AU2016228231A1 - Combination Neisserial compositions - Google Patents

Abstract

Compositions comprising a first biological molecule from a Neisseria bacterium and a second biological molecule from a Neiseria bacterium. The term 'biological molecule' includes proteins and nucleic acids. Preferred Neisseria species are N.meningitidis and N. gonorrhoeae. a)CD rco C) CC) C) 9- U) ~ c~ cr 0) C)~ 0c r--~ LO r- 0 0 LO0 C) (B C C:) 6 6D C) C C) C) :) C= 0 O qT m Cji c- c ~ 1o C:) co E 9- f c C=J C=- C=) C) 0 00 0l 0~ to a L~) 9- ~ ~ - 9- 9-CD LO m

Description

COMBINATION NEISSERIAL COMPOSITIONS

This is a divisional application of AU 2014265011, which is a divisional of AU2013200678, which is a divisional of AU2009200865, which is a divisional of AU2006201153, which is a divisional of AU20050977, which is the Australian National Phase Entry of PCT/1B00/00828, which claims priority from GB9911692.3 filed on 19 May 1999, GB9919705.5 filed on 19 August 1999 and GB0005730.7 filed on 9 March 2000, the entire contents of each of which are incorporated herein by reference in their entirety.

FIELD OF THE INVENTION

This invention relates to compositions comprising combinations of biological molecules from Neisseria bacteria, particularly N. meningitidis and N.gonorrhoeue.

BACKGROUND ART

Neisseria meningitidis and Neisseria gonorrhoeae are non-motile, Gram negative diplococci that are pathogenic in humans.

Based on the organism's capsular polysaccharide, 12 serogroups of N. meningitidis have been identified. Group A is the pathogen most often implicated in epidemic disease in sub-Saharan Africa. Serogroups B and C are responsible for the vast majority of cases in the United States and in most developed countries. Serogroups W135 and Y are responsible for the rest of the cases in the United States and developed countries.

The meningococcal vaccine currently in use is a tetravalent polysaccharide vaccine composed of serogroups A, C, Y and W135. Meningococcus B remains a problem, however. The polysaccharide approach cannot be used because the menB capsular polysaccharide is a polymer of a(2-8)-linked N-acetyl neuraminic acid that is also present in mammalian tissue. One approach to a menB vaccine uses mixtures of outer membrane proteins (OMPs) To overcome the antigenic variability, multivalent vaccines containing up to nine different poring have been constmcted [eg. Poolman JT (1992) Development of a meningococcal vaccine. Infect. Agents Dis. 4:13-28]. Additional proteins to be used in outer membrane vaccines have been the opa and ope proteins, but none of these approaches have been able to overcome the antigenic variability [eg. Ala Aldeen & Borriello ( 1996). The meningococcal transferrin-binding proteins 1 and 2 are both surface exposed and generate bactericidal antibodies capable of killing homologous and heterologous strains. Vaccine 14(1 ):49-53].

Given the propensity for meningococcal disease during non-epidemic periods to be caused by multiple strains or strain variants [Russel et al. (1998) Abstracts of 11th International pathogenic Neisseria conference, page 281] together with frequent temporal shifts in the predominant strains in a community, it seems that a universal meningococcal B vaccine will require more than one antigenic species.

DESCRIPTION OF THE INVENTION

Neisseria! protein and nucleotide sequences arc disclosed in the following documents: • WO 99/24578 • WO 99/36544 • WO 99/57280 • WO 97/28273 • WO 96/29412 • WO 95/03413 • Tettelin el ui (2000) Science 287:1809-1815

For ease of reference, the sequences disclosed in these documents are referred to in the present application according to the following table:

The present invention provides compositions comprising a first biological molecule from a Neisseria bacterium and a second biological molecule from a Neisseria bacterium. The term "biological molecule" includes proteins and nucleic acids.

The compositions may also comprise further biological molecules, preferably also from Neisseria, that is to say the compositions may comprise two or more biological molecules (eg 3. 4. o. 6. 7. 8 etc. I. at least two of which are from a Neisseria bacterium (eg. 3. 4. 5. 6. 7. 8 etc.). Such compositions include those comprising (i) two or more different Neisserial proteins. (ii) two or more different Neisserial nucleic acids, or (iii) mixtures of one or more Neisserial protein and one or more Neisserial nucleic acid.

In one preferred embodiment, the first and second biological molecules are from different Λeisseria species (eg; one is from X meningitidis and one is from X.gonorrhoeae). but they may be from the same species. The biological molecules in the compositions may be from different serogroups or strains of the same species. I he first biological molecule is preferably selected from the group consisting of SEQ IDs 1-8376. More preferably, it is selected from the group consisting of SEQ IDs 1-4002 and/or SEQ IDs 4057-8376. It is preferably a purified or isolated biological molecule.

The second biological molecule is preferably selected from the group consisting of SEQ IDs 1-8376. More preferably, it is selected from the group consisting of SEQ IDs 1-4002 and/or SEQ IDs 4057-8376. It is preferably a purified or isolated biological molecule.

One or both of the first and second biological molecules may be a Neisserial biological molecule which is not specifically disclosed herein, and which may not have been identified, discovered, made available to the public or purified before this patent application was filed.

