:ibody activity, and is compared below ~ith the corresponding activity in a control group of untreated pregnant sows.

Anti-adhesive Acti t~
In serum In colostrum Vaccinatecl sows 256 512 Unvaccinated sows ~-8 32 The anti-adhesive activity in the serum of pi~lets suckled by the ~accinated sows was 128 mits; for the serum of pi~lets sucklecl by the ~mtreated sows it was only 32 units.
The piglets suc~led by the vaccinated sows ~ere far more resistant -than those suckled by non-vaccinated sows to infectior by enteroto~ic K88a,b and K88a,c strai-s o' E. coli.

. . ~

-- 11 --- ***

.; .

`;' i , ~ . :.
:. . . .: - ' . ~ : .
-. , ,:, ", . ~ , - , ' : . : ,~
, ' ' '' '., . ' ' '

Claims (3)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE PROPERTY
OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for the preparation of a vaccine for parenteral administration to pregnant mammals as a means for reducing the incidence of neonatal diarrhoea, in which process K88a,b/adhesion factor of an enteropathogenic strain of E. coli is adsorbed from a cell-free aqueous solution onto erythrocytes possessing their natural surface characteristics, and the composition so formed is rendered storable by the addition thereto of a preservative such as sodium azide or formalin.

2. A process as claimed in claim 1, where the ratio of adsorbed K88a,b/adhesion factor to the volume of erythrocytes present is in the range of 125 to 800 haemagglutination units per ml.

3. A process as claimed in claim 1 or claim 2, wherein said erythrocytes are chicken erythrocytes.