Vaccine formulation

FIELD OF THE INVENTION
This invention relates to a novel vaccine formulation shown to have enhanced immunostimulating properties.
BACKGROUND OF THE INVENTION
Early in the history of vaccine development, it was discovered that addition of foreign substances, including particulate matter, to vaccine formulations increased the immune response to such vaccines. Freund reported use of water-in-oil (W/O) emulsions to enhance antibody response to an antigen. Addition of inactivated Mycobacteria to such emulsions (Freund's complete adjuvant or CFA) further enhances the cell-mediated response and may or may not enhance the humoral response.
The efficacy of vaccine adjuvants appears to be due to a sequestering or concentrating effect on an antigen or to stimulation of an inflammatory response or to both. CFA is not practical other than in laboratory experiments because local inflammation induced by its administration is unacceptably severe. Many other vaccine adjuvants have been employed in W/O as well as in oil-in-water (O/W) emulsions. Such adjuvants include muramyl dipeptide (MDP), inorganic aluminum compounds such as Al(OH).sub.3 and surfactants such as trehalose dimycolate (TDM), lipopolysaccharide (LPS)-lipid A, certain hydrophilic block polymers and many alkyl nitrogenous compounds which are highly lipophilic and water insoluble.
Useful surfactants have a small polar region attached to a large non-polar region; they interact at the interphase of hydrophilic and lipophilic surfaces in W/O and O/W emulsions. Because such surfactants typically cannot reduce the surface tension of immiscible liquids, i.e., oil and water, to form an emulsion, emulsifying agents are generally added. Commonly employed emulsifiers are the Tweens and the Spans.
Freund, Adv. Tuberc. Res. 7: 130 (1956), reviews use of water-oil emulsions as adjuvants.
Saponins, or sapogenin glycosides, are a type of glycosidic surfactant widely distributed in plants. Use of various saponin preparations as vaccine adjuvants, though not in W/O or O/W emulsions, is well-known.
Quil A is a saponin isolated from the bark of Quillaja saponaria Molina, a tree native to parts of South America. It consists of a carbohydrate moiety in glycoside linkage to the triterpenoid, quillaic acid. Its use as a vaccine adjuvant is well established. See, for example, Dalsgaard et al., Acta vet. Scand. 18: 349-360 (1977).
Bomford, Int. Arch. Allergy Appl. Immunol. 67(2): 127-131 (1982), reports use of cholesterol-saponin complexes as vaccine adjuvants. Bonati, U.S. Pat. No. 4,101,642, discloses use of saponin-sterol complexes, for example, aescin, Polygala saponins, tomatin or digitonin complexed with cholesterol, in pharmaceutical compositions.
Charlier et al., Arch. Exp. Vet.-Med. 27: 783 (1973), report studies with various saponin preparations.
SUMMARY OF THE INVENTION
It has now been discovered that a vaccine antigen which is dually adjuvanted with an oil and a saponin is highly effective in stimulating an immune response in an animal. Thus, this invention, in one aspect, in a vaccine which comprises an effective non-toxic amount of an immunostimulating antigen and an immunostimulating, non-toxic amount of a saponin in a water and oil emulsion.
In another aspect, the invention is a method for preparing such vaccine and a method for stimulating an immune response in an animal which comprises internally administering a vaccine of the invention to the animal. These and other embodiments which are fully disclosed hereinbelow are further aspects of a single invention.
DETAILED DESCRIPTION OF THE INVENTION
The vaccine formulation of the invention employs a saponin and an oil as an adjuvant. It is a parenterally administerable mono- or polyvalent vaccine in which the antigen is any antigen or antigenic component for stimulating a desired immune response. Such vaccine can be, for example, a vaccine for protecting a mammal against infection, or against development of a disease state resulting from infection, by a pathogenic or opportunistic bacteria, virus, parasite or other invasive microbe or organism. Such vaccine can also be, for example, a hormone such as leutenizing hormone. In the former case, the antigen can be one or more modified or inactivated bacteria, viruses, parasites or other microbes or organisms or one or more subunits thereof or derivatives of such subunits.
