WO1998026799A1

WO1998026799A1 - Conjugate vaccine for salmonella paratyphi a

Info

Publication number: WO1998026799A1
Application number: PCT/US1996/019978
Authority: WO
Inventors: Edward Konadu; Shousun Szu
Original assignee: The Government Of The United States Of America, Represented By The Secretary, Department Of Health And Human Services
Priority date: 1996-12-18
Filing date: 1996-12-18
Publication date: 1998-06-25
Also published as: AU1420897A

Abstract

A conjugate vaccine for S. paratyphi A comprising lipopolysaccharide from which lipid A has been removed and substantially all O-acetyl groups have been retained conjugated to a carrier. The vaccine elicits bactericidal antibodies and is useful for prevention of enteric and typhoid fever.

Description

CONJUGATE VACCINE FOR SALMONELLA PARATYPHI A

Field of the Invention The present invention relates to conjugate vaccines against bacterial pathogens. More specifically, the invention relates to a conjugate vaccine against Salmonella paratyphi A comprising the O-specific poiysaccharide bound directly or indirectly to a carrier protein.

Background of the Invention

Salmonellosis remains a serious health problem throughout the world, particularly in developing countries.

Recently, the incidence of infection with nontyphoidal salmonella in the United States has risen (Chalker et al., Rev.

Infect. Dis., 10:111-124, 1988). Salmonella infection can cause enteritis which may be complicated by bacteremia (enteric fever) and extraintestinal complications in both normal and immunocompromised individuals. Treatment of enteric fever, including Salmonella paratyphi A, has become more difficult with the emergence of multiantibiotic- resistant strains (Anand et al., Lancet, 335:352, 1990; Bhutta et al., J. Infect. , 25:215-219, 1992). Although used by many countries for decades, TAB vaccine, which contains inactivated cells of Salmonella typhi anά groups A and

B salmonellae, was removed as a licensed product because efficacy was only demonstrable for typhoid fever. The lipopolysaccharide (LPS) of salmonellae is both a virulence factor and a protective antigen. The virulence is due to the lipid A moiety. Although there are approximately 50 LPS serogroups, almost all extraintestinal infections in humans (enteric fever) are caused by serogroups A, B, C-|, C₂ and D. In Southeast Asia, as in developing countries, the most common cause of enteric fever is S. typhi. The second most common cause in this region is S. paratyphi A, which accounts for approximately 15% of cases. Although widespread after World War II, this serogroup is rarely found in other parts of the world.

The surface poiysaccharide of S. paratyphi A is the O-specific poiysaccharide of its LPS. The 0.-specif4C polysaccharides of serogroups A, B and D salmonellae share a common backbone: →2-α-D-Man/?-(1→4)-α-L-Rha - (1→3)-σ-D-Galp-(1→. The serogroup specificity of S. paratyphi A is conferred by an σ-3,6-dideoxyglucose (σ-D- paratose) (factor 2) linked (1→3) to the mannose of the backbone. The cr-L-rhamnose of the backbone is partially O-acetylated at C-3. An cr-D-glucose moiety is linked (1-»6) to the galactose of the backbone. The proposed structure of the repeating unit of the 0-SP is shown in Figure 1.

The LPS of S. paratyphi A can be detoxified by treatment with acetic acid (Chu et al., Infect. Immun., 59:4450-4458, 1991 ) or hydrazine (Gupta et al., Infect. Immun., 60:3201-3208, 1992). Treatment with acetic acid cleaves the lipid A moiety, leaving the O-specific polysaccharides (0-SP) essentially intact. Conversely, cleavage of LPS with hydrazine destroys the majority of the O-acetyl groups, leaving DeALPS. The conjugation of detoxified LPS from various types of bacteria to carrier proteins, including tetanus toxoid (TT), has been described previously (Chu et al., ibid.; Konadu et al., Infect. Immun., 62:5048-5054, 1994).

There are no effective vaccines for nontyphoidal salmonellae. Thus, there is a need for an effective vaccine against S. paratyphi A. The present invention addresses this need. Summary of the Invention One embodiment of the present invention is isolated Salmonella paratyphi A lipopolysaccharide detoxified by removal of lipid A therefrom, said detoxified lipopolysaccharide containing substantially all O-acetyl groups therein. Preferably, the detoxified lipopolysaccharide contains between about 80 and 100% of the O-acetyl groups; more preferably, the detoxified lipopolysaccharide contains about 100% of the O-acetyl groups.

The present invention also provides a conjugate vaccine composition for Salmonella paratyphi A (SPA), comprising SPA lipopolysaccharide from which lipid A has been removed and in which substantially all O-acetyl groups are retained (O-SP), the 0-SP covalently linked to a carrier. Advantageously, the pharmaceutically acceptable carrier is suitable for parenteral administration. According to one aspect of this preferred embodiment, the carrier is a protein. Preferably the protein is tetanus toxin/toxoid, diphtheria toxin/toxoid, detoxified P. aeruginosa toxin A, cholera toxin/toxoid, pertussis toxin/toxoid, Clostridium perfringens exotoxins/toxoid, hepatitis B surface antigen, hepatitis B core antigen, rotavirus VP7 protein or respiratory syncytial virus F and G protein. More preferably, the protein is tetanus toxoid. The composition may further comprise a linker between the 0-SP and the carrier protein. Advantageously, the linker is adipic acid dihydrazide, N-Succinimidyl-3-(2-Pγridγldithio)propionate, e-aminohexanoic acid, chlorohexanol dimethyl acetal, D-glucuronolactone or p-nitrophenγlethylamine; more advantageously, the linker is adipic acid dihydrazide. Most advantageously, no linker is used and the 0-SP is directly linked to a carrier.

