WO2001076623A1

WO2001076623A1 - Iscom polypeptide delivery system for helicobacter pylori antigens

Info

Publication number: WO2001076623A1
Application number: PCT/SE2001/000800
Authority: WO
Inventors: Adam Shapiro; Gautam Sanyal
Original assignee: Astrazeneca Ab
Priority date: 2000-04-12
Filing date: 2001-04-09
Publication date: 2001-10-18
Also published as: AU2001247031A1; GB0008877D0

Abstract

A process for producing a polypeptide delivery system comprising an ISCOM coupled to a Helicobacter pylori polypeptide or a fragment thereof, the process comprising (a) mixing together the polypeptide, cholesterol, a saponin and a phospholipid comprising a polar head group, in the presence of a detergent to form a solution of the polypeptide, cholesterol, phospholipid and saponin, and (b) removing detergent from the mixture so that the ISCOM forms, wherein in step (b) the head group of the phospholipid has a net charge and a terminal charge. The phospholipid can be phosphatidate or a derivative thereof. The invention also relates to a polypeptide delivery system obtainable by the process, a vaccine comprising such a delivery system and the use of such a delivery system in the manufacture of a vaccine.

Description

ISCOM polypeptide delivery system for Hehcobacter pylori antigens

The present invention relates to a process for producing a polypeptide delivery system comprising an ISCOM (ie, immune stimulating complex) coupled to a polypeptide. The present invention also relates to a polypeptide delivery system obtainable by the process, a vaccine comprising such a delivery system and the use of such a delivery system in the manufacture of a vaccine.

BACKGROUND TO THE INVENTION

Reference is made to Chapter 23 of "Vaccine Design, The Subunit and Adjuvant Approach", Ed. M F Powell and M J Newman, 1995, Plenum Press, which discusses the structure and conventional production of ISCOM's. The ISCOM is a cage-like structure with a diameter of 30-40 nm. ISCOM's can be coupled to, or otherwise closely associated with, antigens and therefore make suitable polypeptide antigen delivery systems for administration to a patient, eg for vaccine use, since they are effective for polypeptide antigen presentation to the patient's immune system. ISCOM's comprise at least one saponin, cholesterol and at least one phospholipid. Glycosides of the adjuvant Quil A are suitable saponins, Quil A being a preparation from the bark Quilaja saponaria Molina. Neutral phospholipids such as phosphatidylcholine (PC) and phosphatidylethanolamine (PE) are used in the formation of ISCOM's, these lipids being neutral in the sense that they are used with their polar head groups in the zwitterionic state, and thus the head groups have no net charge when forming the ISCOM's.

There is a problem, however, with the conventional processes for producing ISCOM's, because some polypeptides cannot be coupled to ISCOM's made with these processes. This can be readily detected by carrying out sucrose density centrifugation of a sample of ISCOM's made by a conventional process in the presence of a test polypeptide. Using conventional sucrose density centrifugation, a sucrose gradient is set up in a test tube, with the highest density of sucrose being at the bottom of the tube and the lowest density at the top. The sample to be analysed is placed at the top of the gradient and centrifugation is carried out. Fractions of the gradient are then analysed for cholesterol and polypeptide content. Cholesterol serves as a marker for the location of the ISCOM's. The presence of polypeptide in the ISCOM fraction indicates that the polypeptide is coupled to the ISCOM's. Thus, where a polypeptide does not couple to the ISCOM's, the free polypeptide is separated from the ISCOM's during centrifugation and therefore substantially does not appear in the same fraction as the ISCOM's. Where a polypeptide has the same buoyant density as the ISCOM's, but the polypeptide is not coupled to the ISCOM's, false positives can be identified by carrying out a separate experiment with the polypeptide in the absence of the ISCOM's. The formation of ISCOM's can be confirmed using negative contrast electron microscopy.

WO 00/48630 claims priority dates of 17 February 1999 and 27 July 1999 and was published after the priority date of the present application. WO 00/48630 discloses an immunogenic complex (eg, an ISCOM) comprising a charged organic carrier (eg, phospholipid of an ISCOM) electrostatically associated with a charged antigen. Manufacture of an ISCOM associated with Helicobacter pylori HpE protein is disclosed (Example 2), where the manufacture is carried out at pH8 and the ISCOM is formulated with dipalmitoylphosphatidylcholme (DPPC) or cardiolipin (CDL) or diphosphoryl lipid A (DPL) or dipalmitoylphosphatidyl glycerol (DPPG). There is also a disclosure (Example 7) of manufacturing an ISCOM associated with Helicobacter pylori HpC protein, where the manufacture is carried out at pH7.2 and the ISCOM is formulated with DPPC or DPL. There is a further disclosure (Example 12) of manufacturing an ISCOM associated with Helicobacter pylori HpE protein that is tagged with 6 histidine residues or 6 histidine residues and 6 lysine residues, where the manufacture is carried out at pH7.2 and the ISCOM is formulated with DPPC or DPPC and dipalmitoyl-rac-glycerol-3(8-(3,6-

TM dioxy)octyl-l-amino-N,N-diacetic acid (DPIDA). ISCOPREP 703 is used as a saponin

TM preparation for all manufacture. DPPC has no net charge at pH's 7.2 and 8. ISCOPREP 703 is disclosed as being a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A. Reference is made to

TM

WO 00/48630 for further details of ISCOPREP 703.