The vaccine is formulated to comprise a vaccinal amount, that is, an effective, non-toxic amount of each antigen per dose in accordance with standard procedures for vaccine preparation employing an O/W or W/O emulsion. Typically, this comprises adding an immunostimulating antigen and a saponin to the oil or water phase of an oil and water emulsion prior to combining the oil and water. Usually, the antigen and saponin are added to the water. A saponin is typically added to the aqueous phase in an amount of 15 to 5000 micrograms, preferably 25 to 1000 micrograms, per dose. Any of the saponins or saponin derivatives can be used. See, e.g., Charlier et al., Arch. Exp. Vet.-Med. 27: 783 (1973) and Bonati, U.S. Pat. No. 4,101,652. Preferably, such saponin has lipophilic and hydrophilic regions and therefore can function as a surfactant and emulsifier. Quil A is the preferred saponin. Quil A forms micelles in aqueous solutions at concentrations as low as 0.03% and forms complexes with a wide range of antigens. It is publicly available from a variety of sources including commercial vendors, such as Superfos (Copenhagen, Denmark).
The oil phase comprises one or more parenterally tolerated oils. These include vegetable oil such as soy bean oil and peanut oil, mineral oils, such as Drakeol 6VR, animal oils such as squalene, and intermediate length (C.sub.12 to C.sub.20) alkanes, optionally substituted, such as hexadecane. See, for example, Murray et al., Ann. Allergy 30: 146 (1972). The amount of oil is up to 95% by volume. In O/W emulsions, the amount of oil is preferably 0.2 to 20%, more preferably 0.5 to 10% and in W/O emulsions the amount of oil is preferably 40 to 90%, more preferably 50 to 70%.
The compartmentalization of antigen with Quil A in an O/W or W/O emulsion is believed to make this delivery syste a potent immunostimulator. Since Quil A is an emulsifier, parts of each molecule of Quil A are lipophilic and other parts are hydrophilic. Quil A would therefore collect at the oil-water interface. The antigen, if complexed with Quil A, may also collect at this interface. If the antigen itself possesses hydrophilic and lipophilic characteristics it would be positioned at the interface regardless of whether or not it was attached to Quil A. If the antigen is entirely hydrophilic or entirely lipophilic and not attached to Quil A, it would be in the aqueous or oil phase respectively. When the antigen is in the oil phase, Quil A would be on the surface of the oil droplet. When the antigen is in the aqueous phase Quil A would either be in the water droplet with the antigen or on the surface of the water droplet. In either case, antigens and adjuvants are compartmentalized. Such compartmentalization appears to be a highly efficient method to deliver antigens and adjuvant to immune cells, particularly macrophages.
Quil A is also usually added to the aqueous phase when making W/O emulsions containing 50 to 70% oil. These emulsions are made by adding the aqueous phase while stirring into the oil phase or emulsifying with a high shear mixer emulsifier.
The immunostimulating antigen in an agent which upon internal administration to a mammal induces a humoral or cell mediated response to the antigen. It may be an attenuated or inactivated bacteria, virus or parasite or a polypeptide. The polypeptide can be, for example, an oligopeptide, a polypeptide or a protein derived from such organism or virus and optionally purified and may be glycosylated or otherwise modified. Alternatively, such antigen can be prepared by synthetic or by recombinant DNA techniques. The antigen can also be a protein of different origin, such as a hormone or a derivative thereof. Exemplary antigens are antigens which stimulate an immune response to Canine Parvovirus (CPV), Pseudorabies Virus (PSV), Feline Infectious Peritonitis Virus (FIP), Dirofilaria immitis and leutenizing hormone.
The vaccine formulation is administered internally, for example, intramuscularly, subcutaneously, intraperitoneally or intravenously, to a mammal in accordance with standard protocols for the antigen or antigens of choice. For a non-human mammal, the vaccine is typically administered intramuscularly or subcutaneously in a dose volume of 0.5 to 4 ml, preferably 1 to 2 ml. A typical vaccination protocol often calls for a second vaccination 2 to 6 weeks after the initial vaccination, followed by boosters at regular intervals, for example, every 1 to 2 years.
Additional excipients can be added as desired. For example, sterols have been complexed with saponins to reduce hemolysis associated with administration of saponins in aqueous solution; emulsifying agents such as Tweens and Spans are often employed in O/W or W/O emulsions; Montanide 888 is a monooleate commonly employed as an emulsifying agent. Additional agents or adjuvants, such as muramyl dipeptide, trehalose dimycolate, alumina, LPS and alkylamines, can also be comprised within the vaccine, although addition of further adjuvants would not be expected to enhance further the immune response.