In particular, the invention provides a composition comprising one or more of the following pairs of first and second biological molecules (listed by SEQ ID):

Thus the invention includes each of the 35074500 possible pairs of SEQ IDs 1-8376 (1&2. 1&3. 1&4. 1&5 ... 1&8375. 1&8376. 2&3. 2&4. 2&5 ... 2&8375. 2&8376, 3&4 ... 1000&1001. 1000&1002 ... 1000&8376 ... S374&8375. S374&8376. 837S&8376) although, for reasons oi space, these are not listed in full here.

Details as to how the molecules which make up the SEQ IDs 1-4056 can be produced and used can be found from the relevant international applications and these details need not be repeated here. Similar principles apply to SEQ IDs 4057-8376. SEQ IDs 1-8376 in the compositions of the invention may be supplemented or substituted with molecules comprising sequences homologous (ic. having sequence identity) to SEQ IDs 1-8376. Depending on the particular sequence, the degree of identity is preferably greater than 50% (eg. 65%. 80%. 90%. or more), and include mutants and allelic variants. Sequence identity between the proteins is preferably determined by the Smith-Waterman homology search algorithm as implemented in the MPSRCH program (Oxford Molecular), using an affine gap search with parameters gap open penaity=l2 and gap extension penalty-1. SEQ IDs 1-8376 in the compositions of the invention may be supplemented or substituted with molecules comprising fragments of SEQ IDs 1-8376. Such fragments should comprise at least n consecutive monomers from the molecules and. depending on the particular sequence, n is either (i) 7 or more for protein molecules {eg. 8, 10. 12. 14. 16. 18. 20 or more), preferably such that the fragment comprises an epitope from the sequence, or (ii) 10 or more for nucleic acid molecules (eg- 12. 14. 15. 18. 20. 25. 30. 35, 40 or more).

Where the composition includes a protein that exists in different nascent and mature forms, the mature form of the protein is preferably used. For example, the mature form of the NspA protein (SEQ IDs 4008-4033: W096/29412: Figure 29) lacking the signal peptide mav be used.

In the case of protein molecules. SEQ IDs 1-8376 in the compositions of the invention may be supplemented or substituted with an antibody that binds to the protein. This antibodv may be monoclonal or polyclonal.

In the case of nucleic acid molecules. SEQ IDs 1-8376 in the compositions of the invention may be supplemented or substituted with nucleic acid which can hybridise to the Neisserial nucleic acid, preferably under "high stringency" conditions (eg. 65°C in a 0. lxSSC, 0.5% SDS solution).

It will be appreciated that any nucleic acid in the compositions can take various forms (eg. single stranded, double stranded, vectors, probes ere.). In addition, the term “nucleic acid" includes DNA and RNA. and also their analogues, such as those containing modified backbones, and also peptide nucleic acids (PNA) etc.

In certain embodiments, the composition comprises molecules from different Neisseria species, such as one or more .Y. meningitidis molecule and one or more Λ’ gonorrhoeas molecule. In some embodiments, the composition may comprise molecules from different serogroups and/or strains of the same species, such as strains A and B of N.meningitidis. Further embodiments comprise mixtures of one or more ;Y. meningitidis molecules from different strains and also one or more N.gonorrhoeae molecules.

Many proteins are relatively conserved between different species, serogroups and strains of .V.meningitidis and N.gonorrhoeae (eg. SEQ IDs 52. 54. 58). PCT/ΊΒ00/00642 includes a more detailed experimental analysis of conserved regions in these proteins. To ensure maximum cross-strain recognition and reactivity, regions of proteins that are conserved between different Neisserial species, serogroups and strains can be used in the compositions of the present invention. The invention therefore provides proteins which comprise stretches of amino acid sequence that are shared across the majority of Neisseria, particularly N.meningitidis and N.gonorrhoeae. Preferably, therefore, the composition comprises a protein comprising a fragment of a Neisserial protein (preferably a protein from SEQ IDs 1-8376. or more preferably SEQ IDs 1-4002). wherein said fragment consists of n consecutive conserved amino acids. Depending on the particular protein, n is 7 or more {eg. 8. 10. 12. 14. 16. 18. 20 or more). The fragment preferably comprises an antigenic or immunogenic region of the Neisserial protein. A "conserved" amino acid is one that is present in a particular Neisserial protein in at least x% of Neisseria (or. preferably- in at least x% of combined N.meningitidis and N.gonorrhoeae strains). The value ofx may be 50% or more eg. 66%. 75%. 80%, 90%, 95% or even 100% (ie. the amino acid is found in the protein in question in all Neisseria). In order to determine whether an amino acid is "conserved" in a particular Neisserial protein, it is necessary to compare that amino acid residue in the sequences of the protein in question from a plurality of different Neisseria (a "reference population"). Suitable definitions of "reference populations” can be found in PCT/IB00/00642. Amino acid sequences of different Neissieriae can easily be compared using computers. This will typically involve the alignment of a number of sequences using an algorithm such as CLUSTAL [Thompson el al. (1994) Nucleic Acids Res 22:46734680; Trends Biochem Sci (1998) 23:403-405] or. preferably. PILEUP [part of the GCG Wisconsin package, preferably version 9.0], Conserved amino acids are readily apparent in a multiple sequence alignment - at the amino acid position in question a majority of the aligned sequences will contain a particular amino acid. Conserved amino acids can be made more visually apparent by using a program such as BOXSHADE [available, for instance, at the NIH on-line], PRETTYBOX [GCG Wisconsin, version 10] or JALVIEW [available on-line at EBI],