SUMMARY OF THE INVENTION

The present inventors addressed the limitation of the conventional processes, and of all the components involved in these processes they determined that a change of phospholipid, specifically to the use of a phospholipid having certain charge characteristics, provides a process which has broad applicability to polypeptides, including polypeptides which are unsuited to the conventional processes of ISCOM production as discussed above.

To this end, the present invention provides a process for producing a polypeptide delivery system comprising an ISCOM coupled to a Helicobacter pylori polypeptide or an antigenic fragment thereof, the process comprising

(a) mixing together the polypeptide, cholesterol, a saponin and a phospholipid comprising a polar head group, in the presence of a detergent to form a solution of the polypeptide, cholesterol, phospholipid and saponin, and (b) removing detergent from the mixture so that the ISCOM forms,

wherein in step (b) the head group of the phospholipid has a net charge and a terminal charge,

with the proviso that the following processes are excluded

(i) A process carried out at pH 8 where the polypeptide is Helicobacter pylori HpE protein, cardiolipin (CDL) or diphosphoryl lipid A (DPL) or dipalmitoylphosphatidyl glycerol (DPPG) being used as a sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50- 10% by weight of Fraction C of Quil A; (ii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpC protein, DPL being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50- 10% by weight of Fraction C of Quil A; and (iii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpE protein that is tagged with 6 histidine residues or 6 histidine residues and 6 lysine residues, dipalmitoylphosphatidylcholme (DPPC) and dipalmitoyl-rac-glycerol-3(8-(3,6- dioxy)octyl-l-amino-N,N-diacetic acid (DPIDA) being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A.

Preferably, the present invention is confined exclusively to Helicobacter pylori polypeptides or antigenic fragments that substantially do not couple to ISCOM's when a conventional process employing only neutral phospholipid is used to form the ISCOM's.

The present invention also provides a polypeptide delivery system obtainable by the process of the present invention, eg for use as a medicament. Furthermore, the present invention provides a vaccine composition comprising such a delivery system.

In addition, the present invention provides the use of a polypeptide delivery system obtainable by the process of the invention, in the manufacture of a vaccine for administration to a mammalian patient, to treat or prevent Helicobacter pylori infection in the patient.

FIGURES

Fig. 1 : Analysis of protein-ISCOM association by sucrose gradient density ultracentrifugation, of His-HOP38 (-11) ISCOMs made with equal parts of dimyristoylphosphatidic acid and dimyristoylphosphatidylethanolamine. Fig. 2: Analysis of protein-ISCOM association by sucrose gradient density ultracentrifugation, of His-HOP38 (-11) ISCOMs made with neutral dipalmitoylphosphatidylcholme.

Fig. 3: Analysis of protein-ISCOM association by sucrose gradient density ultracentrifugation, of His-HOP38 (-1 1) ISCOMs made with neutral dipalmitoylphosphatidylethanolamine.

Fig. 4: Analysis of protein-ISCOM association by sucrose gradient density ultracentrifugation, of His-HOP38 (-11) ISCOMs made with dimyristoylphosphatidic acid.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a process for producing a polypeptide delivery system comprising an ISCOM coupled to a Helicobacter pylori polypeptide or an antigenic fragment thereof, the process comprising

(a) mixing together the polypeptide, cholesterol, a saponin and a phospholipid comprising a polar head group, in the presence of a detergent to form a solution of the polypeptide, cholesterol, phospholipid and saponin, and

(b) removing detergent from the mixture so that the ISCOM forms,

with the proviso that the following processes are excluded

(i) A process carried out at pH 8 where the polypeptide is Helicobacter pylori HpE protein, cardiolipin (CDL) or diphosphoryl lipid A (DPL) or dipalmitoylphosphatidyl glycerol (DPPG) being used as a sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50- 10% by weight of Fraction C of Quil A;

(ii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpC protein, DPL being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90%) by weight of Fraction A of Quil A and 50- 10% by weight of Fraction C of Quil A; and

(iii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpE protein that is tagged with 6 histidine residues or 6 histidine residues and 6 lysine residues, dipalmitoylphosphatidylcholine (DPPC) and dipalmitoyl-rac-glycerol-3(8-(3,6- dioxy)octyl- 1 -amino-N,N-diacetic acid (DPIDA) being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A.

Without being bound by any particular theory, we believe that electrostatic interaction between the phospholipid and polypeptide when detergent is removed may be involved in promoting close coupling between the phospholipid and polypeptide.

Table 1 shows illustrative examples of the structure of some common phospholipids at neutral pH. Representative nonpolar tails are shown. In practice, these may vary.

Table 1

Polar Head Group

Nonpolar tails

PHOSPHOLIPID X

Phosphatidate O

Phoshpatidylethanolamine (PE) O-CH₂-CH₂-N H₃

Phoshpatidylcholine (PC) O-CH₂-CH₂-N (CH₃)₃

Phosphatidylserine (PS) O-CH₂-CH(N H₃)(COO^")

Phosphatidylinositol (PI)

Phosphatidylglycerol (PG) O-CH₂-CH(OH)-CH₂(OH) As the general formula in Table 1 shows, the phospholipids have a polar head group. At neutral pH, the head group of PE has no net charge (ie, it is zwitterionic, having one positive charge and one negative charge). The same applies to PC at neutral pH. Each of phosphatidate, PS, PI and PG, on the other hand, has a polar head group with a net charge at neutral pH.