EXAMPLES
The following examples are illustrative, and not limiting, of the invention. Drakeol 6VR, Peen-Drake, Chicago, was employed as the oil phase. To prepare 0.5 to 10% O/W emulsions, filtered sterilized Quil A (25-1000 .mu.g/dose) was added to antigen in the aqueous phase. Emulsions were made by adding the aqueous phase to filter-sterilized oil while stirring or by layering the oil over the aqueous phase and emulsifying with a high shear mixer emulsifier (Silverson, Mixtronics Inc., Cleveland, Ohio).
Antigens that were combined with this Quil A emulsion were: (1) canine parvovirus; (2) pseudorabies virus subunit; (3) feline infectious peritonitis virus; and (4) Dirofilaria immitis (canine heartworm). The Examples demonstrate an unexpectedly, enhanced immune response when animals were vaccinated with vaccine formulations of the invention.
EXAMPLE 1
An inactivated canine parvovirus (CPV) vaccine was prepared by propagating a virulent CPV strain in a stable canine cell line, inactivating virulent virus harvested from the cell culture by treatment with beta-propiolactone (BPI) or a beta-ethylamine (BEI) and combining the inactivated virus with following adjuvants by addition to the water phase prior to emulsification: (1) Five percent mineral oil and Quil A (300 micrograms/one ml dose), or (2) a W/O emulsion containing Montanide 888 as an emulsifier. A third group contained guinea pigs vaccinated with a nonadjuvanted vaccine.
Quil A in a 5% O/W emulsion enhanced the antigenicity of CPV much more so than 5% of the emulsion without Quil A. In the latter vaccine, Montanide 888 instead of Quil A was used as the emulsifying agent. (Table 1).
In another experiment, guinea pigs vaccinated with inactivated CPV adjuvanted with the Quil A and O/W emulsion developed higher serum neutralization (SN) titers especially after the first vaccination than did guinea pigs vaccinated with either inactivated CPV adjuvanted with 5% Alhydrogel (aluminum hydroxide) or with nonadjuvanted vaccine (Table 2).
TABLE 1______________________________________Comparisons of the Adjuvanticity of a 5% Oil-in-Water EmulsionWith and Without Quil A on Inactivated CPV Reciprocal of Guinea Pig Virus Neutralization Titer atAdjuvant No. 3 Week Postvaccination______________________________________5% Oil 1 2048Quil A.sup.1 2 512 3 2048 4 1024 5 2048 6 512 7 64Geometric Mean 7615% Oil 8 64Montanide 888 9 32 10 .ltoreq.32 11 .ltoreq.32 12 32 13 32 14 .ltoreq.32Geometric Mean .ltoreq.35______________________________________ .sup.1 300 .mu.g/1 ml dose.
TABLE 2______________________________________Effect of Different Adjuvants on the Response ofGuinea Pigs to Inactivated Canine Parvovirus Vaccine Reciprocal of Virus Neutralization Titer Guinea 14 Days Post Pig 21 Days Post SecondAdjuvant No. First Vaccination Vaccination______________________________________5% oil 1 2048 8192Quil A.sup.1 2 512 8192 3 2048 ND 4 2048 4076 5 8192 8192 6 512 8192 7 256 16384Geometric Mean 1248 81925% Alhydrogel 8 256 2048 9 1024 4096 10 512 4096 11 512 8192 12 64 ND 13 128 2048 14 256 NDGeometric Mean 283 3566None 15 64 512 16 .ltoreq.64 4096 17 .ltoreq.64 ND 18 .ltoreq.64 1024 19 .ltoreq.64 ND 20 .ltoreq.64 2048 21 .ltoreq.64 1024Geometric Mean .ltoreq.64 1351______________________________________ .sup.1 300 .mu.g/ml (dose = 1 ml)
EXAMPLE 2