Another embodiment of the invention is a method of preventing S. paratyphi A infection in an individual, comprising the step of administering to the individual an effective immunoprotective amount of the conjugate vaccine composition described above. Preferably, the administering step is intramuscular or subcutaneous. In one aspect of this preferred embodiment, the effective amount is between about 10 μg and about 50 μg.

The present invention also provides a method of detoxifying lipopolysaccharide from SPA, comprising, removing lipid A therefrom and leaving substantially all O-acetyl groups therein. Preferably, the lipid A is removed with an acid. Advantageously, the acid is acetic acid.

Still another embodiment of the invention is a method of making a conjugate vaccine against SPA, comprising the steps of: removing lipid A from SPA lipopolysaccharide and retaining substantially all O-acetyl groups therein to produce 0-SP; activating the O-SP with a chemical activator; and covalently binding said activated O-SP to a carrier. Preferably, the removing step comprises treatment with an acid. Advantageously, the acid is is acetic, pyruvic, propionic, methanesulfonic or hydrochloric. Preferably, the carrier is a protein. In one aspect of this preferred embodiment, the protein is tetanus toxin/toxoid, diphtheria toxin/toxoid, detoxified P. aeruginosa toxin A, cholera toxin/toxoid, pertussis toxin/toxoid, Clostridium perfringens exotoxins/toxoid, hepatitis B surface antigen, hepatitis B core antigen, rotavirus VP7 protein or respiratory syncytial virus F and G protein. Advantageously, the protein is tetanus toxoid. Preferably, the activator comprises CDAP. The method may further advantageously comprise binding the chemically activated O-SP to a linker prior to the covalent binding step. Preferably, the linker is adipic acid dihydrazide, N-Succinimidyl-3-(2-Pγridyldithio)propionate, e-aminohexanoic acid, chlorohexanol dimethyl acetal, D- glucuronolactone or p-nitrophenylethylamine; more preferably, the linker is adipic acid dihydrazide.

The present invention also encompasses the use of a conjugate vaccine composition comprising SPA lipopolysaccharide from which lipid A has been removed and in which substantially all O-acetyl groups are retained (O-SP), the O-SP covalently linked to a carrier, as a vaccine for S. paratyphi A.

Brief Description of the Drawings

Figure 1 is a schematic diagram of the proposed structure of the repeating unit of the O-specific poiysaccharide of S. paratyphi A. Figure 2 is a gel filtration profile of the O-specific poiysaccharide.

Figure 3A shows the region of anomeric carbon signals in the C NMR spectrum of the O-SP in D₂0.

Figure 3B is a C NMR spectrum of O-SP in D2O showing the complete spectrum of carbonyl carbon from an acetamido group and methyl groups.

Detailed Description of the Preferred Embodiments

The present invention includes the observation that the O-SP of S. paratyphi Δ, when conjugated to a carrier protein, elicits serum IgG antibodies with bactericidal activity. The O-SP is the LPS from which the lipid A moiety has been removed by treatment with acid, leaving substantially all of the O-acetyl groups intact. In a preferred embodiment, about 80-100% of the acetyl groups are retained; in a more preferred embodiment, about 100% of the acetyl groups are retained. This results in a detoxified LPS, but one in which a protective epitope is left intact. In contrast, treatment of LPS with hydrazine, which removes most of the O-acetyl groups, results in a polysascharide- which, when conjugated to a carrier protein, did not elicit bactericidal antibodies to LPS.

There is a correlation between in vitro bactericidal activity exhibited by antiserum to conjugate vaccines containing surface polysaccharides of Gram-negative bacteria and in vivo efficacy (protection) thereof. This has been demonstrated for Vibrio cholerae (Mosley, Tex. Rep. Biol. Med., 27(Suppl. 1):227-241, 1969; Haemophilus influenzae B (Fothergill et al., J. Immunol., 24:273-284, 1933) and meningococcal vaccine (Goldschneider et al., J. Exp. Med., 129:1307-1326, 1969). Thus, the conjugate vaccine of the invention will also protect against infection by the Gram- negative S. paratyphi k.

As described herein, the lipid A moiety of S. paratyphi A LPS is removed by hydrolysis with acetic acid. The use of acid is imperative, as hydrolysis with a base (hydrazine) results in an detoxified LPS (DeALPS) which does not elicit bactericidal antibodies. Although the use of any organic or inorganic acid for LPS detoxification is within the scope of the invention, organic acids are preferred. Such acids include, for example, acetic acid, pyruvic acid, propionic acid and ethanesulfonic acid. One preferred inorganic acid is hydrochloric acid. The use of other compounds capable of removing the lipid A component while leaving substantially all O-acetyl groups intact is also within the scope of the invention. The pH of the reaction mixture is chosen such that lipid A is removed, but substantially all acetyl groups are retained. The percentage of acetyl groups and iipid A remaining can be determined by one of ordinary skill in the art (Hestrin, J. Biol. Chem., 180:249-261, 1949; Tsai et al., J. Biol. Standards, 14:25- 33, 1986). In a preferred embodiment, the pH is between about 2 and about 4. In a more preferred embodiment, the pH is about 3. Reaction temperatures for acid-mediated LPS detoxification are typically greater than about 25°C. In a preferred embodiment, the reaction temperature is between about 50°C and about 100°C. In a more preferred embodiment, the reaction temperature is about 100°C. The reaction time is typically between about one and two hours, with 90 minutes being preferred. The reaction must be allowed to proceed at least until the O-SP is no longer toxic as determined by the Limulus amebocyte lysate (LAL) assay as described in Example 6.

The hydrolysis reaction that removes lipid A can also reduce the molecular weight of the LPS which can deleteriously affect the O-SP product. Thus, this reaction is controlled using the reaction conditions described above. In a preferred embodiment, the O-SP has a molecular weight of greater than about 50 kDa. In a more preferred embodiment, the O-SP has a molecular weight of greater than about 100 kDa.