Looking at Table 1 again, it can be seen that at neutral pH the head group of phosphatidate has two terminal charges (two negatively charge oxygens bonded to the phosphorus). The polar head group of PE has one terminal charge (positive charge on the nitrogen). The polar head group of PC has one terminal charge (positive charge on the nitrogen). The polar head group of PS has two terminal charges (one negatively charged oxygen of the carboxylate group and one positive charge on the nitrogen). The polar head groups of PI and PG do not possess terminal charges.

Thus, when step (b) is carried out at pH7, a suitable phospholipid for use in the process can be selected from phosphatidate and PS, or a derivative thereof that comprises a polar head group having a terminal charge and a net charge. It will be readily apparent to the skilled addressee how to make conventional changes to the phospholipids (eg, changes to the glycerol backbone and/or changes in the head group) while retaining these charge characteristics. Preferably, the terminal charges of a derivative are provided on the same atoms as in the parent phospholipid. For example, a suitable derivative of PS could retain the terminal negative charge on the oxygen. Additionally, or alternatively, the derivative could retain the terminal positive charge on the nitrogen.

Suitable phospholipids for use in the present invention are, therefore, ones that have the basic structure of a polar head group and a nonpolar tail region, with the head group having a net charge and a terminal charge. In one embodiment, the phospholipid is selected from phosphatidate, PE, PS and PI, or a derivative thereof comprising a polar head group having a terminal charge and a net charge at the pH of step (b). Various hydrophobic tails of the phospholipid are suitable. For phosphatidate, dimyristoylphosphatidate, or dipalmitoylphosphatidate, or mixtures of these are preferred. In general, preferred phospholipids are ones having a fatty acyl chain in each of the 1 and 2 positions. An example is l,2-diacyl-sH-glycero-3-phosphate (eg, the sodium salt thereof) or the corresponding phospholipid having two negatively charged oxygens in the head group. In general, preferably the fatty acyl chains are independently 3C to 24C, preferably 14C to 20C. For example the fatty acyl chains can be independently selected from myristoyl, palmitoyl, oleoyl and stearoyl. The phospholipid l,2-dimyristoyl-_s«-glycero-3- phosphate, or the corresponding phospholipid having two negatively charged oxygens in the head group are the most preferred.

It is possible to use a mixture of phospholipids in step (a), at least one of which has a polar head group with an overall net charge and a terminal charge in step (a). Preferably, where a mixture is used, all but one, or two or three of the phospholipids have an overall net charge and terminal charge in step (a). Most preferably, where a mixture is used, all of the phospholipids have an overall net charge and terminal charge in step (a).

In a particular preferred embodiment, phosphatidylethanolamine (PE) is used together with at least one phospholipid having a polar head group with a net charge and a terminal charge in step (a). In step (a), the head group of the PE preferably also has a net charge and a terminal charge. The inclusion of PE is particularly useful, since PE is predominant in bacterial cell membranes (eg, Helicobacter pylori membranes) and thus the ISCOM of the delivery system approximates the natural phospholipid environment for polypeptides derived from bacterial cell membranes. The likelihood of correct polypeptide folding in the delivery system is therefore promoted by the inclusion of PE. As the experiments below demonstrate, the inclusion of PE is also beneficial for increasing the proportion of complete ISCOM's produced during the process of the present invention.

A preferred phosphatidylethanolamine is one having a fatty acyl chain in each of the 1 and 2 positions of a glycerol backbone. In general, preferably the fatty acyl chains are independently 3C to 24C, preferably 14C to 20C. For example the fatty acyl chains can be independently selected from myristoyl, palmitoyl, oleoyl and stearoyl. The phospholipid dimyristoylphosphatidylethanolamine is the most preferred. In a preferred embodiment, the phosphatidylethanolamine is used in step (a) in a weight ratio of 4:1 to 1 :4 (preferably, 3:1 to 1 :3 or 2: 1 to 1 :2) of phosphatidylethanolamine : the total amount of phospholipid having a net charge and a terminal charge. A ratio of 1 : 1 is most preferred (eg, equal amounts by weight of dimyristoylphosphatidylethanolamine and dimyristoylphosphatidate).

Preferably, in step (a) the weight ratio of cholesterol : saponin : polypeptide : total amount of phospholipid having a net charge and a terminal charge is 0.25-2 : 1-6 : 0.1-1 : 0.25-1 (: 0.25-1), the bracketed information relating to the proportion of PE if this is additionally used in step (a) as described above. In this ratio, a figure of 1 for cholesterol is one example. In the ratio, a figure of 4 for saponin is one example. In the ratio, a figure of 0.4 for the polypeptide is one example. In the ratio, a figure of 0.5 for the total amount of phospholipid having a net charge and a terminal charge is preferred. In the ratio, a figure of 0.5 for PE is preferred. An example of a suitable ratio is 1 : 4 : 1 : 0.5 (: 0.5) of the respective components.