Specific compositions according to the invention therefore include those comprising: two or more biological molecules selected from SEQ IDs 1-4002; one or more biological molecules selected from SEQ IDs 1-4002 combined with one or more biological molecules selected from SEQ IDs 4003-8376; one or more biological molecules selected from SEQ IDs 1-4002 combined with the NspA protein (as disclosed in W096/29412; see also Figure 29 herein), preferably in mature form; one or more biological molecules selected from SEQ IDs 1-S376 (preferably SEQ IDs 1-4002) combined with transferrin binding protein A (TbpA) and/or B (TbpB). such as the TbpA and TbpB disclosed in WOOO/25811 (or immunogenic fragments thereof). one or more fragments of proteins selected from SEQ IDs 1 -4002. with the fragment preferably comprising a stretch of conserved amino acids; a combination of different proteins, wherein the combination as a w hole includes one or more proteins that is recognised by each strain in a reference population, although each individual protein in the combination may not itself be recognised by each strain in the reference population te. each member of a reference population recognises at least one protein in the combination.

The invention also provides the compositions of the invention for use as medicaments {eg. as immunogenic compositions or vaccines) or as diagnostic reagents. It also provides the use of the compositions in the manufacture of: (i) a medicament for treating or preventing infection due to Neisserial bacteria; (ii) a diagnostic reagent for detecting the presence of Neisserial bacteria or of antibodies raised against Neisserial bacteria; and/or (iii) a reagent which can raise antibodies against Neisserial bacteria.

The invention also provides a method of treating a patient, comprising administering to the patient a therapeutically effective amount of a composition according to the invention.

The invention further provides a process for producing a composition according to the invention, comprising the step of bringing one or more of SEQ IDs 1-8376 into combination with one or more other of SEQ IDs 1 -8376.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows SDS-PAGE results for the expression of ORFs 6. 7. 13. 65-1. 72. 73-1. 105-1, 137-1, 143-1 and 147-1. The left-hand lanes show molecular weight markers (set Ml).

Figure 2 shows (A) SDS-PAGE results for ORT9 (B) the position of the N.meningitidis immunoreactive band in a Western blot against a Λ .meningitidis outer membrane vesicle preparation (C) FACS analysis.

Figure3 shows SDS-PAGE results for the expression of ORFs 2-1. 5-1. 22-1. 132-1 and 4. The left-hand lanes show' molecular weight markers (set Ml).

Figures 4 to 28 show a hvdrophilicity plot (upper), antigenic index (middle), and AMPHI regions (lower) for ORFs 2. 5. 6. 7. 9. 13a. 15. 22. 23. 27. 28. 32. 65. 72. 73. 76. 79. 89, 105, 106-1. 132. 137. 138. 143 and 147.

Figure 29 shows sequence variability of NspA from various strains of meningococcus B. These sequences may be used as alternatives to the NspA of WO 96/29412 (SEQ IDs 4008-4033).

Figure 30 shows the binding of polyclonal anti-rNspA by indirect fluorescence flow cytometry to encapsulated and non-encapsulated menB strains. Figure 31 shows similar data for encapsulated strains 8047. CU385 & M986 (31 A) and non-encapsulated strains BZ232. MC58, NG3/88 & NGB165 (3IB),

Figure 32 shows a model of NspA secondary structure.

Figure 33 shows FACS analysis of non-encapsulated strain M7 (37A) and ethanol-treated (to disrupt the capsule) strains 2996. N44776. MC58, 1000. BZ232. BZ133. NG6/88. BZ198, NG3/88, 297-0, BZ147 and BZ169 (37B) using a tetravalent mixture.

Figure 34 shows FACS analysis of M7 strain using a pentavaient mixture at 1:400 (34A), 1:200 (34B) & 1:100 (34C) dilutions.

EXAMPLES

Example 1 - Expression and purification experiments ORFs 6, 7. 13. 65-1. 72. 73-1. 105-1. 137-1. 143-1 and 147-1 disclosed in W099/24578 were expressed in E.coli and purified, as set out in the following table:

NOTE: ORF"'-] was expressed as a fragment tamino acids 41-161)

The protocols used to express tnese ten ORFs were essentially the same as those described in W099/2457&. using pGEX and pET \ ectors. Examples of the PCR primers used to ampiifv the ORFs are in the following table:

SDS-PAGE results for the ten expressed ORFs are shown in Figure i.

The ORF7-His fusion was used to immunise mice. The sera were used in an ELISA assay, essentially as described in WOP9 2457S. and gave positive resuits.

The following proteins were also expressed and purified (results not snowm:

Each of these ORFs can be combined with one or more of SEQ IDs 1-8376.

Example 2 - ORF9 expression and purification ORF9 disclosed in W099/24578 was cloned in the pET vector and expressed in E.coli. The purified ORF-His fusion protein was analysed by SDS-PAGE. as shown in Figure 2Λ. Mice were immunised with the purified ORF9-His and sera were used for Western blot analysis (Figure 2B1 FACS analysis (Figure 20. and ELISA assay. The protocols used were essentially the same as those set out in W099/24578.

The results confirm that 0RF9 is a surface-exposed protein. 0RF9 is suitable for combining with one or more other of SEQ IDs 1 -8376.

Example 3 — Further expression experiments

Further expression and purification experiments were carried out in E.coli for ORFs 2-1. 5-1. 22-1 and 132-1 disclosed in W099/24578. as set out in the following table:

The protocols used to express these four ORFs were essentially the same as those described in W099/24578. using pGEX and pET vectors. Examples of the PCR primers used to amplify' the ORFs are in the following table:

SDS-PAGE results for the four expressed ORFs are shown in Figure 3.

