CA2050635A1 - Class ii protein of the outer membrane of neisseria meningitidis having immunologic carrier and enhancement properties, and vaccines containing same - Google Patents

Abstract

TITLE OF THE INVENTION
THE CLASS II PROTEIN OF THE OUTER MEMBRANE OF
NEISSERIA MENINGITIDIS HAVING IMMUNOLOGIC CARRIER AND
ENHANCEMENT PROPERTIES, AND VACCINES CONTAINING SAME

ABSTRACT OF THE INVENTION
The Class II major immuno-enhancing protein (MIEP) of Neisseria meningitidis, purified directly from the outer membrane of Neisseria meningitidis, or obtained through recombinant cloning and expression of DNA encoding the MIEP of Neisseria meningintidis, has immunologic carrier as well as immunologic enhancement and mitogenic properties.

Description

2 ~

TITLE QF T~E INVENTION
THE CLASS II PROTEIN OF T~E OUTER MEMBRANE OF
NEISS~RIA MENIN&ITIDIS ~AVING IMMUNOLOGIC CARRI~R AND
E~HANCEMENT PROPERTIES, AND VACCINES CONTAINING SAME

BACKGROUND OF THE INVENTION
This application is a Continuation in Part of application USSN 555, 329, f iled on July l9, 1990.
The outer membrane protein complex (OMPC) of : 20 Neisseri~ is used as an ~mmu~Q.lo~ic c~ ~.in vacci~es ~or human uæe. OMPC consists of ve~icles containing a variety of proteins as well as membranous lipidæ, including lipopolyæaccharide (LPS
or endotoxin).
: OMPC has the property o~ immune enhancement, and when an antigen is chemically coupled to it, an increaæed antibody reæponse to:the antigen re ults.

::

105/~HB25 - 2 - 18159IA

OMPC is currently used in vaccines for human infants against infectious a~ents such as ~aemophilus influenz~, and renders the infants capable of mounting an IgG and memory immune response to polyribosyl ribitol phosphate (PRP) of ~. influenzae.
when PRP is chemically coupled to OMPC.
OMPC is a mixture of a variety of proteins and lipids, and it was not known which component or components of OMPC bestows the beneficial immune enhancing effect to the coupled antigens. ~owever, some potentially negative aspects of using OMPC in human vaccines include LPS related reactions such as fever, endotoxic shock, hypotension, neutropenia, activation of the alternative complement pathway, intravascular coagulation, and possibly death.
Furthermore, OMPC-antigen conjugates are quite heterogeneous in that the antigen may become conjugated to any of-the protein moieties which make up OMPC.

OBJE~T~_QF THE INVENTION
It is an object of the present invention to provide substantially pure Class II, major immuno 2s enhancing protein (MIEP) derived directly from the outer membrane of Neisser~ia mening~idis, free from other N~isseria menin~itidis outer membrane components. It is another object of the present invention to provide substantially pure recombinant MIEP of the outer membrane of Nei~seria menin~itidis, produced in a recombinant host cell, comple~ely free of all other Neisseri~a menin~idis proteins. A

further object of the present invention is to provide an efficient immunocarrier protein for the enhancement of an immune response to antigens, comprising either MIEP purified directly from the outer membrane of Nei~seria meningitidis, or recombinant MIEP of Neisseria meningitidis produced in a recombinant host cell. Another object of the present invention is ~o provide a protein which possesses immune mitogenic activity, comprising either MIEP purified directly from the outer membrane Neisseria menin~itidis, or recombinant MIEP of ~eisseria m~ningi~ produced in a recombinant host cell. An additional object of the present invention is to provide vaccine compositions containing either the recombinant MIEP, or MIEP purified directly from the outer membrane of Neisser~ meningitidis. These and other objects will be apparent from the following description.

SUMMARY ~F I~E INVENTION
The present invention relates to the Cla38 II major immuno enhancing protein ~MIEP) of the outer membrane o~ Neisseria meni~gi~i~is, in substantially pure form, free from other ~ontaminating N.
m~ningiti~i~ outer membrane proteins and LPS. The MIEP of the present invention, whether purified directly ~rom the outer membrane of Neisseria menigitidi~ cells, or derived from a recombinant host cell producing recombinant MIFP of Neisseria meningi~idi~, possesses immunologic carrier and mitogenic activity. The MIEP of the present 105/G~B25 - 4 - 18~ 3~3 invention, when coupled to an antigen, is capable of immune enhancement in that the antibody response to the coupled antigen i8 augmented or the antigen is transformed to a T-dependent antigen which ensures that immunoglobulins of the IgG clags are produced.
The antigens which may be eoupled to the MIEP of the present in~ention include ~iral proteins, bacterial proteins and polysacharides, synthetic peptides, other immunogenic antigens, and weak or non-immunogenic antigens.

~ Q ~
105 /G~IB25 - 5 - 18159IA

D:E5TAILED DESCRIPTION OF THE INVENTION
It is known that certain substances which by themselves elicit an immune response which consists of only IgM class antibodies and no memory, can be transformed into fully immunogenic antigens which elicit IgM and IgG anitbodies as well as memory, by chemical coupling to a strongly (T-cell dependent) antigenic substance. This immunologic phenomenon is termed the "carrier effect", while the weak or lo non-immunogenic moiety, and the strongly antigenic substance are termed llhapten" and "carrier", respectively.
Injection of the hapten-carrier complex into an animal will result in the formation of antibodies by B-lymphocytes, some of which will be specific for, and bind to the hapten, and others which will be specific for, and bind to the carrier. An additional aæpect of the carrier effect is that upon a subsequent exposure to the hapten-carrier complex, a vigourous antibody response to the hapten ensues.
This is termed a memory, or anamnestic response.
The carrier effect appears to involve - functions mediated by certain T-lymphocytes, called l1helper T-lymphocytes". The carrier moleeule stimulates the helper T-lymphocytes to assist, in some way, formation of anti-hapten IgG class antibody-producing ~-lymphocytes and a memory response.
Helper T-lymphocytes are normally involved in the production by B-lymphocytesl of antibodies specific for a certain type of antigens, termed ~T-dependent" antigens, but not for other antigens 2 ~
105/~HB25 - 6 - 18159IA

termed "T-independent" antigens. A carrier molecule can convert a T-independent, weak or non-immunogenic hapten into a T-dependent, strongly antigenic molecule. Eurthermore, a memory response will follow a subseguent exposure to the hapten-carrier complex and will consist primarily of IgG, which is characteristic of T-dependent antigens and not T-independent antigens.
The utility of carrier molecules is not lo limited to use wlth T-independent antigens but can also be used with T-dependent an~igens. The antibody response to a T-dependent antigen may be enhanced by coupling the antigen to a carrier, even if the antigen can, by itself, elicit an antibody response.
Certain other molecules have the ability to generally stimulate the overall immune system. These molecules are termed "mitogens" and include plant proteins as well as bacterial products. Mitogens cause T and/or B-lymphocytes to proliferate, and can broadly enhance many aspecte of the immune response including increased phagocytosis, increased resistance to in~ection, augmented tumor-immuni~y, and increased antibody productlon.
Many infectious di ease causing agents can, by themselves, elicit protective antibodies which can bind to and kill, render harmless, or cause to be killed or rendered harmless, the disease causing agent and its byproducts. Recuperation from these diseases usually results in long lasting immunity by ~irtue of protecti~e antibodies generated against the highly antigenic components of the infectious agent.

, .. ' , ~ ~ ~3 3 ~ ~ ~

Protective antibodies are part of the natural defense mechanism of humans and many other animals, and are found in the blood as well aæ in other tissues and bodily fluids. It is the primary function of most vaccines to elicit protective antibodies against infectious agents and/or their byproducts, without causing disea3e.
OMPC from N. menin~iti~iS has been used successfully to induce antibody respon~es in humans when OMPC is chemically coupled to T-cell independent antigens, including bacterial polysaccharides. OMPC
contains severa1 bacterial outer membrane proteins as well as bacterial lipids. In addition, OMPC has a vesicular three dimensional structure.
The efficacy of OMPC as an immunologic carrier was thought to depend on one or more of the bacterial m~mbrane proteins, bacterial lipids, the vesicular three dimensional structure, or a combination of bacterial proteins, lipids, and vesicular structure. Applicants have discovered that one of the proteins, MIEP, posse6ses the immunologic carrier and immune enhancement properties of OMPC
vesicles, and is effective in purified form, free ~rom other N. memingi~idis membrane proteins and lipidæ, and in a non-vesicular three dimensional structure. Applicants have al~o discovered that MIEP, when chemically coupled to bacterial polysaccharide, functions as well as OMPC in inducing an antibody response to the polysaccharide.
Applicants have further discovered that MIEP is the Class II protein of the outer membrane of ~.
mening~ iS. The Class II protein of N.

meningitidis is a porin protein [Murakami, K., et al. . Sl989). Infection And Immunity, 57, pp.2318-23]. Porins are found in the outer membrane of all Gram negative bacteria.
While the present invention is e~emplified by MIEP of N. meningitidis, it is readily apparent to those skilled in the art that any outer membrane protein from any Gram negati~e bacterium, which has immunologic carrier and immune enhancement activity, is encompassed by the present invention. Examples of Gram negative bacteria include but are not limited to species of the genera Nei~sexia, Escherichia, Pseud~mona~, Hemophilus, Salmonella, Shigella, Bordetella, Klebsiella, Serratia, Yer~inia, Vibrio, and Enterobac~r.
MIEP may be employed to potentiate the antibody response to highly antigenic, weakly antigenic, and non-antigenic materials. The term "antigen" and "antigenic material" which are used interchangeably herein include ~ne or more non-viable, immunogenic, weakly immunognic, non-immunogenic, or desensitizing (antiallergic) agents of bacterial, viral, or other origin. The antigen component may consi~t of a dried powder, an aqueous phase such as an aqueouc solution, or an aqueous ~uspension and the like, including mixtures of the same, containing a non-viable, immunogenic, weakly immunogenic, non-immunogenic, or desensitizing agent or agent~.
The aqueous phase may conveniently be comprised of the antigenic material in a parenterally acceptabIe liquid. For example, t~e aqueous phase J ~

105/G~B25 - 9 - 18159IA

may be in the form of a vaccine in which the antigen is dissolved in a balanced salt solution, physiological saline solution, phosphate buffered saline solution, tissue culture fluids, or other media in which an organism may have been grown. The aqueous phase also may contain preservatives and/or substances conventionally incorporated in vaccine preparations. Adjuvant emulsions containing MIEP
conjugated antigen may be prepared employing techniques well kno~n to the art.
The antigen may be in the form of purified or partially purified antigen including but not limited to antigens derived from bacteria, viruses, mammalian cells, fungi, rickettsia; or the antigen may be an allergen including but not limited to pollens, dusts, danders, or extracts of the same; or the antigen may be in the form of a poison or a venom including but not li~ited to poisons or venoms derived from poisonous in~ects or reptiles. The antigen may also be in the form of a synthetic peptide, or a fragmçnt of a larger polypeptide, or any subportion of a molecule or romponent derived from bacteria, mammalian cell, fungi, viruses, rickettsia, allergen, poiæon or venom. In all cases, the antige~æ will be in the form i~ which their toxic or virulent properties have been reduced or deætroyed and which when introduced into a suitable host will either induce active immunity by the production therein of antibodies against the specific proteins, peptides, microorganisms, extract, or products of microorganisms used in the preparation of the antigen, poisons, venoms, or, in the case of allergens, they will aid in alleviating the symptoms of the allergy due to the æpeci~ic allergen.

105/GHB25 - 10 - 1815~IA

The antigens can be used either singly or in combination, for example, multiple bacterial antigens, multiple viral antigens, multiple mycoplasmal antigens, multiple rickettsial anti~ens.
multiple bacterial or viral toxoids, multiple allergens, multiple proteins, multiple peptides or combinations of any of the foregoing products can be conjuga~ed ~o MIEP.
Antigens of particular importance are lo derived from bacteria including but not limited to _.
pertussis, Leptospira pomona, and icterQh~emorrhagiae, ~. paratyphi A and B, C.
diphtheriae, . tetani, C. botuliaum, C. perfringens, ~. fes~ri, and other gas gangrene bacteria, B.
anthracla, P. pestis, P. multocida, Y. cholerae, Nesseria menin~itidis, N. gonorrheae, Hemophilus influen~ae, Treponema palid~, and the like; from ma~malian cells incl~1ding but not limited to tumor cells, virus in$ected cells, genetically engineered cells, cells grown in culture, cell or tissue extracts, and the like; from viruses including ~ut not limited to human T lymphotropic virus (multiple types), human immunodeficiency virus (multiple variants and types), polio virus (multiple types), 2s adeno virus (multiple types), parainfluenza virus (multiple types), measles, mumps, respiratory syncytial virus, influenza virus (various types), ~hipping fever virus (SF4), Western and Eastern equine encephalomyelitis virus, Japanese B.
encephalomyelitis, Russian Spring-Summer encephalomyelitis, hog cholera virus, Newcastle disea~e virus, fowl pox, rabies, feline and canine P3,.3 distemper and the like viruses, from rickettsiae including but not limited to epidemic and endemic typhus or other members of the spotted fever group, from various spider and ænake venoms or any of the known allergens, includin~ but not limited to those from ragweed, house dust, pollen extracts, grasæ
pollens, and the like.
The polysaccharides of this inven~ion may be any bacterial polysaccharides with acid groups, but are not intended to be limited to any particular types. Examples of æuch bacterial polysaccharides include Streptococcus pneumoniae (pneumococcal~ types 6A, 6B, lOA, llA, 18C, l9A, 19f, 20, 22F, and 23F, polysaccharides; Group B Streptococcus types Ia, Ib, II and III; ~acmophilus in~luenzae serotype b polysaccharide; Neisseria meningitidis serogroups A, B, C, X, Y, W135 and 29E polysaccharides; and Escherichia ~Qli Kl, K12, ~13, K92 and K100 polysaccharides. Particularly preferred polysaccharides, however, are those capsular polysaccharides selected from the group consisting of . influenzae æerotype b polysaccharides, such as descri~ed in Rosenberg et al., J. ~iol. Chem., 236, 2845-2849 (1961) and Zamenhof et al., J. Biol. Chem., 203, 695-704 (1953). ~tr~ptoco~cus pne~mQ~iae (pneumococcal) type 6B or type 6A polysaccharide, such as described in Robbins et al., Infection and Immunity, 26, No. 3 1116-1122 (Dec., 1979);
pnemococcal type l9F polysaccharide, such as described C. J. Lee et al., Reviews of Infectious ~iæeaseæ, 3, No. 2, 323-331 ~1981); and pneumococcal type 23F poly~accharide, ~uch a6 de~cribed in 0. Larm et al., Adv. Carbohyd Chem and Biochem., 33, 295-321, R. S. Tip~on et al., ed., Academic Press 1976.

~3~

MI~P can be purified from OMPC derived from cultures of N. meningitidis grown in the usual manner as described in U.S. Patent number 4,459,286 and U.S.
Patent number 4,830,852. OMPC purification can be done according to the methods described in U.S.
Patent number 4,271,147, 4,459,286, and 4,830,852.
MIEP can also be obtained from recombinant DNA engineered host cells by expression of recombinant DNA encoding MIEP. The DNA encoding MIEP
can be obtained from N. meningitidis cells ~Murakami, K. et al., (1989), Infection And Immunity, 57, pp.
2318], or the DNA can be produced synthetically using standard DNA synthysis techniques. DNA encoding MIEP
can be expressed in recombi~ant host cells including but no~ limited to bacteria, yeast, insect, mammalian or other animal cells, yielding recombinant MI~P, The preferred methods of the present invention for obtaining MIEP are purification of MIEP from OMPC and recombinant DNA expression of DNA encoding MIEP
deri~ed from ~. m~ningi~idis. with purification from OMPC most preferred.
Purified MIEP was prepared from OMPC -vesicles by sodium dodecylsulfate (SDS) lysis of the ~ vesicles followed by SDS polyacrylamide gel 2~ electrophoresis (PAGE). The MIEP was eluted from the gel, dialysed against a high pH buffer and concentrated. Standard methods o~ polyacrylamide gel - electrophoresis can be utilized to purify MIEP from OMPC vesicles. Such methods are described in Molecular Cloning: A Laboratory Manual, Sambrook, J.
et al., (1989), Cold Spring Harbor Laboratory Press, New York, and Current Protocols In Molecular Biology, (19B7) Ausubel F.M. et al., editors, Wiley and Sons, New ~or~.

2 ~ 3 Standard methods of eluting proteins from SDS-polyacrylamide gels are described in Eunkapiller, M.W., and Lujan, E., (1986), Purification Of Microgram Quantities Of Proteins By Polyacrylamide Gel Electrophoresis, in Methods of Protein Microcharacterization (J. Shively editor) Humanna Press, Clifton N.J., and Current Protocols In Molecular Biology ~1987), Ausubel, F.M., et al., editor~, Wiley and Sons, New York.
lo MIEP prepared in this manner is readily suitable for conjugation to antigens derived from bacteria, viruses, mammalian cellæ, rickettsia, allergens, poisons or venoms, fungi, peptides, proteins, polysaccharides, or a~y other antigen.
Recombinant MIEP can be prepared by expression of genomic N. meningitidi~ DNA encoding MIEP in bacteria, for e~ample E. coli or in yeast, for eæample ~. cer~vlsiae. To obtain genomic DNA
encoding MI~P, genomic DNA is extracted from N.
menin~itidi~ and prepared for cloning by either random ~ragmentation of high molecular welght DNA
following the technique of Maniatis, T. et al., (1978), Cell, 15, pp. 687, or by cleavage with a restriction endonuclease by the method of Smithieæ, 2s et al., (1978), Science, ~Q~, pp. 1248. The genomic DNA is the~ incorporated into an appro~riate cloning vector, for example lambda phage tsee Sambrook, J. et al., (1989), Molecular Cloning, A Laboratory Manual.
Cold Spring ~arbor Press, New York]. Alternatively, 3~ the polymerase chain reaction (PCR) technique (Perkin Elmer) can be used to amplify specific DNA sequences in the genomic DNA ~Roux, et al., (1989), r~

Biotechniques, 8, pp. 48]. PCR treatment requires a DNA oligonucleotide which can hybridize with specific DNA sequences in the genomic DNA. The DNA seguence of the DNA oligonucleotides which can hybridize to MIEP DNA in the N. meningi~ genomic DNA can be determined $rom the amino acid sequence of MIEP or by reference to the determined DNA sequence for the Class II major membrane protein of N. meningiti~s tMusakami~ k. et al., (1989), Infection and Immunity, 57~ PP- 23l8].
Recombinant MIEP can be separated from other cellular proteins by use of an affinity colum~ made with monoclonal or polyclonal antibodies specific for MIEP. These affinity columns are made by adding the antibodies to Affigel-10 (~iorad), a gel support which is pre-activated with N-hydroxysuccinimide esters such that the antibodies form co~alent linkages with the agarose gel bead support. The antibodie3 are then coupled to the gel via amide bonds with the spacer arm. The remaining activated esters are then quenched with lM ethanolamine ~Cl ~p~
8). The column is waæhed with water followed by 0.23 M glycine HCl (pH 2.6) to remove any non-conjugated antibody or extraneous proteln. The column is then equilibrated in phosphate buffered saline (pH 7.3) and the cell culture ~upernatants or cell extracts eontaining MI~P are slowly paæsed through the column. The column is then washed with phosphate buffered sali~e until the optical density (A280) falls to background7 then the protein is eluted with 0.23 M glycine-HCl (pH 2.6). The protein is then dialyzed against phosphate buffered salihe.

.