The polypeptide in step (b) has a surface charge of opposite sign to the terminal charge of the phospholipid head group. Preferably, the present invention is confined exclusively to Helicobacter pylori polypeptides or fragments that substantially do not couple to ISCOM's when a conventional process employing only neutral phospholipid is used to form the ISCOM's. This is readily determined by sucrose density centrifugation, as discussed above. In this preferred embodiment, the term "polypeptide" in the context of the present invention is to be construed in this way only. By substantially not coupling to ISCOM's we mean that the polypeptide present in the ISCOM-containing fraction represents no more than 50%, 40%, 30%, 20%, 15%, 10%, 1%, 0.5%, 0.25% or 0.1% of the total amount of the polypeptide in all of the fractions. In some cases no polypeptide is present in the ISCOM-containing fraction. The proportion of polypeptide present in the ISCOM- containing fraction can be determined by dividing the amount of polypeptide present in the ISCOM fractions (eliminating false positives as discussed above) by the total amount of polypeptide. Suitable subsets of polypeptides for this embodiment are water soluble polypeptides or those having specified HLB values. Preferably, the polypeptide has a hydrophile-lipophile balance (HLB) of 10 or more; most preferably 13 or more. It will be readily apparent to the skilled person how to determine HLB values. In addition, reference is made to the following publications, which concern the determination of HLB values: Griffin, W. C, J. Soc. Cosmet. Chem. 1949, I, 311; Griffin, W. C, J. Soc. Cosmet. Chem. 1954, 5, 249; Davies, J. T., Proc. 2nd Int. Cong. Surf Acitivity; London, 1957, p. 417; Davies, j. T.; Rideal, E. K., Interfacial Phenomena; Academic: New York-San Fransisco- London, 1963. p 129; Davies, J. T., Progi-ess in Surface Science; Danielli, j. F., Parkhurst, K. G. A., Riddford, A. C, Eds.; Academic: New York, 1964; Vol. 2, p 129. Previous attempts in the art to couple water soluble polypeptides with ISCOM's have been based on denaturation of the polypeptide in order to expose hydrophobic regions of the polypeptide that would otherwise be buried inside the polypeptide's structure, in attempt to provide hydrophobic regions for interaction with the ISCOM's lipids. The denaturation is clearly undesirable, and the present invention avoids this while allowing the use of water soluble polypeptides in their natural conformation.

Water insoluble polypeptides are distinguished from water soluble polypeptides by their ability to associate with non-ionic detergents to form micelles, whereas water soluble polypeptides do not associate with such detergents to form micelles (A Practical Guide to Enzymology, C H Suelter, John Wiley & Sons Publishers, 1985, ISBN 0-471-86431-5, pp 71-72). Water insoluble polypeptide/non-ionic detergent micelles can be easily detected, because the electrophoretic mobility of the polypeptide when incorporated in the micelles is different from the polypeptide's electrophoretic mobility in the absence of the non-ionic detergent. Since water soluble polypeptides do not associate with non-ionic detergents, there is no change in electrophoretic mobility for these polypeptides when in the presence or absence of a non-ionic detergent. The first step is to mix the polypeptide to be tested with a non-ionic detergent, then with an ionic detergent. If micelles are formed (ie the polypeptide is water insoluble), the ionic detergent subsequently added becomes incorporated in the micelles thus changing the electrophoretic mobility of the polypeptide.

Another test for water insoluble polypeptides is as follows (A Practical Guide to

Enzymology, 1985, C H Suelter, John Wiley & Sons Publishers, ISBN 0-471-86431-5, pp

® 71-72). The polypeptide is dispersed in Triton X-114 at O°C. When the temperature of this detergent is raised above 20°C, its cloud point, separation into two phases occurs: an aqueous phase and a detergent phase. Water soluble polypeptides are recovered in the aqueous phase, whereas water insoluble polypeptides are found in the detergent phase.

In one embodiment, the polypeptide is a transmembrane polypeptide, ie it has at least one transmembrane domain. The following reference describes a method for predicting the location of transmembrane beta-strands in membrane proteins of the beta-barrel type: Ponnuswamy, P. K. and Gromiha, M. M. (1993) InternationalJournal ofPeptide and Protein Research 42: 420-431. A variety of prediction algorithms are used for predicting transmembrane alpha-helices, the most well-known being that of Kyte and Doolittle: Kyte, J. and Doolittle, R. F. (1982) Journal of Molecular Biology 157: 105-132.

It is possible to use a mixture of polypeptides in the process of the present invention. In one embodiment, one or more, or all of the polypeptides substantially do not couple to ISCOM's when a conventional process is used.

An example of a suitable polypeptide is a Helicobacter pylori outer membrane protein or one that is a expressed on the surface of Helicobacter pylori. A delivery system comprising a Helicobacter pylori polypeptide or fragment thereof can be used in the manufacture of a vaccine for administration to a mammalian patient, to treat and/or prevent Helicobacter pylori infection in the patient.

In on embodiment, the Helicobacter pylori polypeptide is HOP38. Other suitable polypeptides are those disclosed in WO 96/40893, WO 97/37044, WO 98/18323 and WO98/24475, each of which is incorporated herein by reference specifically to provide details of the polypeptides and homologues thereof that can be used in the present invention.