Each of these ORFs can be combined with one or more of SEQ IDs 1-8376.

Example 4 - ORF4 lipoprotein expression and purification ORF4 is disclosed in W099724578 as containing a lipopeptide signal sequence (LPSS). The full length ORF was amplified using the following PCR primers:

orf4-L for CGCGGATCCCATATGAAAACCTTCTTCAAAACC

orf4~L rev CCCGCTCGAGTTATTTGGCTGCGCCTTC

The amplified DNA fragment was cloned into the vector pET21b+ for expression as a C-terminus His-tagged fusion. A log phase culture of E.coii containing pET21b-i-orf4-LPSS was induced with 1.0 mM 1PTG for 3 h at 30°C. collected by centrifugation for 10 min at 8Q00g. and resuspended in PBS. The suspension was sonicated on ice and Triton X-l 14 added to a final concentration of 0.6°o (v v). The material was incubated on ice for 20 minutes then warmed to 37°C until phase separation occurred (indicated by a high degree of cloudiness). After centrifugation tor 10 min at lO.OOOg at 20°C the upper aqueous phase was discarded, and the lower detergent phase collected without disrupting the bacterial pellet. To the detergent phase was added 13 volumes of 20 mM Histidine. 2 mM EDTA. 30 mM NaCl (pH 5.8). This was centrifuged for 10 mm at 4°C and the supernatant combined batchwise at 4°C for 30 min. with 0 Sepharose Fast Flow resin (Pharmacia). The mixture was centrifuged, the supernatant retained and the resin washed with 20mM Histidine. 2mM EDTA. 30 m.M NaCl (pH 5.8). Triton X-100 0.3° o (v v) and eluted with 1M NaCl in the same buffer. The majoritv of Orf4 lipoprotein was found in the supernatant obtained after binding. Final purification was accomplished by chromatography on Hi-TrapTM 0 (Pharmacia). The bindinu-suDematant was adjusted to pH 7.0 by the addition of 0.1M HC1 ana applied to a Hi-TrapTM Q column equilibrated with 50mM Tris-HCl (pH 7.0). 2mM EDTA. 0.3% Triton X-100. lOmM NaCl. The column was washed with 5.0 ml of the equilibration buffer and a NaCl gradient from lOmM to 1M was applied. Two electrophoretically distinct forms of the protein eluted, One in the wash and the other in the NaCl gradient between 15QmM and 300mM NaCl. The protein obtained in the wash was used for immunization of mice. This form of the protein probably represents the fully processed, lipidated molecule.

The 31kDa purified lipoprotein can be seen in Figure 3. ORF4 is suitable for combining with one or more other of SEQ IDs 1 -8376.

Example 5 - Computer prediction

Computer analysis ofORFs 2. 5. 6a. 7. 9. 13a. 15. 22. 23. 27. 28. 32. 65. 72. 73, 76. 79. 89. 105. 106-1. 132. 137. 138. 143 and 147 (as disclosed in W099/24578) was performed. Figures 4 to 28 show, for each or these ORFs. a hydrophilicity plot (upper), an antigenic index plot (middle), and an AiMPHI analysis (lower). The AMPHI program has been used to predict T-cell epitopes [Gao el al. (1989),/. Immunol. ]_43:3007: Roberts et al. (1996) AIDS Res Hum Retrovir L2:593; Quakvi et al. (1992) Scand J Immunol suppl.ll:9) and is available in the Protean package of DNAS TAR. Inc. (1228 South Park Street. Madison. Wisconsin 53715 USA).

Each of these ORFs can be combined with one or more other of SEQ IDs 1-8376.

Example 6 - Tetravalent mixture A mixture of proteins 919 (WO99/57280). 225 (WO99/57280). ORF4 (W099/24578. example 26) and ORF40 (W099/36544. example 1) was produced and assessed by ELISA and FACS. The ELISA titres against 13 test strains were as follows:

The FACS results are shown in Figure 33. It is evident that the tetravalent mixture gives excellent results, regardless of the particular menB strain used. In addition, antisera raised against the mixture in strain 2996 is bactericidal at up to 1:2048 dilution.

Example 7 - Pentavalent mixture A mixture of proteins 0RF4-L (the lipidated protein - see example 4 above). ORF37 (W099/24578. example 1). ORF40 (W099/36544. example 1), 502 (WO99/57280. pages 687690) and 8 (WO99/57280. pages 165-167) was produced. The ELISA litres against 13 test strains were as follows:

The FACS results are shown in Figure 34. It is evident that the pentavalent mixture gives excellent results, regardless ol the particular mcnB strain used. In addition, antisera raised against the mixture in strain 2996 is bacteriostatic.

Example 8 - trivalent mixture

Proteins ORF1 (e.g. example 77 of W099/24578: see also W()99/55873). ‘287 (e.g. Figure 21 of WO99/57280: also SEQ IDs 3103-3108 therein) and '919' (e.g, WO99/57280 Figure 23 and SEQ IDs 3069-3074 therein) were combined and adjuvanted with Al(OH)3. The proteins were from the 2996 strain of MenB.

This mixture was also combined with a MenC polysaccharide conjugate antigen [e.g. Costantino el al. (1992) Vaccine 10:691-698]. OMVs were used as controls.