The O-SP is then chemically activated to allow attachment to either a linker molecule which is subsequently bound to a carrier protein, or to allow direct attachment to a carrier protein. Two such contemplated activating reagents are cyanogen bromide (CNBr) and 1-cyano-4-dimethylamino pyridinium tetrafluoroborate (CDAP) (Kohn et al., FEBS Lett., 154:209 -210, 1983), with CDAP being preferred. CDAP is used at neutral pH which preserves the majority of O-acetyl groups on the O-SP. CNBr is used at an alkaline pH, resulting in the loss of a portion of 0- acetyl groups as shown in Example 6. As described herein, higher titers of bactericidal antibodies were obtained when conjugates were prepared using CDAP.

Linkers between polysaccharides and conjugate proteins are frequently used in the production of conjugate vaccines. Although adipic acid dihydrazide (ADH) is the preferred linker, any linker capable of stably and efficiently conjugating O-SP to a carrier protein is contemplated. The linkers typically have two of the same reactive. groups, either amino or carboxy (heterodifunctional linkers). Such linkers include N-Succinimidyl-3-(2-Pyridyldithio)propionate (Fatoom et al., Infect. Immun., 60:584-589, 1992), e-aminohexanoic acid, chlorohexanol dimethyl acetal, D- glucuronolactone and p-nitrophenylamine. Many other linkers known to those of ordinary skill in the art are also suitable for use in the invention and are discussed in detail by Dick et al. [Conjugate Vaccines, J.M. Cruse and R.E. Lewis, Jr., eds., Karger, New York, pp. 48-114). The ADH-derivatized O-SP is then reacted with the carrier-protein in the presence of the selective activator 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC).

For the vaccines of the present invention, it is particularly preferred that the O-SP be directly bound to the carrier protein without a linker. Typically, direct linking comprises incubating CDAP-activated O-SP with the carrier protein at a pH of between about 8.0 and 8.5, followed by gel filtration chromatography. O-SP may also be directly covalently bonded to the carrier protein by using, for example, the cross linking reagent glutaraldehyde.

Other methods well known in the art for effecting conjugation of oligosaccharides to carrier proteins are also within the scope of the invention. Such methods are described in, for example, U.S. Patent Nos. 5,153,312 and 5,204,098; EP 0 497 525; and EP 0 245 045. The O-SP-carrier protein conjugates for parenteral administration may be in the form of a sterile injectable preparation, such as a sterile injectable aqueous or oleaginous suspension. This suspension may be formulated according to methods well known in the art using suitable dispersing or wetting agents and suspending agents. The sterile injectable preparation may also be a sterile injectable solution or suspension in a parenterally acceptable diluent or solvent, such as a solution in 1,3-butaπediol. Suitable diluents include, for example. Ringer's solution and isotonic sodium chloride solution. In addition, sterile fixed oils may be employed conventionally as a solvent or suspending medium. For this purpose, any bland fixed oil may be employed including synthetic mono- or diglycerides. In addition, fatty acids such as oleic acid may likewise be used in the preparation of injectable preparations,

The conjugate vaccine of the invention may be in soluble or microparticular form, or may be incorporated into microspheres or microvesicles, including liposomes, by well known methods. Although various routes of vaccine administration including, for example, intramuscular, subcutaneous, intraperitoneal and iπtraarterial are contemplated, the preferred route is intramuscular. The conjugate vaccine of the present invention is administered to mammals, particularly humans, in an amount ranging from about 10 μg to about 50 μg. In a more preferred embodiment, the amount administered is between about 20 μg and about 40 μg. in a most preferred embodiment, the amount administered is about 25 μg. If desired, one or two booster injections are administered containing between about 10 μg and about 25 μg conjugate. The exact dosage can be determined by routine dose/response protocols known to one of ordinary skill in the art.

The examples describe conjugate vaccines using S. paratyphi A strain NTP-6. Vaccines from other S. paratyphi A strains are within the scope of the present invention and are made using the same techniques. Because all S. paratyphi A strains have the same O-SP, any such strain can be used. S. paratyphi A is available from the American Type Culture Collection, Rockville, MD (catalog no. 9150). A multivalent vaccine comprising a mixture of conjugates from different S. paratyphi A strains is also within the scope of the invention. A person of ordinary skill in the art will appreciate that LPS from these other, clinically relevant strains may be detoxified by removal of lipid A therefrom as described in Example 2.

Carriers are chosen on the basis of facilitating two functions: 1 ) to increase the immunogenicity of the poiysaccharide and/or 2) antibodies raised against the carrier are medically beneficial. Polymeric carriers can be a natural or synthetic material containing a primary and/or secondary amino group, an azido group or a carboxyl group. The carrier may be water soluble or insoluble.

The carrier protein may be any medically acceptable biologically compatible protein capable of enhancing the immunogenicity of the O-SP and not having any adverse affects. A variety of carrier proteins may be used in the conjugate vaccine of the present invention. Such classes of proteins include pili, outer membrane proteins and excreted toxins of pathogenic bacteria, nontoxic or "toxoid" forms of such excreted toxins, nontoxic proteins antigenically similar to bacterial toxins (cross-reacting materials or CRMs) and other proteins. Nonlimiting examples of bacterial toxoids contemplated for use in the present invention include tetanus toxin/toxoid, diphtheria toxin/toxoid, detoxified P. aeruginosa toxin A, cholera toxin/toxoid, pertussis toxin/toxoid and Clostridium perfringens exotoxins/toxoid. The toxoid forms of these bacterial toxins is preferred. Particularly preferred carrier proteins are tetanus toxoid, diphtheria toxoid and detoxified cholera toxin. The use of viral proteins (i.e. hepatitis B surface/core antigens; rotavirus VP 7 protein and respiratory syncytial virus F and G proteins) is also contemplated. CRMs include CRM₁₉₇, antigenically equivalent to diphtheria toxin (Pappenheimer et al., Immunochem., 9:891-906, 1972) and CRM3201, a genetically manipulated variant of pertussis toxin (Black et al., Science, 240:656- 659, 1988). The use of carrier proteins from non-mammalian sources including keyhole limpet hemocyanin (KLH), horseshoe crab hemocyanin and plant edestin is also within the scope of the invention. LPS was purified from S. paratyphi A as described below.