In respect of HOP38, reference is made to WO 96/40893 which discloses HOP38 sequence information (SEQ ID NOs: 764, 1086 and 1537). This disclosure relating to HOP38 is specifically incorporated herein by reference, and the skilled person will realise that the disclosed HOP38 polypeptide can be used in the present invention. SEQ ID NO's: 1 to 10 below also represent HOP38 sequence information. It is to be understood that in place of a complete (ie, full-length) HOP38 polypeptide, one can use an immunogenic or antigenic fragment of the complete polypeptide. Thus, we use the term HOP38 herein to include such fragments, with or without fusion to additional sequences (eg, non-H. pylori sequences and/or non-ΗOP38 H pylori sequences).

Preferably, the HOP38 polypeptide has an amino acid sequence that is identical to, or substantially similar to, an amino acid sequence selected from positions 21 to 270 of SEQ ID NO: 2; positions 24 to 273 of SEQ ID NO: 4; positions 24 to 273 of SEQ ID NO: 6; and positions 1 to 250 of SEQ ID NO: 8. In one embodiment, the HOP38 polypeptide comprises an amino acid sequence identical to an amino acid sequence selected from SEQ ID NO: 2, 4, 6, 8 and 10.

By "substantially similar" when referring to the amino acid sequence of the HOP38 polypeptide we mean one or more of the following: (i) the amino acid sequence is at least 60%, 70%, 80%, 90%, 95%, 98% or 99% homologous to an amino acid sequence selected from positions 21 to 270 of SEQ ID NO: 2; positions 24 to 273 of SEQ ID NO: 4; positions 24 to 273 of SEQ ID NO: 6; and positions 1 to 250 of SEQ ID NO: 8; (ii) the amino acid sequence is at least 5, 10, 20, 50, 100, 150 or 200 contiguous amino acid residues of an amino acid sequence selected from positions 21 to 270 of SEQ ID NO: 2; positions 24 to 273 of SEQ ID NO: 4; positions 24 to 273 of SEQ ID NO: 6; and positions 1 to 250 of SEQ ID NO: 8; (iii) the amino acid sequence differs by 1, 2, 3, 5, or 10 residues from an amino acid sequence selected from positions 21 to 270 of SEQ ID NO: 2; positions 24 to 273 of SEQ ID NO: 4; positions 24 to 273 of SEQ ID NO: 6; and positions 1 to 250 of SEQ ID NO: 8; (iv) the polypeptide is encoded by a nucleotide sequence that hybridises under stringent (high, intermediate or low) conditions to the complement of a nucleotide sequence selected from positions 61 to 777 of SEQ ID NO: 1 ; positions 70 to 786 of SEQ ID NO: 3; positions 70 to 786 of SEQ ID NO: 5; and positions 82 to 798 of SEQ ID NO: 7. Immunogenic and antigenic fragments of a full- length polypeptide are therefore included in the definition.

A "complement" of a nucleotide sequence refers to an anti-parallel or antisense sequence that participates in Watson-Crick base-pairing with the original sequence.

The terms "homologous" and "homology" refer to the sequence similarity or sequence identity between two polypeptides or between the nucleotide sequences of two nucleic acid molecules. When a position in both of the two compared sequences is occupied by the same base or amino acid monomer subunit, e.g., if a position in each of two DNA molecules is occupied by adenine, then the molecules are homologous at that position. The percent homology between two sequences is a function of the number of matching or homologous positions shared by the two sequences divided by the number of positions compared x 100. For example, if 6 of 10 of the positions in two sequences are matched or homologous then the two sequences are 60% homologous. By way of example, the DNA sequences ATTGCC and TATGGC share 50% homology. Generally, a comparison is made when two sequences are aligned to give maximum homology. Algorithms for determining homology are NBLAST (for nucleotide sequences) and XBLAST (for amino acid sequences), each utilising default parameters (see http://www.ncbi.nlm.nih.gov).

Preferably, when there is the conservative amino acid replacement of an amino acid with another as shown in Table 2, the amino acids are regarded as being homologous to each other. Table 2 CONSERVATIVE AMINO ACID REPLACEMENTS

For a definition of high and low stringency, see Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1989, 6.3.1-6.3.6 and 6.4.1-6.4.10. This definition is incorporated herein by reference as an example of suitable hybridisation conditions for use in the present invention. Stringency of hybridisation is determined by: (a) the temperature at which hybridisation and/or washing is performed; and (b) the ionic strength and polarity of the hybridisation and washing solutions. Hybridisation requires that the two sequences are at least partly complementary; depending on the stringency of hybridisation, however, mismatches may be tolerated. Typically, hybridisation of two sequences at high stringency (such as, for example, in a solution of 0.5X SSC, at 65° C for both the hybridising and washing steps) requires that the sequences be essentially completely homologous. Conditions of intermediate stringency (such as, for example, 2X SSC at 65 ° C for both the hybridising and washing steps) and low stringency (such as, for example 2X SSC at 55° C for both the hybridising and washing steps), require correspondingly less overall complementarity between the hybridising sequences. (IX SSC is 0.15 M NaCl, 0.015 M Na citrate).

In one embodiment, the saponin is provided by an extract from Quillaja saponaria Molina. The saponin is preferably selected from ISCOPREP 703^™, QS-21^™, QuilA, QHA, QHC and mixtures of these. QHA and QHC are partially purified preparations of QuilA. See Barr, I.G., Sjolander, A., Cox, J.C. ISCOMs and other saponin based adjuvants. Advanced Drug Delivery Reviews 1998, 32:247-271.