The mixture was used in a bactericidal assay against the homologous strain and also heterologous MenB strains. Titres were as follows:

Example 9 - Proteins 287, 919 and 953

Proteins 287. 919 and 953 are disclosed in WO99/57280. These proteins from N.meningitidis serogroup B strain 2996 were expressed and tested in a bactericidal assay against strain 2996, alone and in combinations. OMVs from 2996 were used as a positive control.

Figure 35 shows FACS data for the individual antigens and for the four combinations.

It is evident that the antigen mixtures are more effective than the antigens in isolation and. in particular, that combinations of 919+953 give surprisingly good results.

The individual antigens from 2996 and combinations were also tested against different serogroup A. B & C strains (/.e. heterologous challenge). Bactericidal litres were as follows:

It is apparent that the antigen mixtures are useful in conferring cross-strain activity.

In a second set of experiments, titres lor the individual antigens were as follows:

The three proteins used in this example were expressed and used in the following forms: (1) Protein 287 was expressed in E.coli as a GST fusion: (2) Protein 919 was expressed in E.coli without its leader peptide, without its mature N-terminal cysteine, and without any fusion partners ("919-untag"); and (3) Protein 953 was expressed using a histidine tag.

Three immunisations were administered with Freund's adjuvants - the first included CFA, and the final two included IFA.

Example 10 - further polyvalent combinations

Further combinations of antigens were tested in CD1 mice:

* : "his" indicates expression and immunisation with a histidine-tagged protein; "ORF4-L" is the lipidated form of ORF4; “GST" indicates expression and immunisation with a GST fusion protein: “919-untag" is as defined in Example 9; “MenC glycoconj" is the MenC glycoconjugate described in Example 8.

Further combinations of antigens were tested in guinea pigs:

Evidently the combinations give excellent immunological results.

Example 11 - NspA combinations

NspA protein is disclosed in W096/29412. and is represented herein as SEQ IDs 4008-4033. The academic literature disclosure of this protein [Martin el al. (1997) J. Exp. Med IS5 11731183] reported the protein to be highly conserved between Neisseria strains (99% crossreactivity' ot anti-NspA antibodies with 250 meningococcal A. B & C strains) and also efficient protection against deadly challenge with live bacteria. There have also been reports that NspA adsorbed on alum elicits serum meningococcal bactericidal antibody responses in rabbits and monkeys [Martin et al. (1998) Abstracts of 11th International pathogenic Neisseria conference, page 198], On the basis of these data. rNspA (recombinant NspA) is being developed as a vaccine for the prevention oi meningococcal disease caused by all serocroups.

Oespite sequence conservation, however, it has surprisingly been discovered that rNspA cell surface epitopes are detected on only 65% of the serogroup B strains tested below, and susceptibility to anti-NspA bactericidal activity is also less than that reported by Martin et al. Niese results contrast vvith Martin et al.. and suggest that a rNspA-based meningococcal B vaccines will need to be supplemented with additional antigens in order to be effective.

The Λ. meningitidis strains tested in this example were isolated from patients residing in different countries over a period of more than 30 years (see table on page 72). These strains were selected to be representative of widely divergent ’clonal' groups, as defined by multilocus isoenzyme typing [Seiler et al. (1996) Mol. Microbiol. 19:841-856] and/or multilocus sequence typing [Maiden et al. (1998) PNAS USA 95:3140-45], Strain M7. which is derived from strain NMB. contains a transposon insertion that blocks capsular polysaccharide biosynthesis [Stephens el al. (1991) Infect, kntnun. 59:4097-4102], but all the other strains are encapsulated.

Based on the nucleotide sequence in Martin et al. (1997), PCR primers were designed and the Asp A gene Irom strain 8047 was amplified. The sequence, including the promoter region, was cloned into pSK.^ plasmid (rNspA). A plasmid pTrc.NspA. 1 encoding a protein in which a portion of the signal sequence has been replaced with a poly-histidine tag was also used. Both plasmids were expressed in E.coli strain BL21(DE3) and the proteins were purified. In E.coli. rNspA is secreted, rather than being associated with the outer membrane. The protein was paniallv purified from the culture medium by precipitation with 55% w/v ammonium sulphate, and had an apparent MW oi 18.6kDa. confirmed by Western Blot.

The two forms of NspA (rNspA and denature His-tage NspA) were injected into 6-week old female CD-I mice to raise antisera. The ability of these to bind to the surface of X.meningitidis strain B was determined using flow cytometric detection of indirect fluorescence assay [Granoff et al. (1998) J.Immunol 160:5028-36]. The results for strains NMB and M7 (an acapsulated mutant of NMB) are shown in Figure 30. As expected, anti-group B polysaccharide mAb SEAM-3 [Granoff et al.} only binds to the encapsulated strain, whereas the positive anti-P1.2 (PorA) control mAb binds to both strains. The antisera raised against rNspA is able to bind both strains. Antisera against the His-tag NspA gave negative results, however. These antisera were also negative for strains 8047. CU385 and M986 (Figure 31 A), but by Western Blot these antisera gave positive results.

These data suggest that antibodies prepared using Flis-tag NspA recognise epitopes that are present in denatured NspA. but not native NspA as found on the cell-surface in vivo. In contrast, antibodies prepared against rNspA seem to recognise conformational NspA epitopes.