The conjugates of the present invention may be any stable polysaccharide-MIEP conjugates, coupled through bigeneric spacers containing a thioether group and primary amine, which form hydrolytically-labile covalent bonds with the polysaccharide and the MIEP. Preferred conjugates according to this invention, however, are those which may be represented by the formulae, Ps-A-E-S-B-Pro or Ps-A'-S-E'-B'-Pro, wherein Ps represents a poly-lo saccharide; Pro represents the bacterial proteinMIEP; and A-E-S-B and A'-S-E'-B' constitute bigeneric spacers which contain hydrolytically-stable co~alent thioether bondæ, and which form co~alent bonds (such as hydro-lytically-labile ester or amide bonds~ with the macromolecules, Pro and Ps. In the spacer, A-E-S-B, S is sulfur; E ~s the transformation product of a thiophilic ~roup which has been reacted with a thiol ~roup, and i~ represented by wherein R is H or CH3, and p is 1 to 3; A is -Ch(C~Iz)mY~C~2>n-NH-, J~

wherein W is O or NH, m is O to 4, n is O to 3, and Y
is CH2,0,S,NR~, or CHC02H, where Rl is ~ or Cl- or C2-alkyl, such that if Y is CH2, then both m and n cannot equal zero, and if Y is O or S, then m is greater than 1 and n is ~reater than 1; and B is ~(cH2)pcl~I(c~I2)qD- ~
Z O
wherein q is 0 to 2, Z is NH2, NH~R~, COOH, or H, where R' and p are as defined above, and D is ~, NR', H O O
or N-~(CH2)2~. Then in the spacer, A'-S-E'-B', S
W
i~ sulfur; A' is -~N~(CH2)aR"-, wherein a is 1 to HOY' 4, and R" is CH2, or N~(CH2)p, where Y' is NH2 or NHCOR', and W, p and R' are as defined above, and E' is the transformation product of a thiophilic group which has been reacted ~ith a thiol group, and ~s R
represented by -~H-~ wherein R is as defined above, and B' is -~-, or E' is o ~ N -.
~
o .
. , .

~.3 105/G~B25 - 17 - 18159IA

B~ is -(~H2)p~-, wherein p is 1 to 3. Further, of the bigeneric spacers, A-E-S-B and A~-S-E~-B~, the E-S-B and A'-S-E~ components are determinable and quantifiable, with this identification reflecting the covalency of the conjugate bond linking the side of the thioethersulfur which originates from the covalently~modified polysaccharide with the ~ide of the spacer which originates from the functionalized lo protein Then the conjugates, Ps-A-E-S-B-Pro, accord-ing to this invention may contain spacer3 whose com-ponents include derivati~es of, intçr ~1~: carbon dioxide, 1,4-butanediamine, and S-carboxymethyl-N-acetylhomocysteine; carbon dioxide, 1,5-pentanedia-mine, and S-carboxymethyl-N-acetylhomocyæteine; carbon dioxide, 3-oxa-1,5-pentanediamine, and S-carboxy-methyl-N-acetylhomocysteine; carbon dioxide, 1,4-butane-diamine, and S-carboxymethyl-N-acetyl-cysteine; carbon dioxide, 1,3-propanediamine, and S-carboxymethyl-N-benzoylhomocysteine; carbon dioxide, 3-aza-1,5-pentanediamine, and S~carbogy-methyl-N-acetylcysteine; and carbon dioxide, 1,2-ethanediamine, glycine, and S-(succin-2-yl)-N-acetylhomocysteine. The conjugates, Ps-A'-S-E'-B'-Prs, according to this invention, may contain spacers whose components include derivative~
of, int-~-F~ alia: carbon dioxide and S-carboxy-methylcy6teamine; carbon dioxide and S-(a-carboxy-ethyl)cysteamine; carbon dioxide and S-carbo~y-methylhomocysteaminc; carho~ dio~ide, S-(succin-2-yl)cy~teamine, and glycine; and carbon dioxide and S-carboxymethylcy~tei~e.

105/G~B25 - 18 - 18159IA

In the process of the present invention, the polysaccharide is covalently-modified by (a) solubilizing it in a non-hydroxylic organic solvent, then (b) activating it with a bifunGtional reagent, (c) reacting this activated polysaccharide with a bis-nucleophile, and finally, if necessary, further (d) functionalizing this modified polysaccharide by either reaction, (i) with a reagent generating electrophilic (e.g., thiolphilic) sites or, (ii) with lo a reagent generating thiol groups. The protein is conversely either reacted (i) with a reagent generating thiol groups or (ii) with a reagent generating thiolphilic sites, ~hen the covalently- ~
modi~ied polysaccharide and the functionalized protein are reacted together to form the stable covalently-bonded conjugate and the final mixture is purified to remove unreacted polysaccharides and proteins.
The process of this invention also includes selection of a nucleophile or bis-nucleophile which will react with the acti~ated polysaccharide to form a covalently-modified polysaccharide with pendant electrophilic æites or pendant thiol groups, thereby obviating the need to further functionalize the bis-nucleophile-modified polysaccharide prior to reacting the covalently-modified polysaccharide with the covalently-modified protein. Al~o, the functionalization of the protein to either moiety form may be accomplished in more than one step according to the selection of reactants in these step~.

In the first step toward covalently-modifying the polysaccharide, the solid poly-saccharide must be solubilized.
Since the nucleophilic alcoholic hydroxyl s groups of a polysaccharide cannot compete chemically for electrophilic reagents with the hydroxyls of water in an aqueous solution, the polysaccharide should be dissolved in non-aqueous (non-hydroxylic) solvents. Suitable solvents include dimethyl-formamide, dimethylsulfoxide, dimethylacetamide,formamide, N,N~-dimethylimidazolidinone, and other similar polar, aprotic solvents, preferably dimethylformamide.
In addition to the use of these solvents, converting the polysaccharides (e.g., the capsular polysaccharides of ~. influenzae type b, which are a ribose-ribitol phosphate polymers), which have acid hydrogens, such as phosphoric acid mono- and diesters, into an appropriate salt ~orm, causes the ~; 20 polysaccharides to become readily soluble in the above solvents. The acidic hydrogens in these macro-molecules may be replaced by large hydrophobic cations, such as ~ri- or tetra-(Cl- to C5)alkyl-a~monium, l-azabicyclo~2.2.2]octane,19 8-diazabicyclo [5.4.0]undec-7-ene or similar cations, particularly tri- or tetra-(Cl- to C5)alkylammonium, and the resultant tri- or tetraalkylammonium or similar salts of phosphorylated polysaccharides readily dissolve in the above solvents at about 17-50C, while being stirred for from one minute to one hour.

, 2 ~ C~

.

Partially-hydrolyzed ~. influenzae serotype B polysaccharide has been converted into the tetrabutyl-ammonium salt, then dissolved in dimethylsulfoxide (Egan et al., 1- Amer. Chem. ~Q~., 104, 2898 (1982)), but this product is no longer antigenic, and therefore useless for preparing vaccines. By contrast, Applicants accomplish the solubilization of an intact, unhydrolyzed polysaccharide by passing the polysaccharide through lo a strong acid cation exchange resin, in the tetraalkylammonium form, or by careful neutralization of the polysaccharide with tetraalkyl-ammonium hydroxide, preferably by the former procedure, and thereby preserve the viability o~ the ~olysaccharide for immunogenic vaccine use.
Subæequent steps are then directed to overcoming the other significant physico-chemical limitation to making covalent bonds to poly-saccharides, that being the lack of fu~ctional groups on the polysaccharideæ, other than hydro~yl groups, which are reactive enough with reagents commonly or practically used for functionalization of units with which bonding is desired. Acti~ation of the polysaocharide to ~orm an activated polysaccharide, reaction with bis-nucleophiles to form a nucleophile-functionalized polysaccharide, and functionalization with reagents generating either electrophilic sites or thiol ~roups, are all directed ~o covalently modifying the polysaccharide and developi~g functional groups on the polysaccharide in preparation for conjugation.

~;3~5~ 3 In the next step, the solubilized polysaccharide i3 activated by reaction with a bifunctional rea~ent at about 0O-50C, while stirring for ten minutes to one hour, with the crucial weight ratio of activating agent to poly~accharide in the range of 1:5 to 1:12. In the past, this activation has been accomplished by reaction of the polysaccharide ~ith cyanogen bromide. ~owever, derivati~ee activated with cyanogen bromide, ~hich lo has a "proclivi~y~ for vicinal diols, have shown transient stability during dialysis against a phosphate buffer. Therefore, while activation with cyanogen bromide is still possi~le according to the present invention, this reagent is poorly utilized in activation of polysaccharides and is not preferred.
Instead, preferred bifunctional reagents for activating the polysaccharide include carbonic acid O
derivatives, R2-~-R3, wherein R2 and R3 may be independently, azolyl, such as imidazolyl;
halides; or phenyl ester~, such as ~-nitrophenyl, or polyhalophenyl.
Carbonyldiimidazole, a particularly preferred reagent, ~ill react with the hydroxyl groups to form imidazolylurethanes of the polysaccharide, and arylchloroformate~, including, for example, nitrophenylchloroformate, will produce mi~ed carbonates of the polysaccharide. In each case, the resulting activated polysaccharide is very æuscep~ible to nucleophilic reagentQ, such as amines, and is thereby transformed into the respective urethaneæ.

In the next stage, the activated polysaccharide is reacted with a nucleophilic reagent, such as an amine, particularly diamines, for example, HN(CH2)mY(CH2)n- ~ , wherein m is O to 4, n is O to 3, and Y is C~2, O, S, NR', CHCO2~, where R' is H or a Cl- or C2-alkyl, such that if Y i8 C~2 -then both m and n cannot egual zero, and if Y is O or S, then m is greater than 1, and n is greater than 1, in a gross excess of amine (i.e., for example, a 50-to 100-fold molar excess of amine vs. activating agent used). The reacgion is kept in an ice bath for from 15 minutes to one hour then kept for 15 minutes to one hour at about 17 to 40C.
lS An activated polysaccharide, when reacted with a diamine, e.g., 1,4-butanediamine, would result in a urethane-form polysaccharide with pendant amines, which may then be further functionalized by acylating. Mixed carbonate~ will also readily react with diamines to re~ult in pendant amine groups.
Alternatively, the activated polysaccharide may be reacted with a nucleophile, such a~ a monohaloacetamide of a diaminoalkane, ~or example, 4-bromoacetamidobutylamine (see, W. B. Lawson e~ al., Hoppe SeYler's ~ eiol Chem., 349, 251 (1968)), to generate a covale~tly-modified poly~accharide ~ith pendant electrophilic ~ites. Or, the activated polysaccharide may be reacted with an aminothiol, ~uch as cysteamine (aminoethanethiol~ or cysteine, examples of derivatives of which are well-known in the art of peptide æynthesis, to produc~ a polysaccharide with pendant t~iol groups. In both cases, no additional functionalization is necessary prior to coupling the covalently-modified polysaccharide to the modified bacterial "carrier"
protein.
The last step in preparing the polysaccharide, the ~urther ~unc~ionali~ation, i~
necessary, of the polysaccharide, may take the form of either reacting the nucleophile-~unctionalized polysaccharide with a xeagent to generate electrophilic (i.e., thiophilic) sites, or with a reagent to generate thiol groups.
Reagents suitable for use in generating - electophilic sites, include for example, those for acylating to a-haloacetyl or a-halopropionyl, ~
derivative 3UC~ as X ~X (wherein R is H or CH3; X is Cl, Br or I; and X' is nitrophenoxy, dinitrophenoxy, pentachlorophenoxy, penta~luorophenoxy, halide, 0-(N-hydroxysuccinimidyl) or azido), particularly chloroacetic acid or a-bromcpropionic acid, with the reaction being run at a pH of 8 ~o 11 (maintained in this range by the addition o~ base, if necessary) and at a temperature of about 0 to 35C, for ten minutes to one hour, An amino-derivatized polysaccharide may 2s be acylated with activated maleimido amino acids (see, O. Keller et al, ~Ql~. Chim. ~cta., 58, 531 (1975)) to produce maleimido groups, O ~
--C(CH2)pN

o 105/G~B25 - 24 - 18159IA

wherein p is 1 to 3; with a 2-haloacetyling agent, such as p-nitrophenylbromoacetate; or with an a-haloketone carboxylic acid derivative, e.g., o HO2C ~ CH2Br (Ber., 67~ 1204, (1934)) in order to produce appropriately functionalized poly3accharides susceptible to thio substitution.
Reagents suitable for use in generating thiol groups include, for example, acylating reagents, such as thiolactones, e.g., CH2)p~

2() 0~\

wherein R4 is Cl- to C4-alkyI or mono- or bicyclic aryl, such as C~H5 or CloHl3, and p is 1 to 3;
NHCoR5 -03SSCH2(CH2)mCH-COX', wherein m is 0 to 4, R5 is Cl-to C4-alkyl or C6H5, and X' is as defined above, followed by treatment with HSCH2CH20~; or N~co~5 C2~5-S-S-CH2(CH2)mC~C0X', wherein m, R5 2 ~ 3 105/G~B25 - 25 - 181~9IA

and X' are as defined immediately above, then treat-ment with dithiothreitol. Such reactions are carried out in a nitrogen atmosphere, at about 0~ to 35C
and at a pH of 8 to 11 (with base added, as necessary, to keep th pH within this range), for one to twenty-four hours. For example, an amino-derivatized polysaccharide may be reacted with ~ COCH3 o~

to produce an appropriately-functionalized polysac-charide.
By the e steps then, covalently-modified polysaccharides of the forms, Ps-A-~*- or Ps-A'-SH-, 20 wherein E* i8 -CC~X or /
--C(CH2)pN
~r O

and A, A', R, X and p are as defined above, are produced.

105/G~B25 - 26 - 181~" '3 Separate functionalization of the protein to be coupled to the polysaccharide, involves reaction of the protein with one or more rea~ents to generate a thiol group, or reaction o~ the protein with one or more reagents to generate an electrophilic (i.e., thiophilic) center.
In preparation for conjugation with an electrophilic-functionalized poly3accharide, the protein is reacted in one or two steps with one or lo more reagents to generate thiol groups, such as those acylating reagents used for generating thiol groups on polysaccharides, aæ discussed on pages 15-17 a~ove. Thiolated proteins may also be prepared by aminating carboxy-activated proteins, such as those shown in Atassi et al., Biochem et Biophys. Acta, 670, 300, (1981), with aminothiols, to create the thiolated protein. A preferred embodiment of this process step involves the direct acylation of the pendant amino ~roups (i.e., lysyl groups) of the protein ~ith N-acetylhomocysteinethiolactone at about 0 to 35C and pH ~-11, for from five minutes to two hours, using equiweights of reactants.
- When E'B' is O
~ 11 ~ NCCH2)pC, O

'vi~
105/G~B25 - 27 - 18159IA

the conditions and method of preparing the functionalized protein are as discussed above for preparing the counterpart polysaccharide by reaction with activated maleimido acids.
In preparing for conjugation with a covalently-modified bacterial polysaccharide with pendant thiol groups, the protein is acylated with a reagent generating an electrophilic center, such acylating agents including, for example, XCH2~-X' and X~ X', ~herein X and Xl are as defined above; and ~CO\ O
N(CH2)aC--X~

wherein X' is a~ defined above. Suitable proteins with elec~ophilic centers al80 include1 for e~ample, those prepared by acylation of the pendant lysyl amino groups with a reagent, such a~ activated maleimido acids ~for example,O
~ 1l ~
~ NOC(CH2)n O

2 ~

or by reacting the carboxy-act~vated protein with monohaloacetyl derivatives o~ diamines. In both preparation reactions, the temperature is from 0 to 350C for from five mi~utes to one hour and the pH is from 8 to 11.
Formation o~ the conjugate is then merely a matter o~ reacting any of the covalently-modified polysaccharides havin~ pendant electrophilic centers with of the bacterial protein MIEP having pendant thiol groups at a p~ of 7 to 9, in approximate equiweight ratios, in a nitrogen atmosphere, for from six to twenty-four hours at from about 17 to 40C, to give a covalent conjugate. ~xamples of such reactions include:
OH O N~COCH3 Ps-CNCH2CH2CH2CH2NHCCH2Br + HSCH2CH2CHCO-Pro >

Psl~cH2c~I2cH2cH2N~/~cH2scH~cH2~Hcopro ' wherein an activated polysaccharide which has been reacted with 4-bromoacetamidobutylamine is reacted with a pro~ein whieh has been reacted with N-acetyl-homocysteinethiolactone, to form a conjugate, and:

~ Q ~ v ~ si ~
105/GHB25 - 29 - 1815gIA

P9 CN~--NCCH2 N

H~;CH2CH2NHCCH2CH2CPro PsCNH~' NHCCHz- N~3~ ~cH2cHzNHccH2c}l2cpro O

(where Y" is a C2-C8alkyl radical), wherein an 20 amino-derivatized polysaccharide which has been reacted with activated maleimido acids is reacted with a car~oxy-activated protein which ha~ been : aminated with an aminothiol, to form a conjugate.
Similarly, any of the covalently-modified 2S polysaccharides with pendant thiol groups may be reacted with the bacterial protein MIEP having pendant electrophilic centers to give a covalent conjugate. An example of such a reaction is:

"

~ 3~

O O O ~
Ps~:N}IcH2cH2sH + Pro~cH2cH2~-N(cH2)4N~coc~2Br OH OE O O
PS~C~2CH2SCH2~(CH2)4NH~c~2cH2~pr wherein an activated polysaccharide which has been reacted with a~ aminothiol is reacted with a carboxy-activated protein which has been reacted with monohaloacetyl derivatives of a diamine, to form a 1~ conjugate.
Should the electrophilic activity o~ an excess of haloacetyl groups need to be eliminated, reaction of the conjugate with a low molecular weight thiol, such as n-acetylcysteamine5 will accomplish this purpose. Use of this r@agent, n acetylcysteamine, also allows confirmation accounting of ~he haloacetyl moieties used (see Section D), because the S-carboxymethylcysteamine which i~ formed may be uniquely detected by the method of Spackman, Moore and Stein.
These conjugate~ are then centrifuged at about lOO,OOO x g using a fixed angle rotor for about two hour~ at about 1 to 20C, or are submitted to any of a variety of other purification procedures, including gel permeatio~, ion exclusion chromatography, gradient centrifugation, or other differential ad~orption chromatography, to remo~e non-covalently-bonded polysaccharides and proteins, using the coYalency assay for the bigeneric spacer (see belo~) as a method of following the desired biological activity.

~ ' .
, c~

The further separation of reagents may be accomplished by size-exclusion chromatography in a column, or in the case of very large, non-soluble proteins, separation may be accomplished by ultracentrifugation.
Analysis of the conjugate to confirm the covalency, and hence the stability of the conjugate, is accomplished by hydrolyzing (preferably with 6N
HCl at 110C for 20 hours) the conjugate, then quantitatively analyzing for the amino acid of the hydrol~tically-stable spacer containing the thioether bond and constituent amino acids of the protein. The contribution of the amino acids of the protein may be removed, if necessary, by comparison with the appropriate amino acid standard for the protein involved, ~ith the remaining amino acid value reflecting ~he covalency of the conjugate, or the amino acid of the sp~cer may be designed to appear outside the amino acid standard of the protein in the analysiæ. The covalency aæsay is also useful to monitor purification procedures to mark the enhancement of concentration of the biologicallyacti~e components. In the above exampleæ, hydrolysis of ~9~ NHCOC~3 Pæ NCH2CH2CH2CH2NH~C~2SCH2CH~COPro results in the release of S-carboxymethylhomocysteine, ~92CCH2SCH2CH2~HCO~H; hydrolysis of " ~

105/GHB25 - 32 ~ 18159IA

o o IJ
" ~ o o P~CNHY" NHCCH2 N
~ ~ CH2CH2NHCC~aCH2CPro S

results in the release of the aminodicarboxylic acid, Ho2cc~2c~sc~2cH2N~2; and hydrolysis of H02~
OH OH O O
P3~CH2C~2SC~2~(CH2)4NH~CH2CH2~Pro re~ults in the release of S-carboxymethylcysteamine, H2NCH2CH2SCH2CO~H by,cleavage of the Ps-A-~-S-B-Pro molecule at peptide linkages and other hydrolytically-unstable bonds. Chromatographic methods, such as those of Spackman, Moore, and Stein, may then be conveniently applied and the ratio of amino acid constituents determined.
Optimal production of IgG antibody requires collaboration of B and T lymphocytes with specificity : 25 for the antigen of interest. T lymphocytes are : incapable of reco~nizing polysaccharides but can provide help for anti-polysaccharide IgG antibody : responses if the polysaccharide is covalently linked to a protein which the T cell is capable of re~ognizing .