Example 1 Purification of LPS S. paratyphi A was obtained from the blood culture of a patient in Kathmandu, Nepal (Acharya et al., Am. J. Trop. Med. Hyg., 52:162-165, 1995). S. typhimurium (group B) TML (0:4, 12) was obtained from Uniformed Services University of the Health Sciences, Bethesda, MD (Watson et al., Infect. Immun., 60:4679-4686, 1992). Serogroups were identified by agglutination with grouping antiserum and were confirmed by the laboratory of Microbiology, Clinical Center, National Institutes of Health. LPS was purified as described (Chu et al., ibid.; Koπadu et al., ibid.; Watson et al., ibid.; Westphal et al., Meth. Carbohyd. Chem., 5:83-91, 1965). Briefly, formalin- inactivated cells (200 g) were washed twice in pyrogen-free saline (PFS), suspended in pyrogen-free water (PFW) and mixed with an equal volume of 90% phenol at 68°C for 30 min. The suspension was centrifuged at 7,300 x g at 10°C for 1 hour. The aqueous layer was brought to 75% ethanol, and the precipitate was treated with DNase (50 μg) and RNase (50 μg). Proteinase K (Boehringer Mannheim, Indianapolis, IN; 200 mg) was added, and the suspension was dialyzed against Tris buffer overnight at 37°C and against PFW overnight at 3-8°C. The suspension was centrifuged at 64,000 x g at 10°C for 5 hours and freeze-dried (weight, 1.5 g). LPS was detoxified by two methods, acid hydrolysis and hydrazinolysis, as described in the following two examples.

Example 2 Detoxification of LPS by acid hydrolysis This method is described by Chu et al. (ibid.); Konadu et al., (ibid.); Watson et al., (ibid.); and Westphal et al. (ibid.). Briefly, LPS (300 mg in 30 ml of 1 % acetic acid, pH ~ 3) was heated at 100°C for 90 minutes. The pH of the mixture was brought to 6.8 with 3.5 M NaOH, followed by centrifugation at 64,000 x g at 10° jor 5 hours. The supernatant was freeze-dried, suspended in 5 ml of 0.2 M NaCI, and passed through a desalting column (3 x 46 cm) of Biogel P-10 (Bio-Rad laboratories, Richmond, CA) in PFW. Void volume fractions were pooled and freeze-dried. The powder was dissolved in PFW at 20 mg/ml and passed through a column (2.5 x 90 cm) of Sephadex G-75 (Pharmacia LKB), and the void volume fractions were pooled, freeze-dried and designated O-SP. In another experiment, the powder was dissolved in PFW (20 mg/ml) and passed through a column (2.5 x 90 cm) of Sephadex G-75 of CL-6B Sepharose (Pharmacia LKB) (Figure 2). The first (fractions 30-48) and second (fractions 49-65) peaks were dialyzed against PFW, freeze-dried and designated O-SP(HMW) and O-SP(LMW), respectively. Example 3

^C NMR of O-SP and DeALPS C nuclear magnetic resonance spectroscopy (NMR) spectra of O-SP and DeALPS (25 mg/ml D₂0) were recorded on a Varian GN 300 spectrometer at room temperature with a 5-s decay between acquisition, and a 10-μs pulse and approximately 50,000 free induction decay were averaged for each spectrum. Prior to Fourier transformation, a 3-Hz line broadening was applied and zero filled to 32,000 datum points.

¹³C NMR spectra (Fig. 3) for the anomeric carbons showed six signals (104.5, 104, 103.4, 102.5, 102 and 101.4 ppm) for 0-SP (Fig. 3A) and five signals (104.5, 103.5, 102.5, 102 and 101.4 ppm) for DeALPS. The O-SP spectrum also had a signal at 174 ppm, which is characteristic of the carbonyl group from an O-acetyl group, and two signals (21.5 and 21 ppm), which is characteristic of methyl carbons of 0 acetyl substituents (Fig. 3B). These signals were not detected in the DeALPS spectrum. Each spectrum showed one signal (35 ppm) that was characteristic of the deoxy carbon of paratose and two signals (17.8 and 17 ppm) that were characteristic of the methyl carbons from the paratose and rha nose.

Example 4 Detoxification of LPS by hvdrazinolysis

Detoxification by hydrazinolysis is described by Gupta et al., ibid. LPS (300 mg) was dried over P₂0g for 5 days, suspended in 30 ml of anhydrous hydrazine and stirred at 37°C for 2 hours. The solution was placed on ice and cold acetone was added dropwise to a final concentration of about 90%. The resultant precipitate was removed by centrifugation at 15,000 x g at 10°C for 30 min, washed twice with cold acetone, dissolved in PFS ( ~ 5 mg/ml) and centrifuged at 64,000 x g at 10°C for 5 hours. The supernatant was freeze-dried, dissolved in 0.2 M NaCI (5 ml) and passed through a column (3 x 46 cm) of Biogel P-10 in PFW. The void volume fractions were pooled, freeze-dried and designated DeALPS. As with O-SP, DeALPS was passed through CL-6B Sepharose, and the void volume fractions were pooled, dialyzed, freeze-dried and designated DeALPS(HMW).