The flow cytometric assay was applied to the strains shown in the table on page 72. Figure 31A shows that murine antibodies raised against rNspA bind to the surface of strain 804/ (the strain from which the nspA gene was cloned) and strain CU385. but not to M986. Figure 3IB shows similar negative results for strains BZ232. MC58. NG3/88 and NGP165. In all of these negative cases, however, the anticapsular mAb control was positive.

The table on page 72 summarises the flow cytometry results. Although NspA is reported o be accessible at the surface of all intact N.meningitidis strains tested [Martin et al. (1997) J. Exp. Med 185 1173-1183; Plante et al (1999) Infect. Immun. 67:2855-61], only 11 of the 17 test strains (65%) reacted with the anti-rNspA sera. There was no apparent relationship between cell-surface expression in a given strain and classification (by serotype, subtype, or electrophoretic type), or with year or country of isolation.

In an attempt to explain the differences in reactivity with the anti-rNspA sera, the nspA genes from five of the six negative strains (BX232, NG3/88, NGP165. M136 & M986) and from three of the positive strains (8047, CU385 & NG6/88) were sequenced. The sequence for the sixth negative strain (MC58) was already available from the complete genome sequence.

The nspA sequences for all ten strains were highly conserved, varying at most by 5 nucleotides from the prototype sequence of Martin et al. The most variant protein had only 3 amino acid differences (see Figure 29). With one exception, all of the amino acid variants involved the same respective residues in discrete segments of the protein. These include the signal peptide, which is not present in the mature protein, and two short segments in the 50 C-terminal residues. These differences do not explain the antisera results, as there are examples of identical variant sequences in strains that were positive and those that were negative (compare Ml36 & 8047; NGP165 & NG6/88; MC58 & CU385).

As neither lack of the gene nor polymorphism explained the antiserum results, the amount of NspA protein in the outer membranes of five strains (8047, CU385 & NG6/88 - all positive for anti-rNspA; M986 & Ml36 - both negative) were tested. Bacterial cell pellets were extracted with lauryl sarcosinate, and the insoluble outer membrane fractions were analysed. An 18.6kDa band was seen for all five strains, and this was cross-reactive with anti-His-tag-NspA by Western Blot. Thus strain differences in nspA expression also failed to explain the results.

The ability of anti-rNspA to bind to the bacterial cell surface could be influenced by the amount of polysaccharide capsule present. The quantity of capsular polysaccharide produced by the 17 test strains was therefore assessed by inhibition ELISA.

Extracts of capsular polysaccharide were prepared based on a method described by Com et al. [J.Infect.Dis. (1993) 167:356-64], Individual bacterial clones were grown to an OD^o 0.5-0.7 in 7 mL of Mueller-Hinton broth. Bacteria were collected by centrifugation at 5000g for 15 min, washed in 0.6mL of 10 mM Hepes, pH 8.0, and then resuspended in 0.6 mL of the same buffer containing 10 mM EDTA and incubated at 37°C for 1 hr. The cells were pelleted at 10,000g for 1 minute and the relative amount of meningococcal B polysaccharide antigen released into the supernatant was determined by an inhibition ELISA, performed as described by Azmi et al. [Infect. Immun. (1995) 63:1906-13]. The solid phase antigen in the ELISA was meningococcal B polysaccharide-ADH-biotin absorbed to avidin-coated microtiter plates [Granoff et al.]. The meningococcal B polysaccharide-reactive human paraprotein LIP [Azmi et a 1.] was used as the primary antibody (0.2 pg/ml). In the absence of inhibitor, this concentration of antibody was sufficient to given an OD of ~0.7 to 1.0 after 30 minutes incubation with substrate [Azmi et al.]. The titre of polysaccharide released into the supernatant was measured by determining the dilution of supernatant that resulted in 50% inhibition of antibody binding. Controls in this assay included an EDTA extract prepared from the strain M7, which does not produce any capsular polysaccharide, and purified meningococal B polysaccharide. To ensure that all of the capsular polysaccharide was released by the EDTA treatment, the same inhibition ELISA was performed using the cell pellet resuspended in the same buffer and volume as the capsule extract. The observable inhibitory· activity from the cell pellet was between 0 and 10% of the activity observed in the capsule extracts with the latter, higher percentage coming from cell pellets of strains that produce the largest amounts of capsule.

The results for each strain are shown in the table on page 72. On average, the six negative anti-rNspA strains produced three-fold more capsular polysaccharide than the eleven positive strains (respective reciprocal geometric mean dilutions of 676 vs. 224, p<0.05). This may explain the results obtained with the antiserum -- conceivably, the presence of larger amounts of capsule could interfere with the ability of the anti-rNspA antibody to bind to NspA epitopes which, in strains with lower amounts of capsule, are accessible.

The complement-dependent bactericidal activity of the anti-rNspA antisera were tested using an assay similar to that described by Mandrell el al [J. Infect. Dis. (1995) 172:1279-89]. The complement source was human serum from a healthy adult with no detectable anti-capsular antibody to group B polysaccharide and no intrinsic bactericidal activity against the test strain. Serum bactericidal titres were defined as the serum dilution resulting in a 50% decrease in CFU/ml after 60 minutes incubation of bacteria in the reaction mixture, compared to the control CFU/ml at time zero.