3~

105/G~B25 - 33 - 18159IA

In mice this requirement exists for secondary, as well as primary, antibody responses and is carrier-speci~ic, i.e. a secondary antibody response occurs only if the T helper cells have previously been eensitized with the carrier protein used for the secondary immunization. Therefore, the ability o~ a mouæe to make a secondary antibody response to a PRP-pro~ein conjugate is dependent on the presence of primed T lymphocytes with specificity lo for the carrier pxotein.
Demonstration of the ability of MIEP to provide carrier priming for anti-PRP antibody responses was done in mice adoptively primed with PRP
covalently linked to a heterolo~ous carrier, diphtheria toxoid ~DT). Adoptive transfer was used in order to determine whether the administration of lymphocytes primed with MIEP alone was sufficient to generate effective helper-T cell activity for anti-PRP antibody formation in response to PRP-OMPC.
Comparable secondary anti-PRP antibody responses were elicited by PRP-OMPC when lymphocytes primed with MIEP or OMPC were transferred, indicating that the T
~ cell recognition of OMPC resides in the~MIEP moiety.
~ P~P-MIEP conjugate~ ~ere tested for : ~25 immunogenicity in mice as well a~ infant Rhesus monkeys. The immune response in both o~ these animal : . models~s~are, with infant humans, a dificiency in their ability to generate antibody responses against : T-independent antigens such as bacterial polysaccharides. These animal are commonly used as model~ for assessment of the immune re~ponse of infant humans to various antigens.

;

~ 3 1051GHB25 - 34 - lB15~IA

Likewise, MIEP-peptide conjugates, for example where the peptide is an HIV principal neutralizing determinant (PND) peptide, may be prepared. One method of making such conjugates includes the formation of a bigeneric spacer between activated MIEP and activated XIV PND peptides as described and specifically claimed in copending application USSN _ , _ (Merck case MRL91/125). The linker may includ~ a polysaccharide moiety, as lo described in USSN 55~,558 (Merck case 18068).
The novel conjugate of this invention comprises MIEP, the major immuno enhancing protien of the outer membrane protei~ complex (OMPC) of Neisseria m~nin~ b, covalently linked to ~IV
PND peptides.
The conjugates are prepared by the process of covalently coupling actl~ated peptide to an activated protein. ~he peptide and protein components are separately activated to display either pendant electrophilic or nucleophilic groups so that co~alent bonds will form be-tween the peptide and the protein upon contact.
The covalent conjugate immunogens that result from the æeries o~ reactions described above may conveniently be thought o~ as a conjugate in ~hich multiple peptide functionalities are built upon a ~oundation of MIEP.
When the peptide components of the conjugate are capable of eliciting HIV neutralizin~ immune responses, the conjugates of this invention may be administred to mammals in immunologically effective amountæ, with or without additional i~munomodulatory, .

2~5~ s~

105/G~B25 - 35 - 18159IA

antiviral, or antibacterial compounds, and are useful for inducing mammalian immune responses against the peptidyl portion of the conjugates, for inducing EIV-neutralizing antibodies in mammals, or for ma~ing vaccines for administration to humans to prevent contraction of HIV in~ection or disease including AIDS, or for administration to humans afflicted with ~IV infection or disease including AIDS.
In a preferred embodiment, the conjugate of lo the invention has the general structure:
j(PEP-A-)-MIEP

or pharmaceutically acceptable salts thereof, wherein:

PEI is an HIV PND peptide, or a peptide capable o~
raising mammalian immune responses which recognize HIV PNDs;
MIEP is an immunogenic protein of the outer membrane protein complex ~OMPC) of Neisseria meningitidis b either ~ecombinatly produced or purfied from ~MPC;
-A- is a covalent linkage, preferably a bigeneric spacer;
i is the percentage by mass of peptide in the : coconjugate, and ic preferably bet~een 1% and 50% of the total protei~ mass in the conjugate.
The conjugate of the lnvention may be prepared by any of the common methods known in the art for preparation of peptide-protien conjugates, such as, ~or example, the bi~eneric chemistry disclo~ed in U.S. patent 4,695,624 and Marburg et al.

-~ 3 ~

105/G~B25 - 36 - 181S9IA

J.A.C.S. 108, 5282 (1986), and in Applications USSN
362,179; 55,558; 55~,974; 555,966 and 555,339. In a preferred embodiment, a process that utilizes the available nucleophilic functionalities, ~ound in proteins, such as the amino group of lysine, the imidazole group of histidine, or the hydroxyl groups of serine, threonine, or tyrosine is used. In practical terms, the number of available protien necleophilic sites may be determined by an appropriate as~ay which may comprise thiolation with N-acetyl homocysteine thiolactone, ~ollowed by Ellman Assay [Ellman, G.L., Arch. B~ochem. Biophys., 82, 70 (1959)] for determination of total free sulfydryl groups and/or by alkylation with a bromoacetyl amino acid, assayable by amino acid analysis.
The preferr~d process can be carIied out in several ways in which the sequence, method of activation, and reaction of protein and peptide groups can be varied. The process may compri~e the ~tep~ of:
Proces~_l:
la. reacting the protein nucleophilic groups with a reagent, for example with N-acetyl homocysteine thiolactone, which generate~ thiol 5 groups on the protein; and lb. reacting the product of ætep la. with peptides previously deriva~ized ~o as to append an electrophilic group prefera~ly comprising moleimide, on the peptide. A preferred embodiment of this invention, which may be prepared according to this process, has the structure:

~'J
lOS/GHB25 - 37 - 18159IA

MIEP -(NH-C-R-~
O ~ N-R-C-N,-PEP)~
R O O H

or pharmaceutically acceptable saltg ~hereof, ~ lo wherein:

:~ : PEP, MIEP, and j, are as defined supra;

: -R- is:
a) -lower al~yl-, b) -substituted lower alky-, : : c~ -cycloalkyl-, -: : d) -substituted cyloalkyl-, e) -phenyl-;
20 _~1 is:
a) -hydrogen, ; :~ b) -lower alkyl, or c) -S03E; and ~ 25 -S- is ~ulfur.
: : Li~ewise, a preferred embodiment of the invention~having the structure:

MIEP-(~DH-C -R : O
O N
O ~ -PEP)~

- ` ' ' ~ 33' 105/G~B25 - 38 - 18159IA

wherein all variables are as defined above, may be prepared by process 2, which comprises the steps of:
~ a. reacting the protein nucleophilic groups wi~h a bifunctional e~ectrophilic reagent, such as maleimidoal~anoic acid hydroxysuccinimide ester, so as to generate an electrophilic protein; and 2b. reacting the product of step 2a. with a peptide containing a nucleophile, such as a thiol group.
A highly preferred embodiment o~process 1, is described in detail below and in Scheme A.
According to the scheme, the immu~ogenic protein is the class II protein of the outer membrane protein.
complex (OMPC) of Neisserla meningi~idis b, either purified from the bacterial membrane or produced by recombinant means. The process comprises the s~eps of:
a.i. reacting MIEP (I), having nucleophilic groups, including free amino groups due to the presence of lysineæ or protein amino-termini, with a thiolatin~
agent, preferably N-acetyl homocy~teine thiolactone, to generate MIEP (II) having "m" moles of ~ulfhydryl groups available for reaction with a thiophile; a.ii.
quantitating the number of available sulfhydryl3 appended to MIEP in step la.i. to determine the value of ~m", preferably by Ellman assay ~Ellman, &.L., Ar~h. Biochem. Biochem. Biophy~., 8~, 70 (19~9)]; and b contactin~ the product of step a. ~ith an exces~, (>m), of an HIV PND ~hich has been previously derivatized so as to append an electrophilic group, preferably w;th a maleimido-al~anoic acid, and moæt preferably with maleimido-propionic acid (this J ~
105/G~B25 - 39 - 18159IA

derivatization is acnieved by N-protecting all amino groups on the peptide that should not be derivatized, and reacting the free peptide amino groups with a bifunctional reagent, preferably maleimidoalkanoyloxysuccinimide, and most preferably maleimidopropionyloxysuccinimide), to generate the conjugate of this invention (III).
The conjugate product may be purified by, for example, dialysis in a buffer having an ionic strength between O.OOlM and l~ and a pH between 4 and 11, and most preferably in an agueous medium having an ionic strength of between 0.01 and O.lM and a pH
of between 6 and 10.

.

.,, , . ~

2 ~

SCHEME A

EP
NH- COC~33 0~ .
II.
Ml EP-( NH C~/S~ m [R J$~ I>m . O

: ::
: ::

~ :

OC~
MIEP~ C--f ~ ~ ~ 0 ~ ~ CH2] 2- s~
:: ~ 3 D R1 4~1 I Nl PDP~ m ~ O

:
.

7 ; :` ' ':

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

:
, 105/G~B25 - 41 - 18159IA

The process described above and depicted in Scheme A may be modified so that MIEP is derivatized so a~ to be covalently linked to a thiophiIe, such as a derivative of maleimide, while the peptide is activated so as to be covalently linked to free sulfhydryls. This and other alternate processes, naturally fall within the scope of this disclosure, including variations on these processes, such ~s variations of sequence of reaction of activated lo species, or ratios of reactants.
The process for making the conjugates of this invention may be applied to making any conjugate wherein a peptide-protein conju~ate is desired and is particularly significant where enhanced immunogenicity of the peptide is required.
The conjugates herein described may be included in compositions containing an inert carrier and are useful when appropriately ~ormulated as a vaccine. This may include prior adsorption onto alum or combination with emulsifieræ or adjuvants known in the art of vaccine formulation. Methods of using the covalent conjugate immunogens of thi~ invention include: (a) uBe as a laboratory tool to characterize -~IV PND peptide structure-function relationships; (b~
use as a~ immunogen to rais ~IV-neutralizing antibodies in a mammal which antibodies may be isolated and administered to a human so as to preven~
infection by HIV, or to limit HIV proliferation post-infection, or to treat humans afflicted by ~IV
in~ection or disease including AIDS. (c) use as a vaccine to immunize humanY against infection by HIV
or to treat humans po~t-infection, or to boost an HIV-neutralizing immune response in a human a~flicted with HIV infection or disea~e including AID~.

, 105/G~B25 - 42 - 18159IA

As a laboratory tool, the conjugate is useful when adminiætered to a mammal in an immunologically effective amount, to generate anti-PND peptide, anti-HIV, or HIV-neutralizing immune responses. The mammal may be boosted with additional conjugate to elevate the immune response.
Antiserum is obtained from such a mammal by bleeding the mammal, centriPuging the blood to separate the cellular component from the serum, and isolating lo antibody proteins from the serum if necessary, according to methods known in the art. Such antiserum or antibody preparations may be used to characterize the efficacy of an HIV PND peptide in a conjugate in raising mammalian anti-PND peptide, anti-HIV, or HIV-neutralizing antibodies in a mammal. ELISA assays using the unconjugated peptide and the antiserum are useful in Yi~ro assays for measurin~ the elicit~tion of anti-peptide antibodies. An in vitro assay for measuring th~
~IV-neutralizing ability of antiserum comprises incubating a preparation of live ~IV with a preparation of the antiserumt then incubating the antiserum-treated ~IV preparation with CD4 receptor bearing cells, and measuring the extent of cellular 2~ protection afforded by the antiserum. These assays and the characteristics of antiserum produced by a given conjugate may be used to study ~he PND peptide stucture-function relationship.
The conjugate is useful for inducing mammalian antibody responses as described in the previous paragraph, and such antibodies may be used to passively immu~ize humans to prevent ~IV
infection, or to limit HIV proliferation post-infection, or to treat humans aiflicted with HIV
infection or di~ease including AIDS.

~ ~ ~,7 105/G~B25 - 43 - 18159IA

The conjugate is useful as a vaccine which may be administered to humans to prevent ~IV
infection or proliferation, or to humans suffering from ~IV disease of HIV infection, including AIDS and related complexes, or to humans testing seropositive for the ~IV virus. The conjugate may be administered in conjunction with other anti-HIV compounds, such as AZT, or more general anti-viral compounds, or in conjunction with other vaccines, antibiotics, or immunomodulators (see Table I below).
The form of the immunogen within the vaccine takes various molecular configurations. A single molecular species of the antigenic conjugate III will often suffice as a use~ul and suitable antigen for the prevention or treatment of ~IV disease including AIDS or ARC. Other antigens in the form of cocktails are also advantageous, and consist of a mixture of conjugateæ that differ by, for example, the mass ratio of peptide to total protein. In addition, the conjugates in a mixture may differ in the amino acid se~uence of the PND.
An immunological vector, carrier or adjuvant may be added as an immunological vehicle according to conventional immunological testin~ or practice.
Adju~ants may or may not be added during the preparation o~ the vaccines of this invention. Alum is t~e typical and preferred adjuvan~ in human vaccines, especially in the form of a thixotropic, viscous, and h~mogeneows aluminum hydroxide gel. For example, one embodiment of the present invention is the prophylactic vaccination of patients with a suspension of alum adjuvant as ~ehicle and a cocktail of conjugates a8 the selected set of immunogens or antigens.

.

105/G~B25 - 44 - 18159IA

The vaccines of this invention may be effectively administered, whether at periods of pre-exposure or post-exposure, in combination with effective amounts of the AIDS antivirals, immuno-modulators, antibiotics, or vaccines of Table I
~source: Market Letter, No~. 30, 1987, p. 26-27;
Genetic En~ineering News, Jan. 1988, Vol. 8, p. 23.]

TABL~ Il A. Antivirals Drug Name Manufacturer Indication AL-721 Ethigen ARC, PGL

BETASERON Triton Biosciences AIDS, ARC, KS
(interferon beta) CARRISYN Carrington Labs ~RC --(polymannoacetate) CYTOVENE Syntex CMV
~ganciclovir) 25 DDC Hoffmann-La Roche AIDS, ARC
~dideoxycytidine) FOSCARNET Astra AB HIV inf, CMV
(triæodium retinitis phosphonoformate) ~PA-23 Rhone-Poulenc Sante ~IV infection ~ J~s~
105/G~B25 - 45 - 18159IA

__________________________________ ___________________ lAbbreviations: AIDS (Acquired Immune Deficiency Syndrome); ARC (AIDS related complex); CMV (Cytomegalo-virus, which causes an opportunistic infection resulting in blindness or death in AIDS patients); ~IV (~uman Immunodeficiency Virus, previously known as LAV, ~TLV-III
or ARV); KS (Kaposi's sarcoma); PCP (Pneumonocystis carinii pneumonia, an opportunistic infection); PGL
(persistent generalized lymphadenopathy).

Drug Name ~anufac~urer Indication ORNIDYL Merrell Dow PCP
(eflornithine) PEPTIDE T Peninsula Labs AIDS
(octapeptide sequence) RETICULOSE Advanced Viral AIDS, ARC
(nucleophospho- Research protein) 25 IR Burroughs Wellcome AIDS, advanced (zidovudine; ARC
AZT) pediatrlc AIDS, gS, asympt ~IV, less severe ~IV, neurological in-volvement.

~ ~ ~S tJ ~,3 ~j r5_~

RIFABUTIN Adria Labs ARC
(ansamycin LM 427) (trimetrexate) Warner-Lambert PCP

UAOOl Ueno Fine Chem AIDS, ARC
Industry VIRAZOLF. Viratek/ICN AIDS, ARC, KS
(ribavirin) ~ELLFERONBurroughs Wellcome KS, EIV, in comb (alfa interferon) with RETROVIR

15 ZOVIRAXBurroughs Wellcome AIDS, ARC, in (acyclovir)comb with RETROVIR

20 B. Immunomodulatoræ
Drug Name ~n~lB~ Indica~iQp ABPP Upjohn Advanced AIDS, KS
(~ropirimine) AMPLIG~N DuPont ARC, PGL
(mismatched RNA~ ~EM Re~earch (Anti-human alpha Advanced Biotherapy AIDS, ARC, KS
3D interferon Concepts antibody) J .f c~

Colony Stimulating Sandoz Genetics AIDS, ARC, HIV, Factor ~GM-CSF) Institute KS

CL246,738 American Cynamid AIDS
~CL246,738) IMREG-l Imreg AIDS, ARC, PGL, KS

10 IMREG-2 Imreg AIDS, ARC, PGL, KS

IMUIHIOL Merieux Institute AIDS, ARC
(diethyl dithio carbamate) IL-2 Cetus AIDS, KS
(interleukin-2) 20 Drug Name ` Ma~ufa~t~ Indication IL-2 Hoffmann-La Roche AIDS, KS
(interleukin-2) Immunex INTRON-A Schering-Plough KS
(intexfersn alfa) ISOPRINOSIN~ Newport ARC, PGL, EIV
(inosine pranobex) Pharmaceuticals seropositive patients (methionine TNI AIDS, ARC
enkephalin) Pharmaceuticals MTP-PE Ciba-Geigy KS
(muramyl-tripep-tide) THYMOPENTIN (TP-5) Ortho HIV infection (thymic compound) Pharmaceuticals ROF~RON ~offmann-La Roche KS
(interferon alfa) (recombinant Ortho severe anemia erythropoietin) Pharmaceuticals as~oc with AIDS
& RETROVIR
therapy TREXAN DuPont AIDS, ARC
(naltrexone) TNF (tumor Genentech ARC, in combination 20 necrosis factor) inter~eron gamma C. Antlbi,Qtics PENTAM 300 LyphoMed PCP
(pentamidine iæethionate~

D. Vaccine~

30 Gag Merck AIDS,ARC

, v 105/G~B25 - 49 - 18159IA

It will be understood that the scope of combinations of the vaccines of this invention with AIDS antivirals, immunomodulators, antibiotics or vaccines is not limited to the list in the above Table, but includes in principle any combination with any pharmaceutical composition useful for the treatment of AIDS. The AIDS or HIV vaccines of this invention include vaccines to be used pre- or post-exposure to prevent or treat HIV infection or disease t and are capable of producing an immune respo~se specific for the immunogen.

The conjugates of this invention, when used as a vaccine, are to be administered in immunologically ef~ective amounts. Dosages of between 1 ~g and 500 ~g of conjugate protein, and preferably between 50 ~g and 300 ~g of conjugate protein are to be ad~inistered to a mammal to induce anti-peptide, anti-HIV, or HIV-neutraliæing immune responses. About two weeks after the initial administration, a booster dose may be administered, and then again whenever serum antibody titers diminiæh. The conjugate should be administered intramuscularly or by any other convenient or efficaciou route, at a concentratio~ of between 10 ~g/ml and 1 mg/ml, and preferably between 50 and 500 ~g/ml, in a volume sufficient to make up the total required for immunological efficacy. The conjugate may be preadsorbed to aluminum hydroxide gel or to the Ribi adjuvant (GB 2220211A, US priority document Z12,919 filed 29/06/1988) and suspended in a sterile physiological ~aline solution prior to injection.

105/G~B25 - 50 - - 18159IA

The protein moiety should behave as an immune enhancer. It is desirable, in the choice of protein, to avoid those that result in non-specific activation of the recipient's immune response (reactogenicity). In U.S. Patent 4,695,624, Marburg et al. used the outer membrane protein complex (OMPC) derived from Neisseria meningitidis to prepare polysaccharide-protein conjugates. OMPC has proven to ~e suitable though other immunogenic proteins may lo be used. The instant invention utiliæes the Class II
major immune enhancing protein (MIEP) of OMPC.
Various methods of purifying OMPC from the gram-negative bacteria have been devieed tFrasch et al., J. E~p. Med. 140, 87 <1974); Frasch et al., J.
Exp. Med. 147, 6~9 (1978)i Zollinger et al., US
Patent 4,707,543 (1987); ~elting ~ al., Acta Path.
Microbiol. Scand. Sect. C. 89, 69 (1981); Helting et al., US Patent 4,271;147]. OMPC may be used herein essentially according to the Helting proces3, from which MIEP may be further purified [Murakami, K., et al., Infection and I~m~nitv. 57, 2318 (1989)3, to provide i~mune enhancement necessary to induce mammalian immune responses to HIV PND peptides. MIEP
may be derived by diæsociation of the isolated OMPC, or alternatively, produced through recombinant e~pr~ssion of the desired immuno~enic portions of OMPC. Methods of preparing and using an OMPC æubunit ~re disclosed in co-pending US application serial Nos. 555,329; 555,978; and 555,204 (Merck Case #'s 18159, 18110, and 18160 respectlvely).