O-SP and DeALPS were derivatized with ADH as described in the following example. Example 5

Derivatizatioπ of polysaccharides with ADH

O-SP and DeALPS were derivatized by two methods. In the first method, derivatization was performed as described previously (Chu et al., ibid.; Konadu et al., ibid.; Watson et al., ibid.; Westphal et al. ibid.). Polysaccharides

(5 mg/ml in PFS) were brought to pH 10.5-11 with 0.1 M NaOH, and an equal amount of CNBr (1 g/ml in acetonitrile) was added. The reaction was carried out for 6 min on ice, and the pH was maintained at 10.5-1 1 with

0.1 M NaOH. An equal volume of 0.8 M ADH in 0.5 M NaHC0₃ was added, and the pH was adjusted to 8.5 with

0.1 M HCI. The reaction mixture was stirred at 4°C overnight and dialyzed against PFS at the same temperature for 24 hours. The mixture was freeze-dried, reconstituted in 5 ml PFW and passed through a column (3 x 46 cm) of P-10 in PFW. Void volume fractions were pooled and freeze-dried. In method 2, O-SP was activated by CDAP as described (Kohn et al., FEBS Lett., 154:209-210, 1983).

Briefly, a 60 μl volume of CDAP (100 mg/ml in acetonitrile, Sigma, St. Louis, MO) was added to a solution of O-SP (2 ml, 10 mg poiysaccharide per ml PFS) at room temperature. The pH was maintained at 5.8 - 6.0 for 30 seconds, and 60 μl 0.2 M TEA was added to a pH of 7.0. The reaction was carried out for 2 min, and an equal volume of 0.8 M ADH (Sigma) in 0.5 M NaHC0₃ was added. The reaction was carried out for 2 hours and the pH was maintained at 8.0-8.5 with 0.1 N NaOH. The reaction mixture was dialyzed against PFS and passed through a column (3 x 46 cm) of P-10 in PFW. Void volume fractions were pooled, freeze-dried and designated O-SPC.

Example 6 Analyses The molecular sizes of LPS, O-SP and DeALPS were estimated by gel filtration through Superose 12 (Pharmacia) in 10 mM Tris-HCI, pH 8.0, 0.2 M NaCI, 1 mM EDTA, 10 M Tris-HCI, 0.25% deoxycholic acid using dextrans as standards. The degrees of derivatization of 0-SP and DeALPS with ADH were measured with trinitrobenzolsulfonic acid (TNBS) (Chu et al., ibid.). Protein concentrations were measured using the bicinchoninic acid reagent with bovine serum albumin (BSA) as a standard (Gupta et al., ibid.). Hexose amounts were measured by anthrone reaction with O-SP as a standard (Pollack et al., J. Clin. Invest., 63:276-286, 1979). The amounts of O-acetyl groups were measured by the Hestrin reaction with acetylcholine chloride as a standard (Hestrin, J. Biol. Chem., 180:249-261, 1949). LPS was detected by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS- PAGE; Tsai, J. Biol. Stand., 14:25-33, 1986).

The toxicity of LPS, O-SP and DeALPS were measured by the Limulus Amebocyte Lysate (LAL) endotoxin assay and were expressed in endotoxin units (EU) related to the U.S. standard (Hochstein, "Role of the FDA in regulating the limulus amoebocyte lysate tests, p. 38-49, in Clinical application of the Limulus amoebocyte lysate test, R. B. Prior, ed., CRC Press, Inc., Boca Raton, FL, 1990). Equal volumes (100 μl) of samples and Limulus amebocyte lysate were mixed and incubated at 37°C for 1 hour. Gelation of the lysate at the minimal LPS concentration j/vas_ determined by inverting the mixture. A firm gel was considered a positive reaction (Hochstein et al.. Bull. Parenteral Drug Assoc, 27:139-148, 1973). All reagents were from the U.S. Food and Drug Administration, Bethesda, MD. The sensitivity of the LAL assay is 0.09 EU per ml. By this assay, LPS had 7 x 10³ EU per μg and O-SP and DeALPS had 0.70 EU per μg, each representing an approximately 10,000 fold reduction in toxicity.

High performance liquid chromatography (HPLC) of LPS and 0-SP showed four peaks. LPS had K_ds (coefficients of distribution) of 0.1 1 (130 kDa), 0.34 (35 kDa), 0.49 (14.3 kDa) and 0.59 (8.3 kDa). 0-SP had K_ds of 0.13 (122 kDa), 0.41 (23.4 kDa), 0.56 (9.6 kDa) and 0.64 (6.0 kDa). DeALPS had K_ds of 0.12 (128 kDa), 0.37 (31 kDa) and 0.53 (11.6 kDa). Passage through CL-6B Sepharose yielded two fractions with molecular weights for O-SP(HMW) of 122 kDa and for O-SP(LMW) of 23.4 kDa and only one fraction for DeALPS, denoted DeALPS(HMW), with a molecular weight of 130 kDa.

The 0-acetyl contents of S. paratyphi k polysaccharides were as follows (in percent moles of 0 acetyl per mole repeat unit): LPS, 0.82; O-SP, 0.83; 0-Sp-AH, 0.59; 0-SPC-AH, 0.80; DeALPS, 0.09. Analysis of LPS on SDS- PAGE revealed a characteristic ladder pattern. At 10 μg, O-SP showed a faint smear at the top of the gel and no ladder formation was observed. In contrast, DeALPS showed a faint ladder at 10 μg, indicating that some LPS was still present in the preparation. After treatment of O-SP with CNBr, the amounts of 0 acetyls decreased from 82% to 59% mol/mol. In contrast, activation with CDAP followed by derivatization with ADH did not reduce the O-acetyl content. The adipic acid hydrazide (AH) content of O-SP activated with CNBr was 3.0% and that with CDAP was 2.0% (Table 1). Each repeating poiysaccharide unit within an O-SP can react with ADH. Thus, there may be many AH molecules bound to an O-SP. These percentages indicate the number of O-SP units which have AH moieties bound thereto. For DeALPS activated with CNBr, the AH content was 1.8%. Table 1 shows that the polysaccharide-to-protein ratios of the conjugates ranged between 0.95 and 2.05. The yield of saccharide in the conjugates was between 22 and 75%, with the lower molecular weight 0-SP(LMW)-TT having the lowest yield.