Typically, bacteria incubated with a negative control antibody showed a 150-200% increase in CFU/ml during the 60 minutes of incubation. The positive control antibody [anti-capsular IgG2a mAb SEAM 12, Granoff et al.] showed complement-mediated bactericidal activity against all 17 strains. In contrast, the six strains that were negative for anti-rNspA antisera binding by flow assay were resistant, showing no bactericidal or bacteriostatic effects. Ten of the other eleven positive strains were either killed by complement and the antisera (SWZ107, J351, CU385, NG6/88, BZ198, H44/76, NMB &amp; 8047) or were inhibited (H355 &amp; S3446); strain 1000, however, was not affected.

The ability of the anti-rNspA antisera to confer passive protection against meningococcal B bacteremia was tested in infant rats using a method adapted from Saukkonen [J. Infect. Dis. (1988) 158:209-212]. Briefly, 6-7 day old rats were randomly distributed to nursing mothers. Groups of 5-6 animals were challenged IP with 100μ1 of approximately 5000 CFU of N.meningitidis group B bacteria. One strain negative for NspA surface epitopes (M986) and one positive strain (8047) were tested, each of which having been passaged three times in infant rats. Immediately before administration, the bacterial suspension was mixed with different dilutions of test or control antibody (positive control: anticapsular mAb; negative control: asiti-E.coli). 18 hours after challenge, blood specimens were obtained from the heart. Aliquots were plated onto chocolate agar, and CFU/ml was determined after overnight incubation at 37°C in 5% C02.

The protective activities of the various co-administered antibodies were as follows:

ap>0.5, compared to geometric mean CFU/ml of control rats b £><0.001, compared to geometric mean CFU/ml of control rats

As can be seen, a dose of 2pg per rat of the positive anticapsular control was protective against both strains. A 1:5 or 1:25 dilution of anti-rNspA antiserum protected against bacteremia caused by strain 8047. Neither dilution was effective in preventing M986 bacteremia, however.

Despite the positive conclusions of Martin et ah, therefore, NspA does not seem to be effective in preventing meningococcal B infection. Approximately one third of strains have decreased cell-surface expression of NspA epitopes when grown in vitro, are resistant to anti-NspA induced complement-mediated bacteriolysis, and are resistant to passive antiserum immunisation. These strains produce large amounts of capsular polysaccharide, and would thus be expected to have the greatest virulence. The ability of a vaccine containing only NspA to confer broad protective immunity against meningococcal B thus has to be doubted.

Compositions comprising NspA [SEQ IDs 4008-4033; Figure 29] therefore advantageously comprise further antigens. A preferred aspect of the invention is thus a combination of NspA protein with one or more further Neisserial antigens.

Example 12 - NspA fragments A model of the secondary structure of NspA is shown in Figure 32, containing eight transmembrane β-strands and 4 surface-exposed connecting loops. This fits the pattern of alternating hydrophobic and hydrophilic amino acids in NspA. which is characteristic of many β-barrel porins [Weiss et al. (1990) FEBS Letts 267:268-272]

The grey shaded areas in the model indicate segments that are >40% identical and >70% similar to encoded amino acid sequences of opacity proteins (Opa) from N. meningitidis, N. gonorrhoeae, N. flavius, N. sicca, and Hinfluenzae identified in a BLAST search of the nonredundant GenBank CDS. The alternating sequences are predicted amphiphilic β-strands; vertical segments correspond to transmembrane segments; the top of the figure corresponds to surface exposed segments, labelled as loops 1 to 4 .

According to Martin et al,, the only significant homology between the deduced amino acid sequence of NspA and those of other proteins are weak homologies with the Neisseria opacity protein (Opa) family in two small segments (—20 amino acids) near the C-terminal end of the protein . However, separate comparisons of the N- and C-termini of NspA with GenBank reveals a high degree of homology (>40 % identity and >70% similarity) between NspA and Opa proteins from N.meningitidis, N. gonorrhoeae, N. flavius, N. sicca, and H. influenzae. The Opa proteins are thought to be integral membrane proteins that have eight transmembrane segments and a β-barrel topology in the membrane similar to that of porin [Merker et al. (1997) Mol. Microbiol 23:281-293]. The presence of NspA in detergent-insoluble membrane preparations indicate that NspA is located in the outer membrane, which would be consistent with the Opa-like membrane topology shown in the model. In addition, the segments of NspA that are most homologous to those of the Opa proteins are the putative transmembrane segments indicated in the shaded areas of Figure 32.

The opacity proteins of Neisseria can, under certain circumstances, elicit protective antibody. However, problems with limited antibody accessibility of the opacity proteins in encapsulated bacteria, variability of amino acid sequences in exposed loop segments, and phase variation of protein expression during clinical infection, have limited the ability of Opa to elicit protective antibody consistently [Malomy et al. (1998) J. Infect. Dis. 172:1279-89], In contrast, there appears to be little or no sequence variation in the surface exposed loops of NspA in Figure 32. However, it was recently reported that a panel of anti-TV. meningitidis NspA monoclonal antibodies that reacted with all meningococcal strains tested reacted with only a limited number of JN.gonorrhoeae strains, even though the respective amino acid sequences in the two species are 92% identical. When the respective NspA sequences of the meningococcal and gonococcal strains are compared (Figure 29), all of the respective amino acid differences result in changes in hydrophobicity or charge, and are located in the putative surface exposed connecting loops (Figure 32). This finding suggests that the connecting loops in NspA, which are highly conserved in N.meningitidis, may be important epitopes for antibodies that bind to native NspA. These segments of the molecule, therefore, would appear to be of greatest interest with respect to interacting with protective antibody. Flowever. the putative surface loops of NspA are relatively small (10-14 amino acids) compared to, for example, the highly immunogenic external loops of PorA and Ope (24 to 45 amino acids). The shorter length of the loops may limit the accessibility of NspA surface epitopes for binding interactions with serum antibody, especially in the presence of abundant capsular polysaccharide.