.
, .

~ Y~ f~ 3 105/G~B25 - 51 - 18159IA

The HIV PND peptides that may be used for making species of the conjugate of this invention may be linear or cyclic peptides. The linear peptides may be prepared by known solid phase peptide synthetic chemistry, by recombinant expression of DNA
encoding desireable peptide sequences, or by fragmentation of isolated ~IV proteins. Cyclic ~IV
PND peptides may be prepared by cyclization of linear peptides, for example (a) by oxidizing peptides lo containing at least two cysteines to generate disulfide bonded cycles; (b) by forming an amide bonded cycle; (c) by forming a thioether bonded cycle. Processes for making such peptides are described herein but this description should not be construed as being exhaustive or limiting. The conjugates of this invention are u~eful whe~ever a component peptide is an ~IV PND ~r i~ capable of priming mammalian im~une response3 which recognize HIV PNDs.
PND peptides, both thoæe known in the art and novel compoundæ disclosed herein and separately claimed in co-pending U.S. Application Serial Nos.
555,112 and 555,227, (Merck Caæe Nos. 18149, and 18150) and co~iled U.S. application _ , _ (Merck 2s Case Nos. 1806~IB~ are defined as peptidyl ~equence~
capable of inducing an ~IV~neutralizing immune response in a mammal, including the production of IV-neutralizing antibodies.

: - :

~ ~ P~J ~

A major problem overcome by the instant invention is the ~IV interisolate sequence variability. For example, in the PND which occurs in the third hypervariable region of gpl20 (see below), although certain amino acids have been found to occur at given locations in a great many iæolateæ, no strictly preserved primary sequence motif exists.
This di~ficulty is surmounted by this invention because it allows conjugation of a eocktail o~
peptides having PND sequences from as many different ~IV iæolates as necessary to attain broad protection. Alternatively, a broadly protective cocktail Qf conjugates may be prepared by mixing conjugates, each of which is prepared separately with a peptide moiety providing protection against a single or several ~IV isolates.
The amino acids found near or between amino acids 296 and 341 of gpl20 have been shown to meet the criteria which define a PN~. In the IIIB iæolate of HIV, a 41-amino-acid sequence has been reported as follows (SEQ ID~
-Ile Asn Cys Thr Arg Pro Asn Asn Asn Thr Arg Lys Ser Ile Arg Ile Gln Arg Gly Pro Gly Arg Ala Phe Val Thr Ile Gly Lys Ile Gly Asn Met Arg Gln Ala ~iæ Cys Asn Ile Ser-, with the two cysteines being disulfide bonded to each other to form a loop. The trimer -Gly Pro Gly- iæ exposed at the PND loop tip. Peptides from di~ferent HIV isolates from this same region of gpl20 raise isolate-specific neutralizing antibodies 30 when presented to goat and guinea pig immllne systems as conjugate with keyhole-limpet hemocyanin. The major neutralizing epitope within the 41-mer ~3~J~1, , sequence, presented above, is comprised by the eight amino acids æurroundi~g and including the -Gly Pro Gly- trimer ~Javaherian et al., ~NAS USA 86, 6768 (1989)]. In Table II below a number of linear peptides of different length and composition that can be used to prepare the conjugates of this invention are presented. The name o~ the isolate containing a peptide having the sequence of the tabulated peptide is given, along with a name herein ascribed to that lo peptide for ease of reference. The letter r- on the left hand side of each peptide represents the possibility of linking the peptide to an immunogenic protein, such as the MIEP at that posi~ion. In addition, marker amino acids, such as norleucine and ornithine may form part of r-.
TABLE II
LINEAR ~IV P~ e~LDES

HIV SEQ ID
20 Isolate Peptid~ ~equence ~ame N0:
MN r~Tyr Asn Lys Arg Lys Arg PND142 2 Ile His Ile Gly Pro Gly Arg Ala Phe Tyr Thr Thr Lys Asn Ile Ile Gly Thr SC r-Asn Asn Thr Thr Arg Ser PND-SC 3 Ile His Ile Gly Pro Gly Arg Ala PheTyr Ala Thr Gly Asp Ile Ile &ly Asp Ile 105/GHB25 - 54 ~ IA

IIIB r-Asn Asn Thr Arg Lys Ser Ile PND135 4 Arg Ile Gln Arg Gly Pro Gly Arg Ala Phe Val Thr Ile Gly Lys Ile Gly Asn S

IIIB r-Arg Ile Gln Arg Gly Pro Gly PND135-18 5 Arg Ala Phe Val Thr Ile Gly Lys Ile Gly Asn lO IIIB r-Arg Ile Gln Arg Gly Pro Gly PND135-12 6 Arg Phe Val Thr MN r-His Ile Gly Pro Gly Arg Ala PND-MN8 7 Phe 1$ MN r-Gly Pro Gly Arg Ala Phe PND-MN6 LAV-l r-Ile Gln Arg Gly Pro Gly Arg PND-LAV-l 9 Ala Phe 20 SF2 r-Ile Tyr Ile Gly Pro Gly Arg PND-SF2 lO
Ala Phe NY5 r-Ile Ala Ile Gly Pro Gly Arg PND-NY5 11 Thr Leu 2~
CDC4 r-Val Thr Leu Gly Pro Gly Arg PND-CDC4 12 Val Trp RF r-Ile Thr Lys Gly Pro Gly Arg PND-RF 13 Val Ile : , - ~ . .

, .

~ ~? ~
105lGHB25 - 55 18159IA

ELI r-Thr Pro Ile Gly Leu Gly Gln PND-ELI 14 Ser Leu Z6 r-Thr Pro Ile Gly Leu Gly Gln PND-Z6 15 Ala Leu MAL r-Ile His Phe Gly Pro Gly Gln PND-MAL 16 Ala Leu lD Z3 r-Ile Arg Ile Gly Pro Gly Lys PND-Z3 17 Val Phe This liæt is n~t an exhaustive list of possible PND sequences. Rather, i~ i~ provided as a suggestive and illustrative guide as to useful PND
primary seguences. Therefore, peptides conjugated as ; herein described to form the immunogen of this invention are any of the tabulated peptides or immunologically-equivalent variants on the theme sugge~ted by these peptidyl sequences. The nature of the variations is considered next.
The primary sequence of this HIV PND appears to ha~e a conserved core amino acid sequence, comprised by the tetramer sequence -Gly Pro Gly Arg-, (SEQ ID: 18:~ rJith: increaciD~ divergence on either side of thi~ sequence among ~IV isolates. Some isolates have ~equence that diverge even within the tetramer, having -Gly Pro Gly Lys- (SEQ ID: 19:), -GIy Pro~Gly Gln- (S~Q ID :20::), and even -Gly Leu Gly Gln- (SEQ ID: 21) core sequences. All of these possible sequences come within the scope of this disclosure as bei~g ~eptide ~equences that are advantageous ~or conjugation according to this invention.

, The length of the peptide i~ a significant factor in promoting cross reactive immune responses.
That is, an immune response raised against a given peptidyl epitope may recognize similar epitopes from the same or different EIV isolate based on the number of amino acids in the epitope over and above the critical neutralizing epitope. In addition, the length of the peptide is also responsible for determining the probability of exposure to the immune lo system of the determinant responsible for generating an HIV-neutralizing response.
In order to maximize the probability of relevant epitope preRentation, chemistry was developed whereby the PND peptides may be locked into a given three-dimensional configuration. It is known that the 41-amino-acid PND o~ the HIV IIIB isolate, represented above, is configured as a loop by the presence of the cysteine-to-cysteine disulfide bond.
Diæulfides, however, may be labile under certain conditions and therefore may allow the loop to open and the peptide to exi~t in a linear form.
Therefore, in addition to linear peptides, disulfide-bonded cyclic peptides and novel ~IV PND
peptides havin~ nonlabile cyclic structures disclosed herein but ~eparately claimed as free peptides in : co-pending US application æerial NOR, _,_ (co-filed Merck case 18068IB); 555,112 and 555,227, may all be utilized as the P~P component in the formation of the conjugates of this invention.
The peptides that may be used in formation of these conjugates may be derived as fragments of natural prsteins (gpl20 for example), by recombinant ~;

105/G~B25 - 57 - 18 expression of portions thereof, or by chemical synthesis according to known methods in the art. In addition, novel cyclic PNDs may be prepared synthetically according to the processes herein described. The sequences may contain both natural L-amino acids, or unusual or D-amino acids. In addition, the conjugation chemistry is sufficiently flexible so that the appropriate choice of peptide derivatization reagents allows for successful lo conjugation, Synthetic peptides have been prepared by a number o~ strategies conducted either in solution or on solid support~. Excellent texts covering the basic principles and techniques are: Principles of Peptide Synthesis, Bodansæky. M., Springer-Verlag (1984); Soli~ Phase Peptide Syn~hesis, Stewart J.
M., Young, J. D., Pierce Chemical Çompany (2nd. ed.
1984); and The Peptides, Gross, E., Meienhofer, J., Academic Press, Inc., (1979). The processes described herein, however, are not limited to the disclosure of these texts.
Synthetic cyclic peptides may be prepared in two phases. First, the linear peptide may be synthesized on Milligen 9050 peptide or an A$I 431A
synthesizer using 9-fluorenylmethyloxy-carbonyl (Fmoc) chemistry and side-chain-protected Fmoc-amino acid pentafluorophenyl ester~ which are known reagents or using derivatized Wang resin, Fmoc chemistry, and ~ide-chain protected Fmoc-amino acid symmetrical anhyd rid es, prepared in situ, as rea~ents.

:

105/G~B25 - 58 - 18159IA

Second, the linear peptide may be cyclized, either in solution or with the peptide ~till attached to the solid phase resin. Cyclization may be accomplished by any technigue known in the art, which may comprise, for example: a) incorporating cysteine residues into the linear peptide on either end of the sequence which is to form the loop and allowing disulfide bond formation under oxidizing conditiona known in the art; b) preparing a cyæteine containing peptide as in (a) but retaining the cysteines as free sulfhydryls (or as Acm protected thiols ~hich are deprotected to the free sulfhydryls) and trea~ing the peptide with o-xylylene dibromide or similar reagent, such as the diiodide, dichloride, or a dihalogenated straight or branched chain lower alkyl having between two and eight carbon atoms; such reagents react with the sulfur atoms of the cysteines to form a cyclic structuxe containing two nonlabile thioether bonds to the benzene or the alkyl; c) allowing a free group 2~ on one æide of the loop amino acid~ to become amide bonded to a f ree carboxyl group on the o~her ~ide of the loop amino acids through DPPA, BOP, or æimilar reagent mediated peptide bond formation. Each of these strategies i~ taken up in more detail below, after presentation of a generalized description of the cyclic peptide~ produced by theæe methods.
Thus, without limiting the conjugate invention to the following peptides or methods of producing them, the PND peptides which may be conjugated after removal of appropriate protecting groups aæ nece~sary, according to this invention include those represented by the structure PEP, which includes the linear peptides of Table II above and cyclic peptides:

105/G~IB25 - 59 18159IA

~Pr o -Gly Rl ~ R8 1~ ~ /2 Y h3-R4-R5 R6_______________R7 wherein:
r is the position of linkage ~etween PEP and MIEP, optionally comprising a marker 1~ amino acid, if Rl i5 not a marker amino acid;
.
Rl is:
a) a bond, or b) a peptide of 1 to 5 amino acids, optionally including a marker ami~o acid which migrates at a positio~ in the amino acid analysis spectrum which - -is isolated ~rom the ~ignal of the 20 naturally occuring amino acids;
- preferably norleucine,:gamma aminobutyric acid, ~-alanine, or ornithine;~

2s ~2 i~ ~
~: a) either a bond or a peptide of up to 17 amino acids if R3 i s ~a peptide O:e at :: 1 east 2 amino acids, or~ :
b) a pe~tide o~ between 2 to 17 amino acids, if R3 is a bond;

.
:

'' :

R3 is:
a) either a bond or a pep~ide of up to 17 amino acids if R2 is a peptide of at least 2 amino aclds, or b) a peptide of between 2 to 17 amino acids, if R2 is a bond;

R4 is:
a) -NH-~H-CO-, with R7 bonded to the methine carbon, if R7 is R8, or b) a bond from R3 to R7 and R5, if R7 is carbonyl or -COCH2CH2CH~CONE2)N~CO-;

R5 is:
a) a peptide of one to five amino acids, optionally including a marker amino acid, b) -OH, c) -COOH, d) -CON~2~
e) -NH2~ or f) -absent;

R~ is:
` a~ an amino acid side rhain, selected from the æide chain of any of the common L
or ~ amino acids, (see table of~
Definitios and Abbre~iations), if the optional bond (~ -- ) to R7 i~
absent, b) -R8-S-S-, or -R8-S-R~-R9-R~-S-, if R7 is R8, or .

p~
105iGHB25 - 61 - 18159IA
c~ ~8_NH_ i~ R7 is -C-O, OR -C-CH2-C~2-\~-NH-C=O;
CONH~
R7 is:
a) -R8_, b) -C=O, or c) -I~-CH2-CH2-CH-NH-C=O;
~OM~2 R8 is a bond or lower alkyl of between one and eight carbons;

R9 is:
a) R10 or b) xylylene ~10 is:
a) lower alkyl, or b) -~2--~2-; and every occurrence of a variable is independent of every other occurrence of the same variable. When a peptide haæ been synthesized with a protected amino terminal amino acid, the amino termi~al pro~ecting group:such as benzylo~y carbonyl (Z) for protecting amines, or acetamidomethyl (Acm~: for protecting sulfhydryls, may be removed~according to methods known in the art and exemplified herein. The deprotected group thus revealed may be utilized in co~alent bond formation, through the linker r, ~o the immunogenic protein.

' ~ .
. " . '~

n ~, .v 105/G~B25 - 62 - 18159IA

Hereinafter, amino acids -R2-Gly Pro Gly Arg-R3-, which form the "core" of the PND peptides, and go toward formation of the loop of a cyclic peptide, will be referred to as loop or core amino acids. When the optional bond between R6 and R7 i~
ab~ent however, the structure, P~P, is linear, and encompasses all of the linear peptides of Table II.
Whether the peptide is linear or cyclic, the amino acid sequences comprising R2 and R3 of PEP may be any combination of amino acids, including sequences surrounding the core -Gly Pro Gly Arg-tetramer in any of the sequences of Table II. Thus, the core amino acids represented by -R2-Gly Pro Gly Arg-R3- may be further defined as having the core 15 amino acid structure:
-xnxlX2-Gly Pro Gly Arg~X3~4~m~

wherein:

Xl is a constituent of R2 selected from:
a) serine, b) proline, c) arginine, d) histidine, e) glutaminej or f) threonine;

X2 is a constituent of R2 selected from:
a) isoleucine, b) arginine, c) valine, or d) methio~ine;

~ ~ r.~ ~ _, ,., r,, Xn is is a constituent ~f R2 and is either a bond or a peptide of up to 15 amino acids;

X3 is a constituent of R3 selected from:
a) alanine~
b) arginine, or c) valine;

X4 is a constituent o~ ~3 and is selected from:
lo a) phenylalanine, b) isoleucine, c) valine, or d) leucine;
Xm is a constituent of R3 and is a bond or a peptide f up to 15 amino acids.

The cyclic peptides may be disulfide bonded structureæ OI a cycle formed through a nonlabile bond or structure. The term "nonlabile bond" means a covalent linXage, other than a disulfide bond.
Examples of such nonlabile bonds are amide and thioether bonds as disclosed in co-pending applications USSN 555,112 and 555,227. These covalent linkages may be through a bridge structure, 2~ ~ueh as xylylene, through a lower alkyl, through -C~2-0-C~2, or through an ami~o acid amide bonded bridge. By altering the bridge structure and/or the number and eombination of aminQ aeids included in the peptide, the con~ormation of the loop structure o~
the cycle may be optimized, allowing for fine-tuning of the PND epitope presented to ~he immune ~ys~em.
: For example, use of an o xylylene bridge generate~ a ; ~ . . ~ ' ' , ' ' ' .~ ,, J, - ,~
105/G~B25 - 64 - 18159IA

"tighter" loop structure than when, for example, an eight carbon strai~ht chain lower alkyl i8 used as the bridge. Thus, the conjugates of this invention are useul both as reagentæ to analyze the structure-~unction relationship of the PND epitope in raising anti-peptide, anti-HIV, HIV-neutralizing, and anti-AIDS immune reæponses in mammals, and as components for formulation of anti-~IV disease, including AIDS, vacclnes.
lo Synthetic products obtained may be characterized by fast-atom-bombardment mass spectrometry [FAB-MS~, reverse phase ~PLC, amlno acid analysis, or nuclear magnetic resonance spectroscopy (NMR) .

a. Cyclic Peptides through Disulfide-Bonded ~vsteines:
Peptides containing cysteine residues on either side of the loop amino acids may be cyclized under oxidizing conditions to the d;sulfide-bonded cycles. Method~ for achievin~ disulfide bonding are known in the art. An example of disul~ide bonded peptides uæeful in this invention is given infra in ~xample 10, wherein cPND4 is produced and Example 18 wherein cPND33 is produced. In Example lO, a process utilizing the Acm derivative o~ cysteine to generate disulfide bonded cPNDs is used, ~ut other processes are equally applicable. In ~xample 18, the peptide containg two sulfhydryls is oxidized in dilute acid.
The di~ulfide bonded peptides are preferred in the instant învention.

105/G~B25 - 65 - 18159IA

Thus, in a preferred embodiment of this invention, the peptide has the structure (SEQ ID:
18:~:

Pro - Gly H H oGly Arg H H O
r-Rl-N-C-C-RZ R3-N-C-C-R5 RB ~___-R~
or pharmaceutically acceptable saltY thereof, wherein:

r is:
a) hydrogen, b~
o -co o :

wherein W is preferably -~CH2)2~ or -(CE2)3- or R6, where R6 i~
-. ~ or ~ 'J~ i fr ~"~
105/G~B25 - 66 - 18159IA

wherein R7 is lower alkyl, lower alkoxy, or halo;

Rl is:
a) a bond, or b) a peptide o~ 1 to 5 amino acids, optionally including a marker amino acid;

R2 is : a peptide of 3 to 10 amino acids R3 is: a peptide of 3 to 10 amino acids R5 i3:
a~ -0~, b~ a peptide of 1 to 5 amino acids, optionally including a marker amino acid, or c) -N~2;

R8 is lower alkyl of between one and eight carbons.

Lower alkyl con~ists of straight or branched chain al~yl~ having ~rom one to eight carbons unless otherwi~e specified. ~ereinaftes; amino acids _R2 Gly Pro Gly Arg-R3-, which go t~ward formation of the loop of a cyclic peptide, will be re~erred to as loop amino acid~.
In o~e embodime~t o~ the invention, t~e 0 cyclic peptide haYing the structure ~SEQ ID: 18:~:

~ '' .f't~ ~
105/G~B25 - 67 - 18159IA

Pro --Gly H H 0 Gly Arg H H 0 H-Nle-N-C-C-X X1X2 X3X4Xm-N-C-C-R5 R8. ~ _ R9 /

i~ prepared by cyclizing a linear peptide having the structure:

Pro--Cly H H O Cl y Ar g H H O
H- Nl ~ - N- C - C- XnX1 X2 X3 X~ Xm N- C- C- }~5 I g wherein:
Xl is a constituent of R2 selected from:
a~ serine, b) proline, c) arginine, d) histidinc, : ~ 25 e) glutamine, which i~ preferr~d, or f) threonine;
:
X2 i8 a con~tituent o~ R2 selected from:
a) isoleuc;ne, which is most preferred, 0 b) arginine, which is preferred, c) valine~ or d) methionine;

105/GHB25 - 68 - -181~gIA

Xn is a constituent of R2 and is an amino acid or a peptide of up to 8 amino acids;

X3 is a constituent of R3 selected from:
a) alanine~
b) arginine, or c) valine;

X4 is a constituent of R3 and is selected ~rom:
lo a) phenylalanine, b) isoleucine, c) vali~e, or d) leucine;

Xm is a constituent of R3 and is an amino acid or a peptide of up to 8 amino acids.