As will be readily apparent to the skilled addressee, an appropriate detergent (or a combination of more than one detergent) should be chosen to solubilise the phospholipid and cholesterol used in step (a), and depending on the polypeptide also to solubilise the polypeptide if necessary. To this end, nonionic detergents may be used, as is conventional.

Examples are polyglycol esters and polyglycol ethers with aliphatic or arylaliphatic acids

® and alcohols, eg Triton X-100. Other examples are n-octylglucoside and MEGA- 10 (decanoyl-N-methyl glucamide). An example of another detergent is 3-(4-tert- octylphenoxy)-l-propanesulfonate (TOPPS) or a salt thereof. In step (b) the detergent can be removed by any route conventionally used in the formation of ISCOM's. In this respect, reference is made to Chapter 23 of "Vaccine Design, The Subunit and Adjuvant Approach", Ed. M F Powell and M J Newman, 1995, Plenum Press, which mentions the use of dialysis (eg, for n-octylglucoside and MEGA- 10) or ultracentrifugation through a sucrose gradient when nondialysable detergents such as

® Triton X-100 are used. Diafϊltration or dilution with ultrafiltration can also be used.

The skilled person can readily determine using routine techniques the appropriate workable pH at which steps (a) and (b) are performed. The pH will be selected bearing in mind the charge characteristics of the phospholipid at the selected pH and polypeptide stability (eg, substantially to retain the native configuration of the polypeptide in the product (ie, the product being the delivery system), or a minimum requirement being the maintenance of polypeptide antigenicity or immunogenicity in the product where vaccine use is contemplated for this). In addition, the pH will affect the surface charge of the polypeptide, and at all workable pH values substantially all polypeptides will have a surface charge of opposite sign to the terminal charge of the phospholipid, which we hypothesise plays a part in the close coupling of the polypeptide with the ISCOM's achieved by the present invention.

Where the pH is matched to the protein stability, the phospholipid should then be chosen so that in step (b) there is at least one phospholipid having a net charge and a terminal charge at this pH. Where the polypeptide and phospholipid are selected first, the pH should be matched to these components so that the phospholipid has a net charge and a terminal charge in step (b) and to provide protein stability.

Where a phospholipid with a net charge and a negative terminal charge is required, a lower pH in step (b) will tend to lead to a lower proportion in the mixture of phospholipid molecules bearing the negative terminal charge. The pH should therefore be sufficiently high to ensure that there are molecules of the phospholipid in the mixture that bear the negative terminal charge and have a net charge. For this purpose, a generally applicable preferred pH range for step (b) (and optionally in step (a)) is pH 4 or more. More preferably, the pH is 4 or more, but preferably < pH 8 (or optionally no more than the pi of the polypeptide). A pH of 7 for step (b) (and optionally in step (a)) will be suitable in many cases where the negative terminal charge and net charge are required.

Where a phospholipid with a net charge and a positive terminal charge is required, a higher pH in step (b) will tend to lead to a lower proportion in the mixture of phospholipid molecules bearing the positive terminal charge. The pH should therefore be sufficiently low to ensure that there are molecules of the phospholipid in the mixture that bear the positive terminal charge and have a net charge. For this purpose, a generally applicable preferred pH range for step (b) (and optionally in step (a)) is pH 8 or less. More preferably, the pH is 8 or less, (but optionally equal to or more than the pi of the polypeptide). A pH of 7 for step (b) (and optionally in step (a)) will be suitable in many cases where the positive terminal charge and net charge are required.

PS has both positive and negative terminal charges and a net charge at pH 7. Most polypeptides are stable at this pH, making these conditions of pH and phospholipid suitable for many applications to provide strong coupling of the polypeptide to the ISCOM.

In the polypeptide delivery system produced by the process, the polypeptide and ISCOM are coupled together (ie, closely associated) so that they form a combined unit that can be administered to a patient. Thus, when the delivery system comprises an immunogenic or antigenic polypeptide, the system can be used as a component of a vaccine composition. When provided as part of a vaccine composition, the delivery system can optionally be so provided in combination with a suitable additional adjuvant. The vaccine compositions can be used to treat and/or prevent a disease or infection (depending on the polypeptide antigen(s) of the delivery system) in a mammalian patient by administering an immunologically effective amount of the composition to the patient. The term "immunologically effective amount" means an amount that elicits an immune response by the patient to the polypeptide antigen(s) carried by the delivery system. An "immune response" is a response which eradicates, suppresses, prevents and/or reduces the risk of the infection or disease in the patient. Typically, an appropriate dose of the or each antigen per administration would be approximately lOμg to lOmg, preferably approximately 50μg to 5mg, for oral or intranasal administration. Suitable dosage forms include a frozen dispersion, freeze-dried particles or a liquid dispersion.

The term "immunogenic" indicates a capability of a polypeptide to elicit a humoral and/or cellular immune response in vitro or in vivo (eg, in a patient) with or without provision in a delivery system produced according to the invention.