Table 1

% AH/ Saccharide/ Yield

Conjugate saccharide protein (%)^a

O-SP-TT 3.0 2.05 34.1

0-SP(HMW)-TT 3.0 1.34 74.3 0-SP(LMW)-TT 3.0 0.99 22.2 o-SPC-τ 2.9 1.17 31.5

0-SPC-TT₂ NA 0.95 33.9

DeALPS(HMW)-TT 1.8 1.35 71.6

DeALPS-TT 1.8 1.84 62.8

^a Based on the weight of the saccharide. " NA, Not applicable.

Example 7

Conjugation to carrier proteins

ADH-derivatized poiysaccharide (10 mg) was dissolved in PFS (2 ml). An equal weight of tetanus toxoid (lot GYA, Pasteur Merieux Serum et Vaccins, Lyon, France) passed through a Sephacryl S-300 column (Pharmacia) (Chu et al., ibid.) was added and the pH was maintained at 5.1 -5.5 with 0.1 M HCI. The reaction mixture^" as placed on ice, EDAC (Sigma) was added to a final concentration of 0.05 M, and the pH was maintained at 5.1-5.5 for 4 h in 0.1 M HCI. The reaction mixtures were dialyzed against 0.2 M NaCI for 2 days with three changes of outer fluid and passed through a column (1.5 x 90 cm) of Sepharose CL-6B in 0.2 M NaCI. Void volume fractions were stored at 3 • 8°C. The conjugates prepared with ADH as a spacer were designated 0-SP(HMW)-TT, 0- SP(LMW)-TT, DeALPS(HMW)-TT, deALPS-TT and 0-SPC-TT

In one preparation, 0-SP was first activated by CDAP and 0.2 M triethylamine (TEA). An equal weight of TT was added (no ADH spacer) and the pH was maintained at 8.0-8.5 with 0,1 N NaOH for 2 hours. The reaction mixture was passed through a column (1.5 x 90 cm) of CL-6B Sepharose in 0.2 M NaCI, and the void volume fractions were designated 0-SPC-TT₂. Each conjugate is described in Table 2. Table 2

Conjugate Description 0-SP Acetic acid-detoxified LPS passed through G-75

DeALPS Hydrazine-detoxified LPS passed through G-75

0-SP(HMW)-TT and First peak of O-SP or DeALPS from CL-6B activated DeALPS (HMW)-TT with CNBr, derivatized with ADH and bound to TT.

0-SP(LMW)-TT Second pea k of O-S P from C L-6B activated wit h C N Br, derivatized with ADH and bound to TT 0-SP-TT and DeALPS-TT O-SP or DEALPS, activated with CNBr, derivatized with ADH and bound to TT

0-SPC-TT₁ O-SP activated with CDAP, derivatized with ADH and bound to TT

0-SPC-TT₂ O-SP activated with CDAP and bound to TT

Example 8 Immunization and Seroloπv Hyperimmune LPS antiserum was prepared by injecting adult female general-purpose mice from the NIH colony with heat-killed S. paratyphi A NTP-6. For evaluation of immunogenicity, 5- to 6-week old female general purpose mice from the NIH colony were immunized subcutaneously once, twice or three times at 14-day intervals with 2.5 μg of poiysaccharide alone or as a conjugate. Mice from each experimental group of 10 were exsanguinated^" 7 days after each injection (Konadu et al., ibid.). Double immunodiffusion was performed in 0.8% agarose in phosphate buffered saline (PBS). LPS and TT antibody levels were determined by enzyme-linked immunosorbent assay (ELISA) using Immulon 4 plates (Dynatech, Chantilly, VA) (Konadu et al., ibid.) Each well was coated with 100 μl of 10 μg/ml LPS or 20 μg/ml of protein in PBS. IgG and IgM anti-LPS levels were expressed in ELISA units with the hyperimmune serum, which was assigned a value of 100 Units, as a reference.

Double immunodiffusion of LPS, O-SP, and DeALPS with the hyperimmune serum showed partial identity reactions, with spurs extended from the LPS over O-SP, which had a spur over DeALPS. 0-SP-AH formed a line of identity with O-SP when reacting with the hyperimmune serum. Similarly, DeALPS yielded a line of identity with its AH derivative, indicating that the derivatization procedure did not alter the immunogenicity of the polysaccharides. All conjugates emerged as a single peak in the void volume of the CL-6B Sepharose column, and these fractions and the pool gave identical lines of precipitation with the S. paratyphi A hyperimmune and rabbit anti-TT sera. LPS antibodies in mice

Saline, DeALPS, O-SP, DeALPS(HMW)-TT and DeALPS-TT did not elicit detectable LPS antibodies after any injection. None of the conjugates listed in Table 3 elicited anti-LPS after one injection. After the second and third injections, all conjugates except 0-SP(LMW)-TT elicited low levels of IgM antibodies. Both 0-SP(HMW)-TT and O-SP- TT, prepared with CNBr-activated O-SP, elicited IgG antibodies after two injections. Each conjugate elicited a booster response after the third injection; however, there was no significant difference between the IgG levels of anti-LPS (3.01 vs. 2.05; not significant [NS]). Conjugates prepared with CDAP-activated O-SP (0-SPC-TT₁ and 0-SPC-TT₂) showed similar antibody responses after two or three injections. The antibody levels elicited by 0-SPC-TT₁ (with spacer) and 0-SPC-TT₂ (without spacer) were similar (2.37 vs. 1.72 [NS]). Although the level was slightly higher for 0-SP(HMW)-TT, there was no statistically significant difference in LPS antibody levels after the third injection of these four conjugates (3.01 vs. 2.05, 2.37 and 1.72[NS]).