~2 is preferably Isoleucine.

The novel disulfide bonded cyclic peptides used in this invention (and separately claimed in co-filed Merck cage 18068IB, USSN _ , _ ) may be - prepared in essentially two phases: First the linear peptide is æynthesized on a Milligen 9050 or an 2S ABI-431A peptide ~ynthesizer using 9-fluorenyl-methyloxycarbonyl (Fmoc) chemistry and appropriately slde-chain protccted Fmoc-amino acid : pentafluoro-phenyl esters as reagents or using derivatiz~d Wang resin, Fmoc chemistry, and ~ide-chain protected Fmoc-amino acid symmetrical anhydrides, prepared in situ, as reagents.

} ~ ~; 7~
105/G~B25 - 69 - 1~159IA

Second, the linear peptide is cyclized, either in solution or with the peptide still attached to the solid phase resin by incorporating cysteine residues into the linear peptide at either end of the sequence which is to form the loop, and oxidizing these to the disulfide. In a preferred embodiment, cyclization is accomplished by exposure of the peptide to ~a) ~22~ (b) atmospheric oxygen, (c) aqueous C~3CN containing about 0.1 - 0.5% TFA, or (d) lo about O.lM ferricyanide. The preferred method is exposure to atmospheric oxygen.
Products obtained may be characterized by faæt atom bombardment-mass spectrometry [FAB-MS], reverse phase ~PLC, amino acid analysis or nuclear magnetic resonance spectroscopy (NMR).
Thus, the peptides useful in this inven~ion may be prepared as further described below in (i) and (ii ):

i. Peptide Cvclization in the Solid S~ate: A linear peptide containing Cl and c2 on either side of the loop amino acids, where Cl and c2 are both cysteine or another amino acid containing free sulfhydryl groups in the side chai~, is prepared according to known synthetic procedures (~ee discussion supra).
In the completed cyclic PND, the sul~hydryl containing side chains, (~RB-S~), go to~ard making up the -R8-S-groupæ of the completed cyclic HIV PND structure show~ above. Amino acids to be incorporated which have reactive: side chains (R groups) are used in an appropriately R-group protected form. For example, hiætidine i~ triphenylmethyl (Trt), or Boc protected, and arginine is 4-methoxy-2,3,6-trimethylphenyl sulfonyl (Mtr) protected.

. .. .

$<~
105/G~B25 - 70 - lB159IA

Preferably, a resin is purchased with c2 in its Acm protected form already attached to the resin, for example, Fmoc-L-Cys(Acm)-O-Wang resin. The cysteine incorporated at the amino terminal side of the loop amino acids, Cl, may also be the Acm derivative. Either Cl or c2 may be bound to addi~ional amino acids, R~ or R~ respectively, which may be utilized in the formation of conjugates with carrier molecules or may serve as marker amino acids for subsequent amino acid analysis, such as when norleucine or ornithine is used.
The sulfur of the acetamidomethylated cysteines are reacted, at room temperature for about 15 hours in a solvent compatible with the resin, as a 1-50% concentration of an organic acid, preferably about 10% acetic acid in anhydrous dimethylformamide (DMF), with about a four fold molar excess of a hea~y metal salt, ~uch as mercuric acetate [~g(OAc)2~ for each Acm group. The resulting heavy metal thioether, for example the mercuric acetate thioether of the peptide, PEP(S-HgOAc), is then washed and dried.
Addition of excess hydrogen sulfide in ~ME yields insoluble metal sulfide, e.g. mercuric sulfide ~gS), and the peptide with free ~ulfhydryl groups. The free æulfhydryls are then oxidized by one of the aforementioned methodæ. Alternatively, the Acm protected thiols may be converted directly to the cyclic disulfide by treatment ~ith iodine in a methanol/DMF solvent.

ii. Cyclization Q~ Peptides in Solutisn:
Essentially the same process described abo~e ~ j~ rJ ~ ~ ~. r~

105/G~B25 - 71 - 18159Ih for solid state cyclization applieæ with two main variants: I~ the peptide is cleaved (95% TFA/4a/o ethanedithiol/1% thioanisole) from a pepsyn KA resin, acid labile side chain protecting group~ are also removed, including Cys(Trt) which provides the necessary free -SH function. If however, Cys(Acm) protection is used, then mercuric acetate/hydrogen sulfide cleavage to the free -S~ group is reguired as an independent procedure, with the linear peptide lo either on or off the resin.
One method however, i8 the use of Cys(Acm) protection and Sasrin or Pepsyn K~ resin, and cleavage of the linear, fully protected peptide from the resin with 1/~ TFA/CH2C12. Mercuric acetatel hydrogen sulphide then selectively converts Cys(Acm) to the free -SH group, and cyclization is effected on the other~ise protected peptide. At this point, the peptide may be maleimidated in situ , selectively on the N-terminus. Acid labile side chain protecting groups are c1eaved with 98% TFA/2% thioanisole, and the cyclic peptide is isolated by HPLC. The preferred method, however, is to cleave the peptide from the resin, and allow cyclization by one of the aforementioned methods. The most preferred method is to allow air oxidation for about one to fifty hours of between 1~ and 40C.

Thus, in a particularly prefexred embodiment of this invention, a peptide (CPND 33) ha~ing the ætructure (SEQ ID: 22:):

~ 'S~ '~t~,J

H-Nle Cys Tyr Asn Lys Arg Lys Arg Ile His Ile Gly Pro ¦ Gly Arg Ala Phe Tyr Thr Thr Lys Asn ~CH2 Ile Ile Gly Cys-OH
S S-C~2 is conju~ated to MIEP through either the amino terminal Nle or one of the internal lysines to generate one or a mixture of all of the ætructures:

.

~ . ~

lOS/G~B25 - 73 - 18159IA

o-1) 1 o ~[ ~F~cn~ C.I.. ~ Jn L~ ~rg LyJ ~rg 21~ sly n;

~o~Cy~ O~y 11-- Il- n L~ Thr Sh~ Tyr n~ Ar9 ..
l 5 ~-z) H D.~.Ar9 L~ Ar~ Oly Pro 011 ~r~ Al- ~h-~3.- . ~j~c ~ c~ \ 8 o, 11 ¦ T~r ~S ~Cl~ -C-Y-tCH~ . 1 3Jn Srr ~ C1~ Aon L~ Thr Thr : 25 : : ~

:

': . , :
.

..
, , 13 0 / GHB -- 7 4 -- 1 g 1~ 92 r~ 3 e-3) H O.C-~rg Tl- H~- Il- Ol~ Pro OIJ arg ~ Ph- ~yr ~hr tn~ H-Cj~CHI C*.c4-o \ J~ O H ¦ mr O IIHCOC4 l~--tcb)~-c~ tc*)~-c~ r9 Ly~ ~n Sy~ C~ -C-C~ olg 11- ~ ~n L~ . os o H 11 Ill- COOH

~ o c ~ n 11~ Oly C~ -Cy- Srr ~n L7~ Jlr~ Ly-Sl-Cj,C~CH,-C*~ O H ¦ ~rg l c 4 ~--tc*j~-c-N-tc4~-C-~hr mr Slrr Ph al- llrg 011 Pro Olr 11- H'-~: :
~ 20 ~: :

: ~ : 25 , ~ : :
; 30 ~: :
:: :

~: :

' 2 ~
130/G~B ~ 75 - 18159IA

wherein ~ is the percentage by mass of peptide in the conjugate, and is preferably between 11 and 50% of the total protein mass in the conjugate.

b. Cyclic Peptides through Thioether Linkage to o-Xvlylene or Lower Alkyls:

i. Peptide Cyclization in the Solid State: A linear peptide containing Cl and c2 on either side o~ the lo loop amino acids, where Cl and c2 are both cysteine or another sulfhydryl containing amino acid, is prepared according to ~nown synthetic procedures t ee discussion supra). In the completed cylic PN~, Cl and c2 become part of the R6 and R7 groups of the PEP
structure sho~n above. Amino acids to ~e incorporated which have reac~ive side chains (R
groups) are used in an appropriately R-group-protected form. For e~ample, histidine is triphenylmethyl- (Trt) protected, arginine may be 4-methoxy-~,3,6-trimethylphenyl sulfonyl (Mtr) protected. [Principle~ of Pçptide Syn~hesis, Bodanszky. M., Springer-Verlag ~1984); $ol-id Phase Peptide Synthesis, Stewart J. M.i Young, J. D., ~ Pierce Chemical 50mpany (2nd ed. 1984); and Th~
Peptides, Gross, E., Meienhofer, J., Academic Press~
Inc., (1979)~.
Preferably,~ a resin is purchased with C~ in its Acm-protected form already attached to the resin, ~or example, Fmoc-L-Cys(Acm)-0-Wang resin. The cysteine incorporated at the amino terminal side of the loop amino acids, Cl, may also be the Acm derivati~e. Either Cl or c2 may be bound to additional amino acids, Rl or RS respectively, which .

may be utilized in the formation of conjugates with carrier molecules or may serve as marker amino acids for subsequent amino acid analysis, such as when norleucine or ornithine is used.
The sulfur of the acetamidomethylated cysteines is reacted, at room temperature for about 15 hours in a ~olvent compatible with the resin, such as 10% acetic acid in anhydrous dimethylformamide (DMF), with about a four-fold molar excess o~ a heavy lo metal salt, such as mercuric acetate [Hg(OAc)2] for each Acm group. The resulting heavy metal thioether, for example the mercuric acetate thioether of the peptide, PEP(S-HgOAc~, is then washed and dried.
Addition of excess hydrogen sulfide in DMF yields insoluble metal sulfide, e.g., mercuric æulfide (HgS), and the peptide with free sul~hydryl groups.
A mixture of about an equimolar amount, as comp~red with peptide, of o-xylylene dibromide or dichloride, a dibrominated or dichlorinated lower alkyl, 1,3-dihalogenenated -C~-O-CH-, or similar reagent which will provide a desirable bridge leng~h, is added to the derivatized resin. A large excess of tertiary amine, preferably triethylamine (NEt3) in DMF is added slowly. The reaction with the bis-sulfhydryl peptide moiety is allowed to proceed for about sixtee~ hours at room tempera~ure, yielding the bridge group derivatized cyclic peptide bound to resin. Deprotection of acid sensitive side chain protecting groups and cleavage from the resin is achieved by treatment with 95% trifluoroacetic acid (TFA) in the pre~ence of 4V/~ 1,2-ethanedithiol and 1%
thioanisole. The dissolved cyclic peptide may then ' r~

be separated from the resin by filtration. The filtrate is evaporated and the crude residual product is purified by high performance liquid chromatography (EPLC) according to known methods, for example by reverse phase HPLC.

ii. Cycliza~ion of Peptides in Solution:
Essentially the ~ame process described above for solid state cyclization applies with ~wo main lo variants: If the peptide is cleaved ~95% TFA/4%
ethanedithiol/1% ~hioanisole) from a pepsyn KA resin, acid labile side chain protecting groups are also removed, including Cys(Trt) which provides the necessary ~ree -S~ function. If however, Cys(Acm) protection is used, then mercuric acetate/hydrogen sulfide cleavage to the free -SH group is required as an independent procedure, with the linear peptide either on or off the resin.
The preferred method ho~cver, is the use of Cys(Acm) protection and Sasrin or Pepsyn KH resin, and cleavage of the linear, fully protected peptide from the resin with 1% TFA/CH2C12.. Mercuric acetate/hydrogen sulphide then æelectively converts Cys(Acm) to the ~ree -SH group, and cyclization is e~fected on the otherwise protected peptide. Acid labile ~ide chain prstecti~g groups are cleaved with 95% TFA/4% ethanedithiolll% thioanisole, and the cyclic peptide is isolated by HPLÇ.
Removal of excess reagents, such as unreacted xylylene dibromide, prior to acid cleavage of the protecting groups is conveniently achieved by, for example, a step gradient rever~e pha~e HPLC run prior to more selective gradient elut~on.

130/GHB - 78 - 1815~IR

Cyclic HIV PND peptides prepared according to the method of this subsection include, but are not limited to, the sample cPNDs represented below. The methods of this subsection are generally applicable to small peptides, and particularly applicable to peptides of between 5 and 30 amino acids. An optimal ring size may include between 5 and 10 amino acids, including the -Gly-Pro-Gly- trimer, and this ring size i~ easily maintained by production of cycles from linear peptides ha~lng the appropriate number and combination of amino acids.
Representative peptides resulting from the process described in this subsection k. parts (i).
and (ii) are disclosed inapplication U.S.S.N.
555,227. The conjugate invention ~hould, however, not be construed as being limited to use those particular embodiments of HIV cyclic PND peptides.
Other linear ~IV PND peptide sequences may be cycli ed in essentially the same fashion used to provide those peptides. Series of peptides having divergent primary sequences could be generated and would be beneficial in ~his invention as long as they continue to elicit an anti-peptide, anti-HIV, or EIV-neutralizing immune response.

c. CyclizatiQn through Amide Bond FQrmation;
Novel amide bonded cyclic ~IV PND peptides may be prepared for conjugation in essentially two phases: First, the linear peptide is prepared, for example on an ABI-431A peptide synthesizer, by known solid phase peptide synthetic chemistry, for example using Fmoc chemi~try and appropriately ~ide-chain proteeted Fmoc-amino acids as reagents.

., Second, the linear peptide is cleaved from the resin and cyclized in solution by allowing the free amino terminus of the peptide, the free amino group of an amino terminal isoglutamine, or a free ~-amino or a-amino group of a lysine on one side of the loop amino acids to be amide bonded to a free carboxyl group on the carboxy-terminal side of the loop amino acids through DPPA, BOP, or similar reagent mediated peptide bond formation.
Products obtained may be characterized by fast atom bombardment-mass spectrometry [FAB-MS], reverse phase ~PLC, amino acid analysis, or nuclear magnetic resonance spectroscopy (MMR).
Thus, highly preferred embodiments of this invention are conjugates having covalent linkages from MIEP to an amide bonded cyclic HIV PND, prepared as de~cribed hereinabo~e. Where the PND is from a predominant isolate, such as the ~IV IIIB or the ~IV
MN isolate, a conjugate vaccine, or a mixture of such conjugate vaccines is highly ad~antageous for prophylaxis or treatment of AIDS or ARC. ~xamples o~
such preferred embodiments having the structure:

s ~ ~ ~

130/GHB - 80 ~ 18159IA

~ r~
Ml~ N-C-Cl~-CH~-C}~ O I I H O
O NHCOCH3 l ,,N-C~C~-C-N-Nl~-N~CH~)~-C-C- Hl~-Ile-Gly-Pro-~:ly-~g-Alll-Ph~
~ O H-N--C~ (CH~)~ c7 N-C-o b) ~ H
~EP-- I~-C-C~C~-cq~-s O I I H O
L NHOOC~ N-CN~CH~-C-N-Nl~-~C-C-Gln-Arg-Cly-Pro-Cly-Arg ( CH~)s, Ala ( ~) N--C ~?he _ ~0 or pharmaceutically acceptable ~al~s thereof, wherein:

j is the percentage by mass of peptide in the conjugate, and is preferably between 1% and 50%
of the total protein mass in the conjugate;

are useful for inducing anti-peptide immune responses in mammal~, ~or inducing ~IV-neutralizing antibodies in mammals, for formulating ~accines to prevent : ~IV-di~eaæe or infection, or ~or treating human~
afflicted with ~IV-disease or infection, including AIDS and ARC.

, J ~t "'~

130/G~B - 81 - 1~159IA

One or more of the conju~ate vaccines of this invention may be used in mammalian species for either active or passive protec~ion prophylactically or ~herapeutically against infectious agents such as, in the preferred embodiment of this invention, Haemophilus influenzae serotype B, or human immunodeficiency virus induced diseases. Active protection may be accomplished by injecting an effective quantity capable of producing measurable amounts of antibodies (e.g., about 1 microgram to about 50 ~g, depending on the antigen) of an antigen (e.g. PRP, HIV PND peptides) in the MI$P-conjugate form of each of the conjugates being administered per dose. The use of an adjuvant (e g., alum) is al30 intended to be within the ~cope of this invention.
Passive protection may be accomplished by in~ecting whole antiserum obtained from animals previously dosed with the MIEP-conjugate or conjugate~, or globulin or other antibody-eontaining fractions of said antisera, with or without a pharmaceutically-acceptable carrier, ~uch aæ sterile saline solution. Such globulin is obtained from whole antiserum by chromatography, salt or alcohol fractionation or electrophore i~. Pa~sive pro~ection may be accomplished by ~tandard monoclonal antibody procedures or by immunizing suitable mammalian hosts.
In a preferred embodiment of this inventlon, the conjugate iæ used for active immunogenic vaccination of humans, especially infants, children, or immu~ocompromised individuals. For additional stability, these conjugates may also be lyophilized in the pre~ence of lacto~e (for example, a~ 20 ~g/mL
of P~P/4 mg/mL lactose) prior to u~e.

.

A preferred dosage level is an amount of e~ch of the MIEP-conjugates, or derivative thereof to be administered, corresponding to between approximately 2 to 20 ~g of PRP, or about 1 microgram to 5 milligarams of peptide in the MIEP-conjugate form for conjugates of ~ influenzae serotype B
polysaccharide, or ~IV PND peptide, in a single administration. If necessary, an additional one or two doses of the MIEP-conjugate, or derivative lo thereof, in a dosage comparable to that described above.
The invention is further defined by reference to the following examples, which are intended to be illustrative and not limiting.

E~AMPL~ 1 Preparation of ~eisseria meningitidis Bll Serotype 2 O~IPC - -A. Fer~entation 1. ~ciaLçsi~ menin~idis GIOUP B11 A tube containing the lyophilized culture of Neisseria meningi~idis (obtained from Dr. ~.
Artenstein, Walter Reed Army Institute of Research (WRAIR), Washington, D.C.~ was opened and Eugonbroth (BBL) waæ added. The culture was streaked onto Mueller ~inton agar slants and incubated at 37C with 5% C0~ for 36 hours, at which time the growth was harvested into 10~/~ sklm milk medium ~Difco), and aliguots were frozen at 70C. The identity of the ~ ~, r~
130/G~B - 83 - 181~9IA

organism was confirmed by agglutinatiqn with specif;.c antiserum supplied by WRAIR, and typing serum supplied by Difco.
A vial of the culture from the second passage was thawed and streaked onto 10 Columbia Sheep ~lood agar plates (CBAB-BBL). The plates were incubated at 37~C with 5% C02 for 18 hours after which time the growth was harvested into 100 mL of 10% skim milk medium, aliquots were taXen in 0.5 mL
amounts and ~rozen at -70C. The organism was positively identified by agglutination with specific antiserum, sugar fermentation and gram stain.
A vial of the culture from this passage wa~
thawed, diluted with Mueller-Hinton Broth and streaked onto 40 Mueller-Hinton agar plates. The plates were incuba~ed at 37C with 6% C02 for 18 hours after which time the growth harvested into 17 mL of 10% skim milk medium, aliquotted in 0.3 mL
amou~ts and frozen at -70C. The organi~m was positi~ely identified by Gram stain, agglutination with specific antiserum and oxidase test.
2. Fermentation and collection of cell paste a. Inoculum Development- The inoculum was grown ~rom one frozen vial of Neisseria memingitidis 2S Group B, B~ll from above (passage 4). Ten Mueller-~inton agar slants were inoculated, and ~ix were harvested approximately 18 hours later, and used as an inoculum for 3 250 mL flasks of Gotschlich's yeast dialysate medium at pX 6.35. The O.D.660 was adjusted to 0.18 and incubated until the OD660 was between 1 and 1.8. 1 mL of thi~ culture waæ used to inoculate each of 5 2L. Erlenmeyer ~la~k~ (each containing 1 liter of medium; see beIow) and incubated at 37DC in a shaker at 200 rpm. The O.D.