A pseudorabies subunit vaccine was prepared by attentuating a strain of the virus by serial passage on an established porcine kidney cell line, fractionating cell culture medium on a lectin affinity column, eluting the vaccine antigen with alpha-mannoside and then combining the antigen with the following adjuvants, prior to emulsification: (1) Quil A, (2) Quil A and Bordetella bronchiseptica, and (3) a W/O emulsion containing Montanide 888 and Quil A. The pseudorabies subunit antigen adjuvanted with the W/O emulsion containing 1 mg Quil A/one ml dose stimulated much higher ELISA titers to pseudorabies virus in pigs than did Quil A alone or Quil A and Bordetella (Table 3). This adjuvant was the only adjuvant that stimulated an antibody response following the primary vaccination. This same vaccine antigen-adjuvant combination stimulated the best cell-mediated immune (CMI) response to pseudorabies antigens (Table 4). In another experiment, the W/O emulsion containing Quil A and Montanide 103 was a better adjuvant than 5% alhydrogel for inactivated pseudorabies virus. ELISA titers were much higher following first and second vaccinations and challenge (Table 5 ). In a further experiment, the CMI response to pseudorabies virus was also much higher in pigs inoculated with the vaccine containing the W/O emulsion with Quil A and Montanide 888 than with the vaccine containing alhydrogel (Table 6).
TABLE 3__________________________________________________________________________ELISA Titers of Pigs to a Pseudorabies VirusSubunit Vaccine Combined with Different Adjuvants Reciprocal of ELISA Titer Pig Days Post First Vaccination Days Post Second VaccinationAdjuvant No. 0 7 14 21 7 14 28__________________________________________________________________________Quil A.sup.1 837 0 0 0 0 1620 540 60 838 0 0 0 0 1620 1620 180 839 0 0 0 0 1620 4860 180 840 0 0 0 0 540 540 180 841 0 0 0 20 1620 4860 540 842 0 0 0 0 540 540 180 843 0 0 0 0 180 180 60Geometric Mean 0 0 0 2 865 1012 154Quil A.sup.1 851 0 0 0 0 540 180 60Bordetella 852 0 0 0 0 540 180 180 853 0 0 0 0 60 180 60 854 0 0 0 0 540 540 60 855 0 0 0 0 540 540 180 856 0 0 0 0 540 180 60 857 0 0 0 0 540 540 60Geometric Mean 0 0 0 0 394 288 8250% Oil 865 0 0 0 540 4860 4860 4860Quil A.sup.1 866 0 0 0 180 4860 4860 540Montanide 888 867 0 0 0 1620 43740 14580 4860 868 0 0 0 540 4860 14580 4860 869 0 0 0 20 1620 4860 540 870 0 0 0 0 4860 14580 540 871 0 0 0 540 14580 14580 4860Geometric Mean 137 6652 9105 1895__________________________________________________________________________ .sup.1 1 mg Quil A per dose of vaccine.
TABLE 4__________________________________________________________________________Lymphocyte Blastogenesis Response of Pigs Vaccinated Witha Pseudorabies Virus Subunit Vaccine Combined WithDifferent Adjuvants to Pseudorabies Virus Net Disintegrations Per Minute Pig Days Post First Vaccination Days Post Second VaccinationAdjuvant No. 0 7 14 21 7 14 28__________________________________________________________________________Quil A.sup.1 837 20 408 160 1445 786 94 2049 838 49 0 1326 585 1332 0 926 839 0 223 1688 303 835 5852 1557 840 337 0 164 311 2491 124 430 841 0 2481 2356 467 2029 1947 1285 842 36 4409 2451 1084 4754 171 2980 843 74 3614 3454 0 988 140 334Mean 74 1591 1657 599 1888 1190 1366Quil A.sup.1 851 0 0 0 909 6757 46 728Bordetella 852 0 0 287 0 5179 0 13581 853 0 0 0 2263 11345 225 43 854 0 0 0 5667 9945 929 672 855 0 0 0 2721 6251 773 7178 856 101 238 98 580l 3113 802 14134 857 0 0 801 120 292 280 90Mean 14 343 169 2497 6126 436 520450% Oil 865 0 17355 48778 1797 540 9816 84805Quil A.sup.1 866 88 6417 15583 3586 27565 18853 29046Montanide 888 867 119 1376 52557 5275 30235 76221 74056 868 0 9291 6718 6898 6915 0 34344 869 0 659 4847 1216 9730 6708 9329 870 38 4567 41979 1825 34144 124,420 74545 871 0 4690 47991 6771 31519 78957 109,575Mean 35 6336 31208 3867 20731 44996 59386__________________________________________________________________________ .sup.1 1 mg Quil A per dose of vaccine.