Accordingly, the invention provides the fragments of NspA that are exposed on the cell-surface in Figure 32, namely ssslgsakg, nykapstdfkly, nrasvdlggsdsfsqt, and NY IGKVNTVKNVR s G, and also provides corresponding fragments from allelic variants of NspA. In addition, the invention provides sub-sequences of these fragments, comprising 7 or more contiguous amino acids from the fragments. The invention further provides proteins comprising these fragments. Nucleic acid encoding these fragments and proteins is also provided.

These NspA fragments, proteins comprising the fragments, and nucleic acid, may be used in the compositions of the invention, in particular as substitutes for full-length NspA. In a further aspect, these fragments, proteins and nucleic acids may be used as products in isolation, that is to say they need not exclusively be used in combination with other biological molecules.

It will be appreciated that the invention has been described by means of example only, and that modifications may be made whilst remaining within the spirit and scope of the invention.

Reactivity of anti-rNspA polyclonal antisera with native NspA exposed on the surface of live, encapsulated, Neisseria meningitidis B bacteria in relation to susceptibility to bacteriolysis and capsular production.

Meningococcal BSfrains NspA Cell Anti-rNspA Polysaccharide _______________________________ Surface Bactericidal Capsule

Strain Country Year Serologic ET Complex Reactivity1 Activity(1 /titer) Production __Classification _ ' _(1/titcr + SE) SWZ107" Switzerland 1980 4:P1.2 Γθ4 Positive >64 28±4 NG6/88D Norway 1988 NT: PI. 1 173 Positive 4 II5±23 CU385 Cuba 1980 4: PI.15 5 Positive 4 1I6±1

IH5341 Finland 1985 15:P1.7,16 ND Positive 16 176±6I BZ198 Netherlands 1986 NT:P- 154 Positive >64 362±1 NMB US 1968 2b:P1.2,5 ND Positive 16 244±20

8047 US 1978 2b: PI-.2 ND Positive 16 1125±5Q H44/76 Norway 1976 15: PI.7, 16 5 Positive 24 99±16 1000" USSR 1989 NTP1.5 61 Positive <4 287±12 S3446 IJS 1972 14: PI.23, 14 11 (A 1 cluster) Positive <4 (static— 16) 585±15l

Hiss'1 Norway 1973 15: PI.15 11 cluster Positive <4 (static=4) 656±14l BZ232" Netherlands 1964 NT:P1.2 76 Negative <4 1493±18 NG3/88" Norway 1988 8: PI. 1 A4 cluster Negative <4 498±105 MC58 UK 1985 15:PI.7.l6b 5 Negative <4 627±121 M136* US 1968 11: PI.15 D1 cluster Negative <4 1056±81 M986f US 1963 2a: Pi.5,2 B2 cluster Negative <4 I442±206 NGP165 Norway 1974 NT:P1.2 37 Negative <4 138±6 " Denotes strains that have been characterized further by multilocus sequence typing [Maiden, 1998], * Denotes strains obtained from the Frasch collection, US FDA. 8047 was obtained from W. Zollinger, Walter Reed Army Institute of Research, Washington, D.C. MC58 is the strain selected by TIGR for genomic sequencing. J351 was obtained from M. Sarvas, National Public Health Institute, Helsinki, Finland. The remaining strains are from the collection described by Seiler et al, [Seiler, 1996], ET data are from Caugnant et al [/. Infect. Dis. (1990) 162:867-874], and Seiler et al. 1 By indirect fluorescence flow cytometry with anti-rNspA antisera. J Dilution of anti-rNspA antisera that when incubated for 60 min. with bacterial cells and 20% human complement yielded >50% decrease in CFU/ml, compared to that at time 0. “Static” refers to strains that were inhibited but not killed in the assay (>50% but <100 % survival at 60 mins). *Titre defined as dilution of capsule extract giving 50% inhibition of antibody binding to meningococcal B polysaccharide antigen in an ELISA.

Claims (10)

1. A composition comprising a first biological molecule from a Neisseria bacterium and a second biological molecule from a Neisseria bacterium.

2. The composition of claim 1, wherein the first biological molecule is selected from the group consisting of SEQ IDs 1-8376.

3. The composition of claim 2, wherein the first biological molecule is selected from the group consisting of SEQ IDs 1-4002.

4. The composition of any preceding claim, wherein the second biological molecule is selected from the group consisting of SEQ IDs 1 -8376.

5. The composition of claim 4, wherein the second biological molecule is selected from the group consisting of SEQ IDs 1-4002.

6. The composition of any one of claims 1 to 4, comprising two or more biological molecules selected from SEQ IDs 1-4002.

7. The composition of any one of claims 1 to 4, comprising one or more biological molecules selected from SEQ IDs 1-4002 combined with one or more biological molecules selected from SEQ IDs 4003-8376.

8. The composition of any one of claims 1 to 4, comprising one or more biological molecules selected from SEQ IDs 1-4002 &amp; 4057-8376 combined with the NspA protein disclosed in W096/29412, preferably in mature form.

9. The composition of any one of claims 1 to 4, comprising one or more biological molecules selected from SEQ IDs 1-8376 combined with 953 protein.

10. The composition of any preceding claim for use as a medicament