~ i ~ s ~ ~ ~

was monitored at hourly intervals following inoculation. 4 liters of broth culture, at an O.D.660 of 1.28 resulted.
70 Liter Seed Fermenter- Approximately 4 liters of seed culture was used to inoculate a sterile 70-liter fermenter containing about 40 liters of complete production medium (see below).
The conditions for the 70-liter fermentation included 37C, 185 rpm with 10 liters/minute air sparging and lo constant p~ control at about p~ 7.0 for about 2 hours. For this ba~ch, ~he final O.D.660 was 0.732 after 2 hours.
800-Liter Production Fermenter Approximately 40 liters of seed culture were u~ed to inoculate a sterile 800 liter fermenter containing 568.2 liters of complete production medium (see below). The batch was incubated at 37C, 100 rpm with 60 liters/minute air ~parging and constant pH control at p~ 7Ø For this batch, the final O.D.
was 5.58 thirteen hours after inoculation.
3. Complete Medium for Erlenmeyer flasks and 70-and 800-liter fermenters :
- Fractio~ A
25 L-glutamiC acld ~ 1.5 ~/liter NaCl 6.0 g/liter Na2~P04.anhydrous 2.5 g/liter N~4C1 1.25 g/liter KCl 0.09 g/liter L-cysteine ~Cl 0.02 g/liter . .

, .
:

, ~ , ~ractio~ B (Gotschlich's Yea~t Dialysate~: -1280 g of Difco Yeast Extract was dissolved in 6.4 liters of distilled water. The solution was dialyzed in 2 Amicon DC- 30 hollow fiber dialysis units with three HlOSM cartridges. 384 g MgSO4.7-H2O
and 3200 g dextrose were dissolved in the dialysate and the total volume brou~ht to 15 liters with distilled water. The p~ wa~ adjusted to 7.4 with NaOH, sterilized by passage throu~h a 0.22 ~ filter, and transferred to the fermenter containing Fraction A.
For the ~rlenmeyer flasks: l liter of Fraction A and 25 mL of Fraction B were added and the p~ was adjustcd to 7.0-7.2 with NaOH.
For the 70 liter fermenter: 41.8 liters of Fraction A and 900 mL of Fraction B were added and the pH was adjusted to 7.0-7.~ with NaO~.
For the 800 liter fermenter: 553 liters of Fraction A a~d 15.0 liters of Fraction B were added and the pH was adjusted to 7.1-7.2 with NaOH.
d. ~arvest and Inactivation After the fermentation was completed, phenol was added in a separate vessel, to which the cell broth was then transferred, yielding a final phenol concentration of about 0.5%. The material was held a room temperature with gentle stirri~g until the culture was no longer viable (about 24 hours).
e. Centrifugation After about 24 hours at 4C, the 614.4 liters of inactivated culture fluid ~as centrifuged through Sharples continuous flow centrifuge~. The weight of the cell paste after phenol treatment was 3.875 kg.

130/G~B - 86 - l B. OMPC Isolation S~E~1. Concentration and diafiltration The phenol inactivated culture was concentrated to about 30 liters and dia~iltered in ~terile distilled water using O.L micro-hollow fiber filters (ENKA).

.xtraction An equal volume of 2X TED buffer [0.1 M TRIS
0.01 ~ EDTA Buffer, p~ 8.5, with 0.5~tO sodium deoxycholate~ was added to the concentrated diafiltered cells. The suspension was transferred to a temperature regulated tank for OMPC extraction at 56 C with a~itation for 30 minutes.
The extract was centrifuged at about 18,000 rpm in a Sharples continuous flow centrifuge at a flow rate o~ about 80 mL/minute, at about 4C, The viscou~ supernatant ~as then collected and stored at 4C. The e~tracted cell pellets were reextracted in TED buffer as descri~ed above. The supernatants were pooled and stored at 4C.

Concentration by Ultrafiltration The pooled extract ~as transferred to a temperature regulated vessel attached to AG~Tech 0.1 micron polysulfone filters. The temperature of the extract was held at 25C in the vessel ~hroughout the concentration process. The sample was concentrated tenfold at an average transmembrane pressure of between 11 and 24 psi.

f~ J~ 6 ~ f ,rJ~
130/G~B - 87 - 18159IA

C~llection and Washing of khe OMPC
The retentate from Step 3 was centrifuged at about 160,000 x g (35,000 rpm) at about 70C in a continuous flow centrifuge at a flow rate between 300 to 500 mL/minute, and the supernatant was discarded.
The OMPC pellet was suspended in TED Buffer (190 mL buffer; 20 mL/g pellet) Step 2 and Step 4 were repeated twice (skipping Step 3).

Ste~ ~. Reco~ery of OMPC Product The washed pellets from Step 4 were suspended in 100 mL distilled water with a glass rod and a Dounce homogenizer to insure complete suspension. The aqueous OMPC suspension was the~
filter s~erilized by passage through a 0.22 ~ filter, and the TED buf~er was replaced with water by diafiltration against sterile distilled water using a 0.1 ~ hollow fiber filter.

Preparation of ~ Influenzae Type b Capsular Polysa~chari~e ~PRP~

I~oculu~ a~d Seed De~elopme~t A Stage: A lyophilized tube of Haemophilus influenzae type b, (cultured from Ross 768, received from State University of New York) was suspended in 1 mL of sterile ~aemophilus inoculum medium Ssee below) and this ~uspension was spread on 9 Chocolate Agar 130/G~B - ~8 - 18159IA

slants (BBL). The pH of the inoculum medium was adjusted to 7.2 + 0.1 (a typical value was pH 7~23) and the medium solution was sterilized prior to use by autoclaving at 121C for 25 minutes. After 20 hours incubation at 37C in a candle jar, the growth from each plate was resuspended in 1-2 mL ~aemophilus inoculum medium, and pairs of slants were pooled.

Haemophilus Inoculum Medium Soy Peptone 10 gm/liter-NaCl 5 gm/liter NaH2P04 3.1 gm/liter Na2~P04 13.7 gm/liter K2HP04 2.5 gm/liter .
Distilled Water To Volume ~1 r~ ~, r~

The resuspended cells from each pair of slants was inoculated into three 250 mL Erlenmeyer flasks containing about lO0 mL of Haemophilus Seed and Production medium. The 250 mL flasks were incubated at 37C for about 3 hours until an OD660 of about 1.3 was reached. These cultures were used to inoculate the 2 liter flasks (below).
B Stage: 2 Liter non-baffled Erlenmeyer flasks- 5 mL of culture from "A stage" ~above) were lo used to inoculate each of ~ive two-liter flasks, each containing about 1.0 liter of complete ~aemophilus eed and production medium (see below). The fla~ks were then incubated at 37C on a rotary shaker at about 200 rpm for about 3 hours. A typical OD660 1~ value at the end of the incubation period was about 1Ø

Complete Haemophilus Seed And Production Medium Per liter NaH2P04 3.1 glL
Na2HP04 13 . 7 gl L
Soy Peptone 10 g/L
Yeast extract diàfiltrate (1) 10 glL
K2~P04 2.5 glL
NaCl 5.0 g/L
~lucose (2) 5.0 g/L
Nicotinamide adenine 2 mglL
dinucleotide (NAD~ (3) Hemin (4) 5 mglL

' 7J r~

The salts and soy peptone were dissolved in small vo.umcs of hot, pyrogen-free water and brought to correct final volume with additional hot, pyrogen-free water. The fermenters or flasks were then sterilized by autoclaving for about 25 minutes at 121C, and after coolin~ yeast extract diafiltrate (1), glucose (2), NAD (3), and hemin (4) were added aseptically to the flasks or fermenters prior to inoculation.
(1) Yeast extract diafiltrate: 100 g brewers' yeast extract (Amber) was dissolved in 1 liter distilled water and ultrafiltered usin~ an Amicon DC-30 hollow ~iber unit with H10 x 50 cartridges with a 50 kd cutoff. The filtrate was collected and sterilized by passage through a 0.22 filter.
(2) Glucose was prepared as a s~erile 25%
solution in distilled water.
(3) A stock solution of NAD containing 20 20 mg/mL was sterilized by paæsage through a (0.22 ~ -filter) and added aseptically just prior to inoculation.
(4) A stock ~olution of Hemin 3X was ~ prepared by dissolving 200 mg in 10 mL of 0.1 M NaOH
2~ and the ~olume adjusted with distilled, sterilized water to 100 mL. The solution was sterilized ~or 20 : minutes at 121C and added asep~ically to the final medium prior to inoculation.

, ;.

C Stage: 7Q Liter Seed Fermenter- Three liter~
of the product o~ B Stage was used ~o inoculate a fermenter containing about 40 liters of Complete ~aemophilus Seed And Production medium (prepared as described above) and 17 mL UCON B625 antifoam agent.
The p~ at inoculation was 7.4.
D Stage: 800 Liter Product;on Fermenter-Appro~imately 40 liters of the product of "C Stage"
was used to inoculate an 800 liter fermenter lo containing 570 liters of Haemophilus Seed and Production medium ~prepared as described above), scaled to the necessary volume, and 72 mL of UC~N
LB625 antifoam agent was added.
The fermentation was maintained at 37C with 100 rpm agitation, with the O.D.660 and p~ levels measured about every t~o hours until the O.D.660 was stable during a two-hour period, at which time the fermentation was terminated (a typical final O.D.660 was about 1.2 after about 20 hours).
~ARVEST AND INACTIVATION
Approximately 600 liters of ~he ba~ch was inactivated by harvesting into a "kill tank"
containing i2 liters of 1% thimerosal.
CLARIFICATION
After 18 hours of inactivation at 4C, the batch was centrifuged in a 4-inch bowl Sharples contiuous flow centrifuge at a flow rate adjusted to maintain product clarity (variable between 1.3 and 3.0 liter~ per minute). The supernatant obtained after centrifu~ation (lS,000 rpm) was used for product recovery.

," ~ ,.~
1301&HB - 92 - 1815~IA

ISOLATION AND CONCENTRATION BY ULTRAEILTRATION
The supernatant from two production fermentations was pooled and concentrated at 2 to 8~C
in a Romicon XM-50 ultrafiltration unit with twenty 50 kd cut-off hollow fiber cartridges (4.5 m2 membrane area; 2.0 Lpm air flow and 20 psi>.
Concentration was ~uch that after approximately 4.5 hours, about 1,900 liter~ had been concentrated to 57.4 liters. The filtrate was discarded.

48% AND 61% ETHANOL PRECIPITATION
To the 57 . 4 ~ iters of ultrafiltration rctentate, 53 liters of 95% ethanol was added dropwise over 1 hour with stirring at 4C to a final concentration of 48% ethanol by volume. The mixture was stirred two additional hours at 4C to insure complete precipitation, and the superna~ant was collected following passage through a single 4-inch Sharples continuous flow centrifuge at 15,000 rpm at a ~low rate of about 0.4 liters per minute. The pellet was di3carded and the clarified fluid was brou~ht to 82% ethanol with the addition of 40.7 liters of 95% ethanol over a one hour period. The mi~ture was stirred for three additional hours to 2s insure complete precipitation.

RECOVERY OF THE S~COND PELLET
The resulting 48% ethanol-soluble-82%
ethanol-insoluble precipitate was collected by centrifugation in a 4 inch Sharples continuous flow centri~uge at 15,000 rpm with a flow rate of about 0.24 liters per minute and the 82% ethanol supernatant wa3 disc~rded. The crude product yield was about 1.4 kg of wet paste.

J ~ , " ~ ~

CALCIUM CHLORIDE EXTRACTION
The 1.4 kg grams of 82% ethanol-insoluble m~terial, was mixed in a Daymax dispersion vessel 2-80C with 24.3 liters of cold, distilled water. To this mixture, 24.3 liters of cold 2M CaC12.2H20 was added, and the mixture was incubated at 4C for 15 minutes. The vessel was then rinsed with 2 liters of 1 M CaC12.2H20, resulting in about 50 liters final volume.

23% ET~IANOL PRECIPITATION
The 50 liters of CaC12 extract was brought to 25% ethanol by adding 16.7 liters of 95~tO etha~ol dropwise, with stirring, at 4C over 30 minutes.
After additional stirring for 17 hours, the mixture was collected by passage through a Sharples continuous ~1GW centrifuge at 4C. The supernatant was collected and the pellet was discarded.

38% ET~ANOL PRECIPITATION AND
COLLECTION OF CRUDE PRODUCT PASTE
The 257/o ethanol-soluble ~upernatant was brought to 38% ethanol by the addition of 13.9 liters of 95% ethanol, dropwise with stirring, over a 30 minute period. The mi~ture was then allowed to ætand with agitation for one hour, then without agitation for 14 hours, to insure complete precipitation. The resulting mixture was then centrifuged in a 4 inch Sharples continuous flow centrifuge at 15,000 rpm (flow rate of 0.2 liters per minute) to collect the precipitated crude ~. influenæae polysaccharide.

~ , ~s, 3 TRITURATION
The pellet from the centrifugation was transferred to a 1 gallon Waring Blender containing 2 to 3 liters o~ absolute ethanol and blended for 30 seconds at the highest speed. Blending was continued for 30 seconds on, and 30 seconds off, until a hard white powder resulted. The powder was collected on a Buchner funnel with a teflon filter disc and washed sequentially, in sit~, with two 1 liter portions of absolute ethanol and two 2 lites portions of acetone. The material was then dried, in vacuo, at 4OC for 24 hours, resulting in about 337 g (dry weight) of product.

P~ENOL EXTRACTION
About 168 grams o~ the dry material ~rom the trituration step (about half of the total) was resuspended in 12 liters of 0.488 M sodium acetate, pH 6.9, with the aid of a Daymax dispersion vessel.
The ~odium acetate solution was immediately extracted with 4.48 liters of a fresh aqueous phenol solution made as follows: 590 mL of 0.488 M sodium acetate, pH

6.9, was added to each of eight 1.5 kg bottles of phenol (Mallinckrodt crystalline~ in a 20 liter pressure vessel and mi~ed into ~uspension. Each phenol extract was centri~uged for 2.5 hours at 30,000 rpm and 4C in the K2 Ultracentrifuge (Electronucleonics). The agueous effluent was extracted three additional times with fresh aqueous phenol solution as described above. The phenol phases were discarded.

~, 130/G~B - 95 18159IA

ULTRAFILTRATION
The aqueous phase from the first phenol extraction above (12.2 liters) was diluted with 300 liters of cold, distilled water and diafiltered at 4C on an Amicon DC-30 ultrafiltration apparatus using 3 HlOP10, 10 kd cutoff cartridges, to remove the carryover phenol. The Amicon unit was rinsed and the rinse added to the retentate, such that the final volume was 31.5 liters. The ultrafiltrate was lo discarded.
67% ETHANOL PRECIPITATION
0.81 liters of 2.0 M CaC12 was added to the 31.5 liters of dialysate from the previous step (final CaC12 concentration was 0.05 M~ and the solution was brought to 82% ethanol with dropwise addition and rapid stirring over one hour, of 48.5 liters of 95% ethanol. After 4 hours of agitation, then standing for 12 hour~ at 4C, the supernatant was siphoned off and the precipitate was collected by centri~ugation in a 4 inch Sharples continuous flow centrifuge (lS,OOO rpm), at 4C for 45 minutes. The resulting polysaccharide pellet was triturated in a 1 gallon Waring blender using 30 second pulses with 2 liters of absolute ethanol, eollected on a Buchner funnel fitted with a teflon filter disc, and washed, in ~i~, with four 1 liter portions of absolute ethanol followed by two 1 liter portions of acetone.
The sample was then dried in a tared dlsh, in va~uo, at 4C for 20 hours. The yield was about 102 grams of dry powder. The yield from all phenol extractions was pooled resulting in a total of 212.6 grams of dry powder.

~, v, s~ f,l 130/G~B - 96 - 18159IA

ULTRACENTRIFUGATION IN 29% ETHA~OL
AND COLLECTION OF FINAL PRODUCT
The 212.6 grams of dry powder from above was dissolved in 82.9 liters of distilled water, to which was added 2.13 liters of 2 M CaC12.2~20, (0.05M
CaC12), 2.5 mg polysaccharide/mL), and the mixture was brought 29% ethanol wi~h the dropwise addition of 29.86 liters of 95% ethanol over 30 minutes. The mi~ture was clarified immediately by centri~ugation lo in a K2 Ultracentrifuge containing a K3 titanium bowl and a Kll Noryl core (30 9 000 rpm and 150 mL per minute flow rate) at 4~C. The pellet waæ discarded and the supernatant was brought to 38% ethanol by the : addition of 17.22 liters o~ 95% ethanol over 30 - 15 minutes with stirring. After stirring 30 additional minutes the mixture was allowed too stand without agitation at 4C for 17 hours and the precipitate was collected using a 4 inch Sharples continuous flow centrifuge at 15,000 rpm (45 minutes was required).
The resulting paste wa~ transferred to a l-gallon Waring blender containing 2 liters of absolute ethanol and blended at the highest speed by 4 or 5 cycles of 30 seconds on, 30 seconds off, until a hard, white powder ~ormed. Thiæ powder was collected on a Buchner funnel fitted with a Zitex teflon`disc and rinsed sequentially, in ~, with two fresh 0.5 liter portions and one 1 liter portions of absolute ethanol, and with two 1 liter portion of acetone. The product was removed from the funnel and transferred to a tared dish for drying, in vacu3, at 4C (for 25 hours). The final yield of the product was 79.1 grams dry weight.

6~,rS ?~
130/GHB - 97 - 1815gIA

Cloning of Genomic DNA Encoding MIEP.
About 0.1 g of the phenol inactivated N.
meningitidis cells (see Example 1) was placed in a fresh tube. The phenol inactivated cells were resuspended in 567 ~L of TE buffer tlOmM TRIS-HCl, lmM EDTA, pH 8.0]. To the resuspended cells was added 30 ~L of 10~/o SDS, and 3 ~L of 20 mg/mL
proteinase K (Sigma). The cells were mixed and incubated at 37C for about 1 hour, after which 100 ~L of 5 M NaCl was added and mixed thoroughly. 80 ~L
of 1% cetyltrimethylamonium bromide (CTAB) in 0.7 M
NaCl was then added, mixed thoroughly, and incubated at 65C for 10 minutes. An equal ~olume (about 0.7 to 0.8 mL) of chloroform/isoamyl alcohol (at a ratio of 24:1, respectively) was added, mixed thoroughly and centrifuged at about 10,000 x g for about 5 minutes. The aqueous (upper) p~ase was trans~erred to a new tube and the organic phase was diecarded.
An equal volume of phenol/chloroform/isoamyl alcohol (at a ratio of 25:24:1, respectively) wa added to the aqueous phase, mixed thoroughly, and centrifuged at 10,000 x g for about 5 minutes. The aqueous phase (upper) was transferred to a ne~ tube and 0.6 yolumes (about 420 ~L) of isopropyl alcohol was added, mixed thoroughly, and the precipitated DNA was pelletted by centrifugation at 10,000 x g for 10 minute~. The supernatant was discarded, and the pellet was washed with 70X ethanol. The DNA pellet was dried and resuspended in 100 ~L of TE buffer, and represents N.
meningitidis genomic DNA.