TABLE 5__________________________________________________________________________ELISA Titers of Pigs Vaccinated withan Inactivated Pseudorabies Vaccine Combinedwith Two Different Adjuvants Reciprocal of ELISA Titer at Days Post Pig 1st Vaccination 2nd Vaccination ChallengeAdjuvant No. 0 7 14 21 7 14 7 14__________________________________________________________________________Alhydrogel 956 Neg. Neg. 20 20 540 20 4860 14580(5%) 961 Neg. Neg. 180 180 1620 540 14580 14580 964 Neg. Neg. 60 60 180 540 4860 14580 905 Neg. Neg. 60 60 540 540 4860 4860 968 Neg. 20 20 20 540 540 4860 4860Geometric Mean Neg. 2 48 48 540 279 6054 939567% Oil 952 Neg. Neg. 540 1620 14580 14580 14580 14580Quil A.sup.1 954 Neg. 60 540 540 14580 14580 14580 43740Montanide 103 955 Neg. 20 540 540 4860 14580 14580 43740 959 Neg. 20 180 540 14580 14580 43740 14580 960 Neg. Neg. 540 1620 14580 14580 14580 14580Geometric Mean Neg. 8 433 838 11704 14580 18163 22626__________________________________________________________________________ .sup. 1 1 mg Quil A per dose of vaccine.
TABLE 6__________________________________________________________________________Lymphocyte Blastogenesis Response of Pigs Vaccinated with EitherInactivated Pseudorabies Virus VaccineCombined with Two Different Adjuvants Net Disintegrations Per Minute Days Post Days Post Days Post Pig 1st Vaccination 2nd vaccination ChallengeAdjuvant No. 7 14 21 7 14 7 14__________________________________________________________________________Alhydrogel 956 1546 8547 287 1133 0 12429 3580(5%) 961 1142 9391 4979 11040 15212 5937 9071 964 4274 7910 284 2858 13950 4269 11228 965 2352 11237 953 14647 29430 4117 41780 958 490 5391 3330 21236 21918 8535 37196Mean 1961 8495 1967 10183 16503 7077 20571Killed Whole 952 20074 77039 45123 60763 25930 10375 38830Virus Quil A.sup.1 954 6508 97956 64158 59615 21567 14118 19263Drakeol oil/ 955 7081 39984 20685 14057 18323 44796 22402Montanide 103 959 2152 28225 84912 64995 39485 49584 23191 960 5164 58585 64816 77443 26834 9899 23272Mean 8196 60358 55939 55375 26428 23954 25392None 984 177 ND ND 985 170 202 1145 986 53 644 8489 987 141 ND ND 988 7 ND NDMean ND ND ND ND 110 423 4817__________________________________________________________________________ .sup.1 1 mg Quil A per dose of vaccine.
EXAMPLE 3
A feline infectious peritonitis (FIP) peplomer vaccine was prepared by propagating a virulent strain of FIP on an established stable cell line, fractionating the cell culture medium on a polyacrylamide gel to adsorb the vaccine antigen and eluting and combining the antigen with either Quil A in a 5% O/W emulsion, Quil A in a 67% W/O emulsion, or Quil A combined with polyol L121 in a 5% O/W emulsion, prior to emulsification. Twenty-five ug of Quil A was used in each adjuvant preparation. Quil A combined with either the 5% O/W, W/O emulsion, or 5% O/W emulsion and L121 stimulated a better antibody response as measured by ELISA to the FIP peplomer than did peplomer and Quil A alone (Table 7).
TABLE 7______________________________________Antibody Response of Cats to FIPVPeplomer Following Vaccination withPeplomer and Different Adjuvants Cat ELISA Absorbance .sup.1 at 3 WeeksAdjuvant No. Post Third Vaccination______________________________________Quil A.sup.2 TI5 .544 TY3 .481 TY4 .394 UG3 .413Mean .458Quil A.sup.2, Polyol L121 TI2 1.0375% Oil TY2 .764 UH2 .919Mean .90667% Oil, Quil A.sup.2, TY5 .682Montanide 103 UG2 .583 UH1 .797Mean .6875% Oil, Quil A.sup.1 TI3 .961 TY1 .903 UG4 .682Mean .848______________________________________ .sup.1 Absorbance at 405 nm. .sup.2 25 .mu.g Quil A per dose of vaccine.
EXAMPLE 4
Dogs were vaccinated with canine heartworm antigens from an extract of adult heartworms adjuvanted with either 300 .mu.g/ml Quil A and 5% oil, 300 .mu.g Quil A and a 50% W/O Montanide 888 emulsions, Quil A, 5% oil and Polyol L121, or Quil A alone. All groups of dogs showed strong antibody responses to heartworm antigens as measured by a passive hemagglutination test (Table 8).