Screening immunogenic polypeptides can be accomplished using one or more of several different assays. For example, in vitro, peptide T cell stimulatory activity is assayed by contacting a polypeptide known or suspected of being immunogenic with an antigen presenting cell which presents appropriate MHC molecules in a T cell culture. Presentation of an immunogenic polypeptide in association with appropriate MHC molecules to T cells in conjunction with the necessary costimulation has the effect of transmitting a signal to the T cell that induces the production of increased levels of cytokines, particularly of interleukin-2 and interleukin-4. The culture supernatant can be obtained and assayed for interleukin-2 or other known cytokines. For example, any one of several conventional assays for interleukin-2 can be employed, such as the assay described in Proc. Natl. Acad. Sci USA, 86: 1333 (1989) the pertinent portions of which are incorporated herein by reference. A kit for an assay for the production of interferon is also available from Genzyme Corporation (Cambridge, MA).

Alternatively, a common assay for T cell proliferation entails measuring tritiated thymidine incorporation. The proliferation of T cells can be measured in vitro by determining the amount of ^H-labelled thymidine incorporated into the replicating DNA of cultured cells. Therefore, the rate of DNA synthesis and, in turn, the rate of cell division can be quantified.

The term "antigenic" refers to the presence of an antigenic determinant that is recognised by an antibody, B-cell or T-cell (eg, one from a mammal). For example, the antigenic determinant may bind to such an antibody. Standard ELISA or other immunoassays can be used to detect this. The antigenic determinant is provided by a polypeptide of a delivery system produced according to the invention. The preferred administration route is parenteral or mucosal administration. Examples of suitable mammalian mucosa include the buccal, nasal, tonsillar, gastric, intestinal (small and/or large intestine), rectal and vaginal mucosa. Appropriate corresponding administration routes include oral, nasal, rectal and vaginal administration, with the oral, rectal and nasal routes being the most preferred.

EXPERIMENTAL

1. Use of PA & PE for the Formation of A Polypeptide Delivery System

Materials

• Tris base (Sigma)

Quilaja saponin fraction QH-A

Quilaja saponin fraction QH-C

«-octyl-β-D-glucopyranoside (nOG) (Sigma) cholesterol (Sigma) • l,2-dimyristoyl-s«-glycero-3-phosphoethanolamine (DMPE) powder (Avanti Polar

Lipids)

1 ,2-dimyristoyl--? 7-glycero-3 -phosphate (DMPA) powder (Avanti Polar Lipids)

3-(4-tert-octylphenoxy)-l-propanesulfonic acid, sodium (TOPPS) (Calbiochem)

His-Hop38(-l 1) protein contained in a buffer consisting of 20 mM Tris-HCl (pH 8.0) 150 mM NaCl, 0.1 mM EDTA, 2% (w/v) sucrose, 0.5% (w/v) Brij-35 (Fluka). His-

HOP38(-l 1) has the sequence set out in SEQ ID NO: 10 of the sequence listing. Method of preparation

To prepare a buffer, a 1M solution of Tris-HCl (pH 8.0 at room temperature) was diluted 100-fold with Milli-Q water. The resulting pH was 7.

To prepare a fresh solution of saponin 703, QH-A saponin at 7 mg/ml and QH-C saponin at 3 mg/ml were dissolved in buffer. This was kept on ice until use.

To prepare a fresh lipid/nOG solution, cholesterol at 10 mg/ml, DMPE at 5 mg/ml, DMPA at 5 mg/ml, and nOG at 40% (w/v) were dissolved in buffer. This required heating at 60°C for an hour or so.

To prepare a fresh TOPPS solution, TOPPS at 10% (w/v) was dissolved in buffer.

Concentrated His-Hop38(-l 1) was taken from storage at -80°C and thawed by swirling the container in a bath of cold tap water. This was placed on ice immediately.

The following ingredients were mixed together, in order, as listed: List 1 • 4 parts by volume of saponin 703 solution.

• 1 part by volume of lipid/nOG solution

• 1 part by volume of TOPPS solution

• buffer, then His-Hop38(- 11), such that the final concentration of His-Hop38(- 11 ) is 0.4 mg/ml and the total volume of His-Hop38(-l 1) and buffer was 4 parts by volume.

This was mixed by gentle vortexing.

The mixture was dialysed against 1000 volumes of buffer with a moderate rate of stirring, at room temperature, using dialysis tubing with 10,000 molecular weight cut-off. There was a change of dialysis buffer after 6-8 hours and dialysis overnight. The dialysis buffer was changed in the morning, and once more after 6-8 hours, followed by dialysis overnight. The total dialysis time was 2 days.

Concentration was performed to a sufficiently small volume using a Millipore Ultrafree-15 10,000 molecular weight cut-off centrifugal ultrafiltration device.

An alternative mixture is as follows:- List 2

4 mg/ml saponin 703 solution 4% nOG

1 mg/ml cholesterol 0.5 mg/ml DMPA 0.5 mg/ml DMPE 1% TOPPS

1 mg/ml His-HOP38(-l l)

0.09X HOP38 storage buffer (20mM Tris-HCl pH8, 150mM NaCl, 0.1 mM EDTA, 2% sucrose, 0.5% Brij-35)

Note that experiments are carried out at pH 7, where the DMPE is neutral (ie, has no net charge).