7' ~: ~ ^"-.
" "i ~,; "",;

Two DNA oligonucleotides were synthesized which correspond to the 5~ end of the MIEP gene and to the 3' end of the MIEP gene [Murakami, E.C. et al., (1989), Infection and Immunity, 57, pp.2318-23]. The sequence of the DNA oligonucleotide specific for the 5' end of the MIEP gene was:
5~-ACTAGTTGC MTGAAAAAATCCCTG-3~; and for the 3~ end of the MIEP gene was: S'-GAATTCAGATTAGG M TTTGTT-3'.
These DNA oligonucleo~ides were used as primers for l-o polymerase chain reaction (PCR) amplification of the MIEP ~ene using 10 nanograms of N. meningi~idis genomic DNA. The PCR amplification step was performed according to the procedures supplied by the manufacturer (Perkin Elmer~.
The amplified MIEP DNA was then digested with the restriction endonucleases ~1 and ~coRI.
The 1.3 kilobase (kb) DNA fragment, containing the complete coding region of MIEP, was isolated by electrophoreæis on a 1.5% agarose gel, and reco~ered from the gel by e~ectroelution ~Current Protocols in Molecular Biology, (1987), Ausubel, R.M., Brent, R., gingston, R.E., Moore, D.D., Smith, J.A., Seidman, J.G. and Struhl, K., eds., Greene Publishing Assoc.~
The plasmid vector p~C-19 was digested with ~QI and E~Q~I. The gel puri$ied ~QI-EcoRI MIEP DNA
was ligated into the SpeI-E~oRI pUC-19 vector and was u~ed to transform E. coli strain D~5. Transformants containing the p~C-19 vector with the 1.3 kbp MIEP
DNA were identi~ied by restriction endonuclease mapping, and the MIEP DNA was seguenced to ensure its identity.

~ J~ -~

~ XAMPLE 4 Construction of the pcl/l.GallOp(B~ADHlt vector.
The Gal 10 promoter was isolated from plasmid YEp52 [Broach, et al., (1983) in Experimental Manipulation of Gene Expression, Inouye, M(Ed) Academic Press pp. 83-117J by gel purifying the 0.5 kilobase pair (Kbp) fragment obtained after cleavage with Sau 3A and Hind III. The A~Hl terminater was isolated from vector pGAP.tADH2 ~Kniskern, et al., 91986), Gene, 46, pp. 135-141] by gel purifying the 0.35 Kbp fragment obtained by clea~age with Hind III
and SpeI, The two fragments were ligated with T4 DNA
ligase to the gel purified pucl3~Hind III vector (the ind III site ~as eliminated by digesting pUC18 with Hind III, blunt-ending with the Klenow fragment of E. ~oli DNA polymerase I, and ligating with T4 DNA
ligase) which had been digested with Bam~I and S~hI
to create the parental ~ector pGallo-tADHl. This has a unique Hind III cloning æite at the GallOp.AD~lt junction.
The unique ind III cloning site of pGallO.tAD~l was changed to a unique Bam~I cloning site by digesting pGallO.tADHl with ~ind III, gel purifying the cut DNA, and ligating, using T4 DNA
liga~e, to the following El~ am~I linker:
5'-AGCTCGGATCCG-3' 3'-GCCTAGGCTCGA-5'.

The resulting plasmid, pGallO(B)t~DHl, has deleted the ~in~_~lI site and generated a uni~ue BamHI cloning site.

130/G~B - 100 - ~5 ,$~

The GallOp.tADHl fragment was isolated from pGallO(B)tAD~l by digestion with SmaI and SphI, blunt-ended with T4 DNA polymerase, and gel purified. The yeast shuttle vector pCl/l ~Brake et al., (1984), Proc. Nat~l. Acad. Sci. USA, 81 , pp.4642-4646] was digested with SphI, blunt-ended with T4 DNA polymerase, anpurified. This fragment ~as ligated to the vector with T4 DNA ligase. The ligation reaction mixture was then used to transform E. coli HB101 cells to ampicil~in resistance, and transformants were screened by hybridization to a single strand of the 32P-labelled HindIII BamXI
linker. The new vector construction, pcill.GallOp(B)AD~lt was confirmed by digestion with ~indIII and Ba~HI.

;~XA~
Construction of a Yeast MIEP Expression Vector with MI~P + L~d~r DN~ Sequen~es _ _ A DNA fragment containi~g the complete coding region of MIEP was generated by digestion of pUC19.MIEP #7 with SpeI and coRI, gel purification of the MIEP DNA, and blunt-ended with T4 DNA
polymeraæe.
The yeast internal expression vector pCl/l.GallOp(B)AD~lt wa~ disgested with Bam ~I, dephosphorylated with calf intestinal alkaline phosphatase, and blunt-ended with T4 DNA polymerase.
The DNA was gel purified to remove uncut vector.

~. ' 130/GHB - 101 ~ s, , The 1.1 Kbp blunt-ended fragment of MIEP was ligated to the blunt-ended pcl/l.GallOp(B)ADHlt vector, and the ligation reaction mixture was used to transform competent ~. coli D~5 cells to ampicillin resistance. Transformants were screened by hybridization to a 32P-labelled DNA oilgoncleotide:
5'... AAGCTCGGATCCTAGTT&CAATG...3', which was designed to be homologous with se~uences overlapping the MIEP vector junction. Preparations f DNA were made from hybridization positive transformants and digested with ~LI and SalI to verify that the MIEP fragment was in the correct orientation for expression ~rom the GallO promoter.
Further confirmation of the DNA conætruction was -obtained by dideoxy 3equencing from the GallO
promoter into the MIEP coding re~ion.
Expression of MIEP by the transformants was detected by Western blot analysis. Recombinant MIEP
produced in the transformants comigrated on polyacrylamide gels with MIEP purified from OMPC
vesicles, and was immunologically reactive wi~h antibodies specific for MIEP.

Construction of yeast MIEP expression vector with a 5'-~odified MI~p DNA Se~nee. ~ _ A DNA oligonucleotide containing a HindIII
site, a conserved yeast 5' nontranslated leader (NTL), a methionine start codon ~ATG), the first 89 codons of the mature MIEP (beginning with Asp at position +20) and a ~1 site (at position +89) was ~, :

generated using the polymerase chain reaction (PCR) technique. The PCR was performed as specified by the manufacturer (Perkin Elmer Cetus) using the plasmid pUC19MIEP42#7 as the template and the following DNA
oligomers as primers:
5 CTAAGCTTAACAAAATGGACGTTACCTTGTACGGTACAATT3 , and 5 ACGGTACCGM GCCGCCTTTCAAG3 .
To remove the 5' region of the MIEP clone, plasmid pUC19MIEP42#7 was digested with ~al and lo ~indIII and the 3.4 Kbp vector fragment was agarose gel purified. The 280 bp PCR fragment was digested with ~al and HindIII, agarose gel purified, and ligated with the 3.4 Kbp vector fragment.
Transformants of E. coli HBlOl (BRL) were screened.by DNA oilgonucleotide hybridization and the DNA from positive transformants was analyzed by restriction enzyme digestion. To ensure that no mutations were introduced during the PCR step, the 280 bp PCR
generated DNA of the positive transformants ~as sequenced. The resulting plasmid containæ a ~indIII
- EcoRI insert consisting of a yeast NTL, ATG codon, and the entire open reading frame (ORF) of MIEP
~eginning at the Asp codon (amino acid ~20).
The yeast MIEP expression vectors were constructed as follows. The pGAL10/pcl/l and pGAP/pCl/l vectors ~Vlasuk, G.P., et al., (1989~
J.B.C., ~64, pp.l2,106-12,112] were digested with BamXI, flush-ended with the Klenow fragment of DNA
polymerase I, and dephosphorylated wlth calf intestinal alkaline phosphatase. These linear vectors were ligated with the Klenow treated and gel puri~ied ~indIII - ~nRI fragment described above, which contains the yeast NTL, ATG and ORF of MIEP are forming pGallO/pcl/MIEP and p&AP/pGAP/pCl/MI~P.

l30/GHB - 103 - 18159IA

Saccharomyces cerevisiae strain U9 (gallOpgal4-) were transformed with plasmid pGallO/p/pCl/MIEP. Recombinant clones were isolated and examined for expression of MI~P. Clones were grown at 37C with shaking in synthetic medium (leu-) containing 2% glucose (w/v) to an O.D.660 of about 6Ø Galactose was then added to 2% (w/v) to induce expression of MIEP from the GallO promoter. The cells were grown for an additional 45 hours following galactose induction to an O.D.600 of about 9Ø The cells were then harvested by centrifugation. The cell pellet was washed wi~h distilled water and frozen.

Western ~lot For Reeombinant ~IEP:
To determine whether the yeast was expressing MIEP, Western blot analysis was done.
Twelve percent, 1 mm, lO to 15 well Novex Laemmli gels are u~ed. The yeast cells were broken in ~2 using glass beads (sodium dodecylsulfate (SDS) may be used at 2% during the breakin~ process). Cell debris was removed by centrifugation for l minute at lO,OOO
x g.
The supernatant was mixed with sample running buffer, as described for polyacrylamide gel purification o MIEP. The samples were run at 3S mA, - using OMPC as a reference control, until the dye part l~aves the gel.
Proteins were transferred onto 0.45 ~ pore size ~itrocellulose paper, using a NOV~X transfer apparatus. After transfer the nitrocellulose paper was blocked with 5% bovine serum albumin in phosphate ~ ,rl i~ r, r. ~

buffered saline for 1 hour, after which 15 mL of a 1:1000 dilution of rabbit anti-MIEP (generated from gel purified MIEP using standard procedures) was added. After overnight incubation at room temperature 15 mL of a 1:1000 of alkaline phosphatase conju~ated goat anti-rabbit IgG was added. After 2 hours incubation the blot was developed using FAST
RED TR SALT (Sigma) and Naphthol-AS-MX phosphate (Sigma).

~XAMPLE 7 Bact~rial Expression Of RecQm~inant MIEP
Plasmid pUCl9-MIEP containing the 1.3 kilobase pair MIEP gene insert, was digested with restriction endonucleases SpeI and ~coRI. The l.l~bp fragment was isolated and purified on an agarose gel using standard techniques known in the art. Plasmid pTACSD, containing the two cistron TAC promoter and a unique ECORI site, was digested with ~ORI. Blunt ends were formed on both the 1.3 kbp MIEP DNA and the pTACSD vector, using T4 DNA polymerase (Boehringer Mannheim) according to the manufacturer's directions. The blunt ended 1.3 kbp MIEP DNA was ligated into the blunt ended vector using T4 DNA
ligase (Boehringer Mannheim) according to the manufacturer's direction~. The ligated DNA was used - to transform E. coli strain DH5aIQMAX (BRL) according to the manufacturer's directions. Transformed cells were plated onto agar plates containing 25 ug kantamycinlmL and 50 ug penicillin/mL, and incubated for about 15 hours at 37 C. A DNA oligonucleotide , with a sequence homologous with MIF.P was labelled with 32p and used to screen nitrocellulose filters containin~ lysed denatured colonies from the plates of tranæformants using standard DNA hybridization technigues. Colonies which were positive by hybridization were mapped using restriction endonucleases to determine the orientation of the MIEP gene.
Expression of MIEP by the transformants was lo detected by Western blot analysis. Recombinant MIEP
produced in the transformants comigrated on polyacrylamide gels with MIEP purified from OMPC
vesicles, and was immunologically reactive with antibodies specific for MIEP.

Preparation of Purified MIEP from OMPC Vesicles or From Recombinant Cells by Polyacrylamide Gel Electrop~oresis _ _ Acrylamide/BIS (37.5:1) gels, 18 x 14 cm, 3 mm thick were used. The stacking gel was 4Z
polyacrylamide and the æeparating gel was 12%
polyacrylamide. Approximately 5 ~g of vesicle 2s protein, or recombinant host cell protein, was used per gel. To 1 mL of OMPC ~eæicles was added 0.~ mL
of sample buffer (4% glycerol, 300 mM DTT,- 100 mM
TRIS, O. OOl~/o Bromophenol blue, pH 7.0). The mixture was heated to 105C for 20 minutes and allowed to cool to room temperature be~ore loading onto the gel. The gel wa~ run at 200 400 milliamps, with cooling, until t~e Bromophenol blue reached the .

J ~, J

bottom of the gel. A verticle strip of the gel was cut out (about 1-2 cm wide) and stained with Coomassie/cupric acetate (0.1%). The strip was destained until the MIEP band (about 38 Kd) became visible. The strip was then placed into its original gel position and the MIEP area was excised from the remainder of the gel using a scalpel.
The excised area was cut into cubes (about 5 mm) and eluted with 0.01 M TRIS-buffer, pH 0.1.
After 2 cycles of elution the eluate was evaluated for purity by SDS-PAG~. The eluate was combined with a common pool of eluates and dialysed against borate-buffer (0.1 M boric acid, pH 11.5). Af~er dialysis the eluted protein was concentrated using an 1~ Amicon stirred cell with YM10 membranes (10,090 molecular weight cutoff). The material was further purified by passage through a PD10 sizing column (Pharmacia, Piscataway, NJ), and waæ stored at room temperature in borate buffer.

EXAMPL~ 9 Carrier activi~ Q~ MIEP in covalent PRP-O~C
conju~a~
Immunizations: Male C3~1HeN mice (Charles River, Wilming~on, MA) were im~unized intraperi~oneally (IP) with PRP covalently linked to OMPC (PRP-OMPC; comprising 2.5 ~g PRP and 17 ~g OMPC) or PRP coupled to DT ~PRP-DT; containing 2.5-7.5 ~g PRP and 1.8-5.4 ~g DT) (Connaught Laboratories, Willowdale, ONT), ~uspended in O.5 mL of O.01 M
pho~phate-buf~ered saline (PBS). A ~econd group of 130/GHB - 107 ~ 18159IA

male C3~-HeN mice, received either 17 ~g of MIEP, 17 ~g of OMPC, or OMPC-IAA (OMPC derivatized with N-acetyl homocysteine thiolactone, and capped with iodoacetamide). Cell donors for adoptive transfer experiments were twice immunized IP, 21 days apart, and spleen cells were collected 10 dayæ after the second immunization. Adoptive transfer recipients were male C3H/HeN mice given 500R total body gamma-irradiation ~rom a 137cs source and immediately lo reconstituted by intravenous injection of 8 x 107 spleen cells from each of two syngeneic donors separately primed with PRP-DT, and OMPC, MIEP, or OMPC-IAA. Control mice received 8 x 107 spleen cells from one donor mouse primed with PRP-OMPC and an equal number of spleen cells from an unprimed donor mouse.
ELI~A for anti-P~P anti~Q~y: Reactive amines were introduced into purified ~ fluenzae PRP by treatment with carbonyldiimidazole and reaction with butanediamine a~ described by Marburg et al., U.S. Patent 4,882,317. This derivatized PRP
was chromatographed on Sephadex G-25 in O.lM sodium bicarbonate buffer, p~ 8.4.
N-hydroxysuccinimidobiotin (Pierce Chemical, Rockford, IL) in dimethylsulfoxide was added to the eluate to a final concentration of 0.3 mg/mL and reacted in the dark for 4 hours at ambient temperature (about 25-28OC). Unreacted N-hydroxysuccinimido-biotin was removed by gel filtration over Sephadex G-25 in PBS. Costar (Cambridge, MA) polyvinyl chloride ELISA plates were coated with 50 ~g/well of avidin ~Pierce Chemical) at ' 10 ~g/mL in 0.1 M sodium bicarbonate buffer, p~ 9.5, overnight at ambient temperature and 100% humidity.
Plates were washed 3 times with 0.05 M TRIS-buf~ered saline, pH 8.5, containing 0.05% Tween-20 (TBS-T), and blocked with TBS-T plus 0.1% gelatin (blocking buffer) at ambient temperature and 100% humidity for 1 hour. Plates were blotted without waæhing, 50 ~g/well PRP-biotin in PBS at 15-40 ~g/mL was added, and the plates were incubated for 1 hour. Plates lo were ~ashed 3 ~imes with T~S-T prior to sample addition. Samples were added in two-fold serial dilutions in blocking bu~fer, and incubated for 2 hours at ambient temperature and 100% humidity. The - plates were then washed 3 timeæ with TBS-T, and appropriate al~aline-phosphatase conjugated anti-immunoglobulins diluted in blocking buffer were added. The an~ibodies used were goat anti-mouse IgM
(Jackson Immunoresearch, West Grove, PA), IgG ~Fc) (Jackson Immunoresearch), IgGl (gamma) (BRL, 2~ Gaithersburg, MD), IgG2a (g~mma) (BRL), IgG2b (gamma) (Southern Biotechnology Associates, Birmingham, AL), IgG3 (gamma) (Southern Biotechnology Associates), and goat anti-rabbit IgG (Jackson ImmuDoresearch).
Plates were incubated for 2 hours at ambient temperature and 100% humidity, washed with blocking buffer, and ~u~strate de~elopment was carried out using p-nitrophenyl phosphate (1 mg/mL in 1 M
diethanolamine, Kirkegaard and Perry, Gaithersburg, MD~. Dilutions were considered positive if the sample absorbance exceeded the mean absorbance plus 3 times the standard deviation of 8 reagent blanks, and the difference in absorbance bet~een successive dilutions was 0.01 or greater. Endpoint titers were defined as the reciprocal of the highest dilution which gave a positive reaction in the ELISA as described above. Logarithms of reciprocal titers were compared between treatment groups by two-way analysis of variance [Lindeman, R.H. et al., (1980), Introduction to Bi~ariate and Multivariate Analysis, Scott Foresman (pub.), New York].
RIA for anti-PRP antibodv ~uantitation:
The experimental samples of serum to be tested for the amount of an~i-PRP antibodies were diluted 1:2, 1:5, and 1:20, using fetal calf serum as the diluent. 25 ~L of each diluted sexum sample was transferred, in duplicate, to 0.5 mL RIA vials (Sarstedt). A solution of PRP labelled with 125I was diluted to yield between 300 and 800 counts per minute (cpm) per 50 ~L, using phosphate buffered saline as the diluent. 50 ~L of diluted 125I-PRP was transferred to each RIA vial, mixed thoroughly and incubated for about 15 hours at 4C. 75 ~L of a saturated solution of ammonium sulfate at 4C was added to each RlA vial, mixed thoroughly and incubated at 4C for 1 hour. The RIA vials were then centrifuged for 10 minutes at 10,000 x g, the supernatant was discarded and the cpm in the pellet was measured in a gamma eounter (LKB).
A standard curve consisting of serial two-fo~d dilutions o~ an antiserum containing a known quantity of anti-PRP antibodies was prepared as described above and were assayed concomitantly with the experimental serum samples. The quantity of anti-PRP antibodies in the standard curve was between 14 ~g/mL a~ the highest quantity of antibodie~ and 0.056 ~g/mL as the lowe~t quantity of antibodies.
amples were run in duplicate.

130/GHB - 110 - 18159 ~ ~r~~

The average CPM of the duplicate samples was compared with the standard curve to calculate the amount of anti-PRP antibodies present in the experimental serum 6amples.
Antibody_~çs~ona~~ of adoptive transfer L~cipients; Antibody responses of ad~pt~ve transfer rec1pient~ receiving spleen cells primed separat~ly with PRP-Dr and MIEP, or OMPC, or IAA-OMPC, were mea-sured by ELISA and RIA in blood samples taken on ~he indicated days post-immunization with PRP-OMP~. Rec~- -pients of spleen cells primed separately with PRP-DT, and eith~r MIEP or OMPC or IAA-OMPC, responded to im-munization with PRP-OMPC by production of equivalent amounts of serum IgGl and IgG2a anti-PRP antibody w~thin ~ d~y~. Irradiated mice reconstituted with ~pleen cell~ whieh were carrier-primed with MIEP or 0MPC or IAA-0MPC, had ~ignificantly higher IgG1 (p<0.001) and IgG2a (p<0.04) anti-PRP antibody titer~ after immunization with PRP-OMPC than control mice, given PRP-DT-primed but not OMPC-primed ~pleen cells. The serum antibody re~pon~e~ to immunization with PRP-0MPC in mice given ~pleen cell~ prim2d 8eparately with PRP-DT a~d eit~er ~IEP or OMPC or IAA-OMPC were comparable to those in mice given ~pleen eell~ primed 2s with PRP-OMPC (p~0.12 for Ig&l antibody o~ day~ 6-13.
and p>0.5 Ior IgG2a antibody on day~ 9-13). ~o antibody re~pvn~e was ~een when irradiated n~ice reconætituted with PRP-DT-primed and eitber MIEP or ûMPC~primed splecn cell~ were immunized with PRP
without a protei~ carrier. Sta~icial a~aly~i~ was done by two-wa8 analysis of variance (ANOVA) ~Lindeman, ~.~. et al., Introduction to Bivariate and Mul'ciYariate Analysis, (1980), Scott Foresmall, New Yo~
These results demonstrate that MIEP
functioned i~ mice as wcll as OKPC ~o induce a carrier T helper cell recponse for t~e generation of a~ti-~P IgG antibodies.