TABLE 8__________________________________________________________________________Effect of Adjuvant, Route of Vaccination and AntigenConcentration on the Antibody Titer of Dogs toD. immitis Antigens Geometric Mean Haemagglutination Titer.sup.1 Antigen Days Post First Days Post Second.sup.2 Concentration Vaccination VaccinationAdjuvant Route ug/ml 0 7 14 21 7 14 21 28 56__________________________________________________________________________Quil A.sup.3 Sc 100 21 73 203 128 1171 169 256 81 127 Sc 500 46 102 335 106 671 335 128 128 78 IM 100 64 46 367 233 2130 >876 305 323 274 IM 500 40 84 465 335 1691 398 147 106 18145% Oil/5% Sc 100 47 133 738 465 2344 671 .gtoreq.533 335 301Montanide 888/ Sc 500 37 73 369 133 1172 316 406 116 145Quil A.sup.3 IM 100 40 106 465 671 1861 641 .gtoreq.465 369 .gtoreq.409 IM 500 58 81 406 406 1024 465 168 406 2935% Oil Sc 100 40 58 738 256 >3381 533 423 335 286Polyol L121 Sc 500 53 46 256 161 2344 335 233 161 142Quil A.sup.3 IM 100 64 18 211 168 1172 369 185 106 105 IM 500 37 46 266 84 930 161 67 553 475% Oil Sc 40 .sup. 20.sup.4 55 443 543 1253 887 627 512 192Quil A.sup.3 IM 40 .sup. 23.sup.4 181 .gtoreq.543 .gtoreq.384 326 617 314 362 157__________________________________________________________________________ .sup.1 Geometric mean of 3 dogs. .sup.2 Dogs were revaccinated 21 days post first vaccination. .sup.3 300 .mu.g Quil A per dose of vaccine. .sup.4 Geometric mean of 2 dogs.
EXAMPLE 5
A Moraxella bovis bacterin was prepared substantially as described by Gerber, European patent application No. 146,523. The bacterin was adjuvanted with (i) 5% Drakeol 6VR and Quil A (500 .mu.g/ml) or (ii) with Quil A (500 .mu.g/ml) alone. Bovine animals were vaccinated twice, two weeks apart, and then were challenged. Three out of 7 vaccinates receiving the bacterin with Quil A alone developed IBK. Only 1 out of 7 vaccinates receiving the bacterin with Quil A and oil developed IBK. Of six unvaccinated animals, four developed IBK. The bacterin adjuvanted with Quil A and oil maintained a stable homogeneous composition.
These examples demonstrate the utility of Quil A in an oil and water vaccine formulation and the enhancement in cellular and humoral immune response achievable with such formulations over other oil and water emulsions or formulations comprising other adjuvants. It is particularly surprising that such Quil A-oil formulation would induce a greater immune response than Quil A with other adjuvants, e.g., Bordetella.
The above descriptions and examples fully describe the invention and the preferred embodiments thereof. The invention, however, is not limited to the precise constructions herein disclosed but, rather, includes all modifications and improvements encompassed within the scope of the following claims.

Number	Date	Country
109942	May 1984	EPX
180564	May 1986	EPX

Vaccine formulation

Information

Patent Number

Date Filed

Date Issued

Inventors

Original Assignees

Examiners

Agents

CPC

US Classifications

Field of Search

US

International Classifications

Abstract

Description

Claims

US Referenced Citations (1)

Foreign Referenced Citations (2)

Non-Patent Literature Citations (8)

Entry
Egerton et al., "Effect of Quil A . . . ", Vet. Sci. Communic. 2(1978) 247-252.
Scott, et al., "Adjuvant . . . Glycoprotein", Int. Archs Allergy Appl. Immun. 74: 373-377 (1984).
Maharaj, et al., "Immune Responses . . . Saponin", Can. J. Microbiol. vol. 32, 1986.
McColm, et al., "Comparison of Saponin . . . ", Parasite Immunol. 1982 4, 337-347.
Freund, Adv. Tuberc. Res. 7:130 (1956).
Dalsgaard et al., Acta vet. Scand. 18:349-360 (1977).
Bomford, Int. Arch. Allergy Appl. Immunol. 67(2): 127-131 (1982).
Charlier, et al., Arch. Exp. Vet.-Med. 27:783 (1973).