Table 3

GM antibody level (ELISA U)

One Two Three immunogen injection injections i njections

IgG IgM IgG IgM IgG IgM

0-SP(HMW)-TT 0.03 0.03 0.18 0.14 3.01 0.41

0-SP(LMW)-TT 0.05 0.05 0.07 0.04 0.19 0.03

0-SP-TT 0.04 0.03 0.35 0.26 2.05 0.28

0-spc-τ 0.02 0.02 0.30 0.23 2.37 0.33

0-SPC-TT₂ 0.02 0.04 0.26 0.17 1.72 0.31

Bactericidal antibodies

Complement-mediated bactericidal activity was assayed against the NTP-6 strain (Gupta et al., /Zw -- Konadu- et al., ibid.). Briefly, five-fold anti-LPS serum dilutions in 1 % peptone were mixed with equal volumes of about 10⁴ cells/ml in tryptic soy broth (TSB) supplemented with 5% rabbit serum as the source of complement. Mixtures were incubated at 37°C for 1 hour, 50 μl was spread onto TSB-agarose and the plate was incubated overnight at 37°C. Titers were expressed as reciprocals of the highest dilution of serum that yielded 50% bactericidal activity. Controls included complement alone, sera from mice injected with saline and hyperimmune sera without complement. Saline, O-SP and DeALPS did not elicit bactericidal antibodies to S. paratyphi k. As shown in Table 4, sera with high titers, as measured by ELISA, from individual mice injected with O-SP conjugates exhibited significant bactericidal activity against S. paratyphi A. These sera were specific to S. paratyphi A, as no effect was observed against S. typhimurium strain TML. Sera from mice injected with DeALPS(HMW)-TT and DeALPS-TT, which had no detectable anti-LPS by ELISA, had no bactericidal activity (Table 4).

Table 4

Anti-LPS titer Bactericidal

Immunogen (ELISA U) titer

IgG IgM

S. paratyphi A^a 100.0 100.0 32,000 0-SP-TT 7.73 0.54 160

0-SP-TT 5.89 0.10 160

0-SP(HMW)-TT 7.38 0.76 160

0-SP(HMW)-TT 6.14 0.90 160 0-SPC-T 8.49 0.26 > 640

0-SPC-T 9.93 0.42 1,280

0 SPC TT₂ 7.51 0.70 320

0-SPC-TT₂ 6.97 0.39 640

DeALPS(HMW)-TT 0.20 0.04 0 DeALPS-TT 0.01 0.03 0 ^a Pooled sera from mice injected i.v. with heat-killed S. paratyphi A. The correlation coefficients between IgG or IgM with bactericidal titer are 0.7 (P = 0.015) and 0.02, respectively. Sera were from individual mice taken after the third injection of the conjugate.

Protein antibodies

All conjugates, including those with DeALPS, elicited TT antibodies with booster responses. Mice injected with 0-SPC-TT₂, which does not contain a spacer, elicited the highest anti-TT levels after one or two injections (P < 0.005). Table 5

GM antibody levels

Immunogen

One injection Two injections Three injections

Saline ND ND 0.06 0-SP(HMW)-TT 0.18 2.25 11.03

0-SP(LMW)-TT 0.38 10.01 13.94 0-SP-TT 0.36 6.83 8.31

0-spc-τ 0.24 3.55 14.91

0-SPC-TT₂ 0.77 26.04 62.02 DeALPS(HMW)-TT 0.13 3.72 4.91

DeALPS-TT 0.12 1.96 4.90

^a TT antibody levels expressed in ELISA units with a high titer responder as the reference and assigned a value of 100 U. ND, not done.

Applicants have discovered that 0 acetyl groups are essential for eliciting anti-LPS with bactericidal activity against S. paratyphi A. Treatment of S. paratyphi A LPS with acetic acid resulted in retention of 80% of the 0- acetyl groups. In contrast, hydrazinolysis, a clinically acceptable method for detoxification of LPS, removed the 0- acetyl groups. Conjugates prepared with hydrazine-detoxified LPS (DeALPS) did not elicit anti-LPS with bactericidal activity.

CDAP is preferable to CNBr for synthesis of S. paratyphi A conjugates. First, activation by CDAP occurred at neutral pH and did not reduce the O-acetyl content of the O-specific poiysaccharide. Second, CDAP-activated poiysaccharide could be bound directly to TT without adding ADH as a spacer. 0-SPC-TT₂, which had TT directly bound to O-SP without a spacer, elicited the highest level of TT antibodies (P < 0.005).

Example 9 Administration of conjugate vaccine to humans

Two S paratyphi A conjugate vaccines, synthesized according to methods described in the examples, were tested in adult and teenage (13-17 years) volunteers in Vietnam. Conjugate #1 was 0-SP(HMW)-TT in which O-SP was derivatized with an ADH linker prior to conjugation to TT (see Examples 5 and 7). Conjugate #2 was CDAP- activated O-SP directly conjugated to TT (no linker) (see Example 7). Each volunteer received one injection containing 25 μg of conjugate. No fever or other side effect was reported. As shown in Tables 6 and 7, both conjugates elicited statistically significantly higher levels of anti-LPS antibodies in adults as determined by ELISA. In adults, 9.82 versus 1.24 (conjugate #1) and 27.3 versus 1.68 (Conjugate #2) have P values of 0.0001. 85% of volunteers exhibited anti-LPS IgG levels at least 4-fold higher compared to preimmune serum (seroconversion).