Results

Figure 1 shows the analysis, by sucrose gradient density centrifugation, of His-HOP38 (- 11) ISCOMs made with equal parts of dimyristoylphosphatidic acid and dimyristoylphosphatidylethanolamine according to List 1 above. The position of the ISCOMs is indicated by the cholesterol component. The coincidence of the peaks of cholesterol and polypeptide clearly indicates that the polypeptide is associated with the ISCOM's (compare this with the results below where PE only or PC only was used). 2. Use of PC for the Formation of A Polypeptide Delivery System

The protocol set out in part 1 above was followed, with the exception that dipalmitoylphosphatidylcholme (DPPC) was used in place of the PA and PE. At the experimental pH (pH 7), the PC was neutral.

The results are shown in Fig. 2, where it can clearly be seen that the polypeptide failed to co-localise with the ISCOM's, indicating that the polypeptide was not associated with the ISCOM's.

3. Use of PE Only for the Formation of A Polypeptide Delivery System

The protocol set out in part 1 above was followed, with the exception that dipalmitoylphosphatidylethanolamine (DPPE) was used in place of the DMPA and DMPE. At the experimental pH (pH 7), the PE was neutral.

The results are shown in Fig. 3, where it can clearly be seen that the polypeptide failed to co-localise with the ISCOM's, indicating that the polypeptide was not associated with the ISCOM's.

4. Use of PA Only for the Formation of A Polypeptide Delivery System

The protocol set out in part 1 above was followed, with the exception that only DMPA was used in place of the PA and PE.

The results are shown in Fig. 4, where it can be seen that polypeptide co-localised with the ISCOM's, indicating association of polypeptide with ISCOM's.

Claims

CLAIMS:

1. A process for producing a polypeptide delivery system comprising an ISCOM coupled to a Helicobacter pylori polypeptide or an antigenic fragment thereof, the process comprising

with the proviso that the following processes are excluded:-

(i) A process carried out at pH 8 where the polypeptide is Helicobacter pylori HpE protein, cardiolipin (CDL) or diphosphoryl lipid A (DPL) or dipalmitoylphosphatidyl glycerol (DPPG) being used as a sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A;

(ii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpC protein, DPL being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90% by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A; and (iii) A process carried out at pH 7.2 where the polypeptide is Helicobacter pylori HpE protein that is tagged with 6 histidine residues or 6 histidine residues and 6 lysine residues, dipalmitoylphosphatidylcholme (DPPC) and dipalmitoyl-rac-glycerol-3(8- (3,6-dioxy)octyl-l-amino-N,N-diacetic acid (DPIDA) being used as the sole phospholipid, and the saponin being provided by a saponin preparation comprising from 50-90%) by weight of Fraction A of Quil A and 50-10% by weight of Fraction C of Quil A.

2. The process of claim 1, wherein the phospholipid is selected from phosphatidate, phosphatidylethanolamine, phosphatidylserine and phosphatidylcholine, or a derivative thereof comprising a polar head group having a terminal charge and a net charge.

3. The process of claim 2, wherein the phospholipid is phosphatidate or a derivative thereof.

4. The process of claim 3, wherein the phosphatidate is dimyristoylphosphatidate or dipalmitoylphosphatidate, or a mixture thereof.

5. The process of any preceding claim, wherein the terminal charge is negative and step (b) is carried out at a pH >4.

6. The process of claim 5, wherein step (b) is carried out at a pH < 8.

7. The process of claim 5, wherein step (b) is carried out at a pH no more than the pi of the polypeptide.

8. The process of any one of claims 1 to 4, wherein the terminal charge is positive and step (b) is carried out at a pH <8.

9. The process of claim 8, wherein step (b) is carried out at a pH equal to or more than the pi of the polypeptide.

10. The process of any preceding claim, wherein step (b) is carried out at pH 7.

1 1. The process of any preceding claim, wherein in addition to said phospholipid having the net charge and terminal charge, phosphatidylethanolamine is used in step (b).

12. The process of claim 1 1, wherein the phosphatidylethanolamine is dimyristoylphosphatidylethanolamine.

13. The process of claim 10 or 11, wherein the phosphatidylethanolamine is used in step (a) in a weight ratio of 2:1 to 1 :2 of phosphatidylethanolamine : the total amount of phospholipid having a net charge and a terminal charge.

14. The process of any preceding claim, wherein a mixture of phospholipids is used in step (a), at least one of the phospholipids comprising a polar head group having a terminal charge and net charge.

15. The process of any preceding claim, wherein in step (a) the weight ratio of cholesterol : saponin : polypeptide : total amount of phospholipid having a net charge and a terminal charge is 0.25-2 : 1-6 : 0.1-1 : 0.25-1.

16. The process of any preceding claim, wherein said Helicobacter pylori polypeptide is an outer membrane polypeptide of Helicobacter pylori.

17. The process of any preceding claim, wherein the polypeptide is HOP38.

18. A polypeptide delivery system obtainable by the process of any preceding claim.

19. The polypeptide delivery system of claim 18 for use as a medicament.

20. A vaccine composition comprising the polypeptide delivery system of claim 19.

1. Use of a polypeptide delivery system, in the manufacture of a vaccine for administration to a mammalian patient, to treat or prevent Helicobacter pylori infection in the patient, wherein the delivery system is obtainable by the process of any one of claims 1 to 17 or is obtainable by the following process:-

A process for producing a polypeptide delivery system comprising an ISCOM coupled to a Helicobacter pylori polypeptide or an antigenic fragment thereof, the process comprising

wherein in step (b) the head group of the phospholipid has a net charge and a terminal charge.