Table 6

Coniuqate #1 Preimmune Immune (6 wk)

Number of subjects (N) 10

Geometric Mean IgG level 1.24 9.82

Table 7

Coniuqate #2 Preimmune Immune (6 wk)

Number of subjects (N) 9 8

Geometric Mean IgG level 1.68 27.3

The results for teenage volunteers are shown in Tables 8 and 9. Both conjugates elicited statistically significantly higher levels of anti-LPS antibodies in teenage children as determined by ELISA (1.70 vs. 12.1 for conjugate #1 and 1.68 vs. 17.4 for conjugate #2; P = 0.0001 for both conjugates). Again, 85% of volunteers exhibited anti-LPS IgG levels at least 4-fold higher compared to preimmune serum. Table 8

Conjugate #1 Preimmune Immune (6 wk) Number of subjects (N) 53 43 Geometric Mean IgG level 1.70 12.1

Table 9

Conjugate #2 Preimmune Immune (6 wk)

Number of subjects (N) 51 40 Geometric Mean IgG level 1.68 17.4

It should be noted that the present invention is not limited to only those embodiments described in the Detailed Description. Any embodiment which retains the spirit of the present invention should be considered to be within its scope. However, the invention is only limited by the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

I . Isolated Salmonella paratyphi A (SPA) lipopolysaccharide (LPS) detoxified by removal of lipid A therefrom, said detoxified LPS retaining from at least about 80 to 100% of its O-acetyl groups after said detoxification and having a molecular weight of greater than about 50 kDa.

2. The detoxified LPS of Claim 1 containing about 100% of said 0-acetyl groups.

3. The detoxified LPS of Claim 1 having a molecular weight of greater than about 100 kDa.

4. A conjugate vaccine composition for Salmonella paratyphi A (SPA), comprising SPA .lipopolysaccharide from which lipid A has been removed and in which between about 80 and 100% of its O-acetyl groups are retained (O-SP), said O-SP covalently linked to a carrier.

5. The composition of Claim 4, in a pharmaceutically acceptable carrier suitable for parenteral administration.

6. The composition of Claim 4, wherein said carrier is a protein.

7. The composition of Claim 6, wherein said carrier protein is selected from the group consisting of tetanus toxin/toxoid, diphtheria toxin/toxoid, detoxified P. aeruginosa toxin A, cholera toxin/toxoid, pertussis toxin/toxoid, Clostridium perfringens exotoxins/toxoid, hepatitis B surface antigen, hepatitis B core antigen, rotavirus VP7 protein and respiratory syncytial virus F and G protein.

8. The composition of Claim 7, wherein said carrier protein is tetanus toxoid.

9. The composition of Claim 6, further comprising a linker between said O-SP and said carrier protein.

10. The composition of Claim 9, wherein said linker is selected from the group consisting of adipic acid dihydrazide, N-Succinimidyl-3-(2-Pyridyldithio)propionate, e-aminohexanoic acid, chlorohexanol dimethyl acetal, D- glucuronolactone and p-nitrophenylethylamine.

I I . The composition of Claim 10, wherein said linker is adipic acid dihydrazide.

12. A method of preventing S. paratyphi A infection in an individual, comprising the step of administering to said individual an effective immunoprotective amount of the conjugate vaccine composition of Claim 5.

13. The method of Claim 12, wherein said administering step is intramuscular or subcutaneous.

14. The method of Claim 13, wherein said administering step is intramuscular.

15. The method of Claim 12, wherein said effective amount is between about 10 μg and about 50 μg-

16. A method of detoxifying lipopolysaccharide from SPA, comprising removing lipid A therefrom wherein between about 80 to 100% of O-acetyl groups are retained therein.

17. The method of Claim 16, wherein said lipid A is removed with an acid.

18. The method of Claim 17, wherein said acid is acetic acid.

19. A method of making a conjugate vaccine against SPA, comprising the steps of: removing lipid A from SPA LPS and retaining from between about 80 to 100% of O-acetyl groups therein to produce O-SP; activating said O-SP with a chemical activator; and covalently binding said activated O-SP to a carrier.

20. The method of Claim 19, wherein said removing step comprises treatment with an acid.

21. The method of Claim 21, wherein said acid is selected from the group consisting of acetic, pyruvic, propionic, methanesulfonic and hydrochloric.

22. The method of Claim 19, wherein said carrier is a protein.

23. The method of Claim 22, wherein said carrier protein is selected from the group consisting of tetanus toxin/toxoid, diphtheria toxin/toxoid, detoxified P. aeruginosa toxin A, cholera toxin/toxoid, pertussis toxin/toxoid, Clostridium perfringens exotoxins/toxoid, hepatitis B surface antigen, hepatitis B core antigen, rotavirus VP7 protein and respiratory syncytial virus F and G protein.

24. The method of Claim 23, wherein said carrier protein is tetanus toxoid.

25. The method of Claim 19, wherein said activator comprises CDAP.

26. The method of Claim 19, further comprising binding said chemically activated O-SP to a linker prior to said covalent binding step.

27. The method of Claim 26, wherein said linker is selected from the group consisting of adipic acid dihydrazide, N-Succinimidyl-3-(2-Pyridyldithio)propionate, e-aminohexanoic acid, chlorohexanol dimethyl acetal, D- glucuronolactone and p-nitrophenylethyiamine.

28. The method of Claim 27, wherein said linker is adipic acid dihydrazide.

29. Use of a conjugate vaccine composition comprising SPA lipopolysaccharide from which lipid A has been removed and in which between 80 and 100% of O-acetyl groups are retained (O-SP), said O-SP covalently linked to an carrier, as a vaccine for S. paratyphi A.