6 -",'~ n ~? i 130/G~B ~ 18159IA

Mitogeni~ Activitv of MIEP
MIEP purified from N. menin~itidis OMPC was tested for mitogenic activity in a lymphocyte - 5 proliferation assay. Murine splenic lymphocytes were obtained from C3H/HeN, C3H/FeJ, C3E/HeJl or Balb/c mice. The mice were either naive or had previously been vaccinated with PRP-OMPC. The spleen cells were passed through a sterile, fine me~h screen to remove the stromal debris, and suspended in K medium [RPMI
1640 (GIBCO) plus 10% fetal calf serum (Armour>, 2 ~M
Glutamine (GI3CO), 10 mM ~epes (GIBCO), 100 u/mL
penicillin/100/~g/mL streptomycin (GIBCO), and 50 ~M
~-mercaptoethanol (Biorad)]. Following pipetting to disrupt clumps of cells, the ~uspension was centrifuged at 300 x g for 5 minutes, and the pellet was resuspended in red blood cell lysis buffer [90%
0.16 M ~H4Cl (Fisher3, 10% 0.7 TRIS-XCl (Sigma), pH
7.2] at room temperature, 0.1 mL cells/mL buffer ~or two mi~utes. Cells were underlayered with 5 mL of ~etal calf serum and centrifuged at 4,000 x g for 10 minutes, then washed with K medium t~o times and resuspended in K medium at 5 x 106 cells/mL. These cells were plated (100 ~L/well3 into 96 well plates along with 100 ~L of proteiIl or peptide sample, in triplicate.
The MIEP of N. menin~itidis was purified as previously described in E~ample 7. Control proteins included bovine serum albumin, PRP-OMPC and OMPC
itself, and lipopolysaccharide (endotoxin). All samples ~ere diluted in K medium to concentrations of 1, ~.5, 13 9 26 7 52, 105l and 130 ~gjmL, then plated ::

:

rJ ~

as described above such tha~ their final concentrations were one-half of their original concentrations. Triplicate wells were al~o incubated for each type o~ cell suspended in K medium only, to determine the baseline of cell proliferation.
On day 3, 5, or 7 in culture, the wells were pul~ed with 25 ~L of 3H-thymidine (Amer~ham) containing 1 mCi/25 ~L. The We11B were harYeRted 16-18 hour~ later on a Skatron harve~ter, and counts per minute (CPM) was measured ;n a liquid scintillation counte~. The net change in cpm wa~
calculated by ~ubtracting the mean numbe~ of cpm : ~aken up per well by cells î~ ~ ~edium alone, ~rom the mean of the e~perimental cpm. The stimulation.
index was determined by divid~ng the mean experimental cpm by the mean cpm of the control wells.
Lymphocyte proliferation as~ay for ~itogenic activity of MIEP, in vitro. The increase in 3H-thymidine incorporation into cellular DNA wa~ measured 20 following exposure of the cells to bovine seru~n albumin (BSA), PP~P-OMPC, OMPC, ~IEP, or CNBr. HIEP as well as ONPC and PRP-O~IPC vaccine resulted in proliferation of lymphocytes from previously vaccinated mice. Th~s ~itogenic act~vity did ~ot appear to be due to lipopolysaccharlde (LPS) ~ince the ~IEP was free of detectable L~S, measured by rabbit pyrogenicity assays, and the proliferatiYe effact was greater than that which could hav2 been caused by LPS
present in amounts below the level of detectability on silYer stained polyacrylamide gels.

130/G~B - 113 - 18159IA

Conjugation of ~. influenzae type-b PRP
polvsa~charide to N. meningitidis MI~P
Chemical conjugations were conducted according to the method disclosed in U.S. Patent number 4,882,317.
10 mg of MIEP in 3 mL of 0.1 M borate buffer, pH 11.5, was mixed with 10 mg of ethylenediamine tetraace~ic acid disodium salt (EDTA, lo Sigma chemicals) and 4 mg of dithiothreitol (Sigma Chemicals). The protein solution was flushed thoroughly with N2 125 mg of N-acetylhomocystei~ethiolactone (Aldrich Chemicals) was added to the MIEP ~olution, and the mixture was incubated at room temperature for 16 hours. It was then twice dialyzed under N2 against 2 L of 0.1 M
borate buffer, p~ 9.5, containing 4 mM EDTA, for 24 hours at room temperature. The thiolated protein was then a~sayed for thiol content by Ellman's reagent (Sigma Chemicals) and the proteln concentration was determined by Bradf~rd reagent (Pierce Chemicals).
For conjugation of MIEP to P~P, a 1.5 fold excess (wt/w~) of bromoacetylated ~. influenzae serotype b PRP was added to the MIEP olution and the pE was adjusted ts 9 - 9.5 with 1 N NaO~. The mixture was allowed to incubate under N2 for 6 to 8 hour~ at room temperature. At the end of the reaction time, 25 ~L
of N-acetylcysteamine (Chemical Dynamics) was added to the mixture, and was allowed to stand for 18 hours under N2 at room temperature. The conjugate solution was acidified to between pH 3 to 4 with 1 N HCl, and centrifuged at lO,OGO x g for 10 minutes. l mL of ~ ~ C;'~ '7 ~
130/G~B - 114 - 18159IA

the supernatant wa~ applied directly onto a column of FPLC Superose 6B (1.6 x 50 c~, Pharmacia) and the conjugate was eluted with PBS. The void volume peak which contains the polysaccharide-protein conjugate ~PRP-MIEP), was pooled. The conjugate solution was then filtered through a 0.22 ~ filter ~or sterilization.
~XAMP~ 12 Demo~trat~on ~f Immun~g~nici~y of PR~IEP c~nj~te~
Immunizations: Male Balb/c mice (Charle River, Wilmington, MA) were immunized IP with PRP
co~alently conjugated to ~I~P a~ de~c~ibed in Example 11, using 2.5 ~g PRP in 0.5 mL o~ preformed alum.
Control mice were immunized with equivalent amou~ts ~ PRP given a6 PRP-CRM (2.5 ~g P~P16.2~ ~g C~M; 1/4 of the human do~e), ~P-DT (2.5 ~g PRP/1.8 ~g DT;
1/10 of the human dose), and PR~-OMPC (2.5 ~g PRP/35 ~g OMPC; l/4 of the human dose~.
Infant ~hesu~ monkeys, 6-13.5 week~ of age, were immunized with PRP-MI~P conjugates adsorbed onto alum. Each ~onkey receivet 0.25 mL of conjugate at t~o different 8ite6 of injection, ~or a total dose of 0. 5 ~L. The mon~eys were immunized on day 0, 28, and 56, and blood ~ample6 were taken every two to fou~
wee~s.
Antibody re~ponseR were mea~ured by the ELISA de~cribed in Example 9, which di~tingui~hes the cla~s and ~ubcla~s of the immuno~lobulin response.
An RIA which quantitateæ the total anti-PRP antlbody (see Example 9) was also used to evaluate the monkey re~ponse. PRP-MIEP conjugates were tested for immu-nogenicity in mice as well as infant rhesus monkeys.
Antibody responses were measured by ELISA and ~IA.

The results show that PRP-MIEP conjugates are capable of generating an immune response in infant Rhesus monkeys and mice, consisting of IgG
anti-PRP antibody and a memory response. This is in contrast to the PRP-CRM and PRP-DT which do not elict measurable anti-PRP antibody. Thus, MIEP functions as an immunologic carrier protein for PRP and is capable of engendering an anti-PRP antibody response when covalently conjugated to the PRP antigen.
Purified MIEP is therefore an effective immunologic carrier protein replacing the heterogeneous OMPC in construction of bacterial polysaccharide conjugate vaccines.

E~AMPLE 13 PREPARATION OF MIEP - cPND15 CONJUGATE: -To 10.5 mL of a MIEP solution (1.85 mg/mL, 19.4 mg total) contained in a 50 mL flask ~as added 2O6 mL of a 0.1 M, pH 11 borate buffer. The pH was 20 adjusted to 10.8 with 5N NaO~ after addition of 37 mg EDTA and 11 mg dithiothreitol. Then 34~ mg of N-acetylhomocysteine thiolactone was added and the p~
again adjusted to 11 with 5N NaO~. Thiæ solution was - degassed, the air replaced with nitrogen and the solution aged for 23 hours under an atmosphere of ~itrogen.
The sample was then dialized against 4L of p~9.5 borate containing 10 mL, EDTA for 7 hr; against a fresh 4L for 22 hrs and finally against a pH 9.5 O.OlM borate buffer containing 1.9 mg DTT for 16 hr~. This treatment afforded a ~olution that contained a total of 4.84 ~moles o~ thiol (by Ellman assay). This equates to 249 nanomoles S~/mg protein.

~Z~ 3~

A 10 mg sample of maleimidated cPND15 from Example 13 was dissolved in 1 mL of H20 and 50 ~L of this was used for a maleimide assay by the reverse Ellman method, to reveal 5.4 ~moles (total) of maleimide. A 0.~ mL (4.88 ~moles) ali~uot of the solution was added to the thiolated MIEP solution (pH
9.~), which immediately became turbid and after 3 hrs and 40 minutes no thiol titer ~by Ellman assay) remained.
lo The solution (14 mL) was dialyzed twice ~s 4L of a pH 9.5, 0.01M borate buffer for Z7.5 and 38 hrs respectively. An assay on 100 ~L for amino acid composition gave the following reæults:
nanàmoles/0.1 mL sample: norleucine 1~.9 ~-alanine 13.7 lysine 48.8 A Bradford protein assay on 100 ~L showed 0.95 mg/mL. Using a molecular weight of 1111, this translates as 176.7 ~g/mL of peptide. Thus the peptide to protein loading was 18.67..

E~AMPLE 14 PREPA~ATI~ OF MI~P-cPND31 ~NJUGAT~:
To 6.5 mL of a MIEP solution (1.7 mg/mL) was added 1.5 mL of a p~ 11, 0.1 M borate buffer and the pH adjusted to 11 with 5 ~L of 5N NaOH. To this was ~dded 21 mg of EDTA and 6.5 mg of DTT and solution was effected by tumbling for 15 min~ Then 200 mg of N-acetylhomocysteine thiolactone was added, the solution degassed and the air replaced by N2. After , , J ~j1 130/G~B - 117 - 18159IA

aging in the N2 box for 1.5 hrs., the pH was adjusted to 10.66 with 5N NaO~, the degassing process repeated, and ageing continued for 20.5 hrs.
The solution was dialyzed vs 4L of 0.lM
pH9.5 borate con~aining 0.01 M EDTA for 6.5 hr followed by 4L of 0.1~ pH 9.6 borate, 10 mM EDTA
containing 1 mg dithiolthreitol for 17 hr. An Ellman assay indicated 2.27 ~moles (total) of thiol which i6 equivalent to 205 nanomoles S~/mg protein.
To this thiolated protein solution was added O.55 mL of maleimidated cPND31 from Example 14 (3.77 ~moles/mg, by reverse Ellman assay, 2.07 ~moles total). An instant turbidity was ~oted. ~n additional 0.5 mg of maleimidated cPND31 was added and the mixture was aged for 1 hour.
To remove unconjugated peptide, the mixture was dialyzed in dialysis tubing, having a molecular weight exclusion limit of 12,000-14,000, vs 4L of pH
9.48 0.1M b~rate for ~.25 houræ and ~s 4L of pH 9.68 0.01M borate for 66 hrs. A total of 8 m~ of solution remained from which 200 ~L was removed for amino acid analysis:
norleucine 22.8 nanomoles/200 ~L .
lysine 85.9 nanomoles/200 ~L.

The solution was then dialyzed vs 200 mL of p~ 7.07 0.1 M phosphate buffer which was 5 M in urea, affording a final volume of 6.5 mL. A Bradford protein assay revealed 1.26 mg protein/mL (8.2 mg total). Thus, 0.912 ~moles peptide (8 mL X 22.8 nanomoles/0.2 mL) at a molecular weight of 1204 a 1.1 mg of peptide (total). Therefore, in thi~ case, a peptide to protein loadi~g of 13% was achieved.

~ 3 ~ f.

Solid Sta~ç_Svnthesis of Di~lfide-~onded cPND4:
A linear PND peptide was prepared on Wang resin using an ABI-431A peptide synthesizer, starting from Fmoc-L-Cys~Acm)-0-Wang resin (0.61 meq/gram).
Fmoc chemistry and Fmoc-Amino Acid symmetrical anhydrides (4X excess, prepared in situ) were used as reagents on a 0.25 mmole scale to generate 745 mg of the peptide:
Acm ~tr Fmoc-Nle-Cys-Hi~-Ile-Gly-Pro-Gly-Arg-Ala-Phe-Cys-O-Wang Re~in.

Trt Acm A solution of iodine in 5% methanol/anhydrous DME (1 ml) was added to the dried, derivatized Wang resin shown above and stirred at room temperature for 4 hours. The resin was filtered, washed with anhydrous DMF (5 ~ 2 ml), and finally resuspended in DMF (2 ml). Two drops of a 0.1 M solution of sodium thioæulphate in water were added, and stirred for a : 2~ few æeconds. The resin was washed with aqueous 95%
DMF (3 ~ 2 ml), anhydrous DMY ~2 ml), methylene chloride (3 x 2 ml), ether (3 ~ 2 ml) and dried.
The Fmoe and other protecting groups were : remo~ed by treatment with 20V/o piperidine in DME over 20 minutes, and the resin was washed and dried. The resin was cleaved from the diæulfide bonded cyclic peptide by treatment with 95% TFA/4~b ethane dithiol/1% thioanisole (1 ml) at room temperature ~or 6 hours. The solution ~as filtered, the resin washed with additional 100% TFA (3 x 1 ml), and the combined filtrate dried. Material that waæ insoluble in ether was removed by extraction (3 x 2 ml) and the solution redried.

, ,' -: ~ , :
, .

~ ~ f `s ~ J r~
,_ Preparative ~PLC using two 2.12 x 25 cm Vydac C18 reverse phase columns in series and a gradient elution of 20 to 24% C~3CN over 90~ allowed isolation of a sharp peak eluting at 36.~6' under these conditions. Analytical HPLC yielded a single peak upon co-chromatography of a known disulfide bonded cyclic standard with ~he product obtained from preparative HPLC. FAB-MS gave a [M+H]~ of 1171, which is consistent with the the disulfide bonded cyclic structure cPND4 (SEQ ID: 23:):

H-Nle-~ys-~is-Ile-Gly-Pro-Gly~Arg-Ala-~he-Cys-COOH
S C~2 1. Solution Synthe~i~ of Peptide ~onded cP~D15._ The linear peptide Cbz-Nle-Lys(Boc)-Gln-Arg(Mtr) Gly-Pro-Gly-Arg(Mtr)-Ala -Phe was synthesized following solid-phase methods on an ABI 431A peptide synthesizer using 373 milligrams (O.1 mmole~) of commercially available Fmoc-Phenylalanyl-p- alkoxybenzyl alcohol resin.
With the e~cepticn of norleucine, which was purchased in the benzylo~ycarbonyl ~Cbz) protected form, L-amino acids used were the fluorenylmethoxycarbonyl (Fmoc) derivatives having the a~propriate acid-la~ile side chain protecting groups. The polypeptide-derivatized resin product was transferred to a sintered glass funnel, washed with dichloromethane, and dried, to yield 0.6 g of polypeptide-re~in product.

rJ, ~

The peptide was cleaved from the resin by treatment with 6 ml of a 95:2:3 mixture of.TFA:1,2 ethanediol:anisole for 16 hours. The reaction mixture was filtered through a sintered glass funnel, the resin washed with 10 ml TFA, and the filtrates combined. Following conce~tration to about 1 to 2 ml of yellow oil, the linear peptide was recovered by trituration with 400 ml of diethyl ether, in 50 ml portions, and filtration on a sintered glass funnel.
Dissolution with 100 ml 1% TFA followed by lyophiliza~ion yielded 298 mg of linear pep~ide.
The peptide powder wa~ dissolved in 800 ml DMF, neutralized with 0.42 ml diisopropylethylamine, and treated with 0.077 ml diphenylphosphorylazide.
The solution was stirred in the dark for 70 hours at 4OC to allow formation of the cyclic lactam. After quenching by addition of 3 ml glacial acetic acid, the react~.on mixture was concentrated to about 1 to 2 ml of oil, dissolved in 10% aqueous acetic acid~ and lyophilized.
The cyclic peptide was purified by G-15 size exclusion chromatography using 5% acetic acid as the mobile phase. Fractions, monitored by W detection, cont~ining the peptide were pooled a~d lyophilized to yield 135 mg of dry cyclic peptide. All result3 obtained were consistent with the structure ePND15:

D ~
Z-Nle-C-N-L\ys-Gln-Arg-Gly-Pro~Gly-Arg-Ala-Phe (OC)C\ 2 ~=
H2C ,N~I -- - C ( ) ~3[2 ~2 ,, , which may also be represented as:

Z-Nle-Lys-Gln-Arg-Gly-Pro-Gly-Arg-Ala-~he (C~2)4 N C=0 ¦ () 2. Deprotection Qf cPND15 to yield the hydrogen ~orm:
Deprotection of cPND15 was achieved by dissolving the cyclic peptide in 20 ml of 30% aqueous acetic acid and hydrogenation at 40 pæi for 16 hours o~er 100 mg of 10% palladium o~ carbon. The reaction mixture was filtered over celite to remove the catalyst, and the filtrate was lyophilized. Reverse phase HPLC using a Vydac C~8 semi-prep column was utilized to obtain ~.5 mg of pure deprotected cyclic peptide. This method of deprotection is-applicable to all peptides æynthesized as the b~nzylo~ycarbonyl N-protected peptide, to yield the fre@ hydrogen form f the peptide which may now be acti~ated at the amino terminus ln preparation ~or conjugation. The structure of the product wa~ confirmed by FAB-MS, analytical ~PLC and amino acid analysis, and all results were consistent with the structure cPND15:

O ~ ~
H-Nle~ -Lys-Gln-Arg-Gly-Pro-Gly~Arg-Ala-~he (a)~2: );~=0 ~2C
2 ~2 which may also be represented as:
H-Nle-L ~ ln-Arg-Gly-Pro-Gly-Ar~-Ala-Phe (CH2~ =0 H

~ XAMPLE 17 Synthesis of cPND31:
Two grams ~0.6 meq/gram) of Fmoc-Phe-~a~
resin was loaded on an ABI 431A synthesizer. Fmoc : single coupling protocols were used to add Fmoc-Ala, Fmoc-Arg(Tos)~ Fmoc-Pro, Fmoc-Ile, Fmoc-His(Trt), : Boc-Lys(Fmoc), and Cbz-Nle to produce 3.7 grams of linear peptide resin:ha~ing the sequence:
Boc-Lys(N~-Z-Nle)-Hi~(Trt)-Ile-Gly-Pro-Gly-Arg(Tos)-Ala-Phe.
The peptide was cleaved from the resin by treating with 95Vb TFA, 5% water for two hour~. The 2~ resin was removed by fil~ration, the TFA removed from the filtrate by evaporation in vacuo, and the residue was-triturated with diethyl ether. The precipitate wa3 recovered by ~iltration and drying to yield 1.7 grams of linear peptide having the sequence:
H-Lys(NE-Z-Nle)-~i6-Ile~Gly-Pro-Gly-Arg(Tos)-Ala-Phe.
The peptide was treated with Bo~-isoglutamine-ONp (0.71 grams, 2 nmoles,) and DIEA ..
(0.35 ml, 2 mmoles) in~DMF (10 ml) overnight at room temperature. The DME was evaporated, and the residue treated with diethyl ether. The precipitate was recovered by filtration and ~ashed wi~h ethyl acetate. The dried peptide (l.9 gramg) was treated rl ~ ~

with TFA (lO0 ml) for 0.5 hours. The TFA was evaporated in vacuo, the residue triturated with diethyl ether and the precipitate was reco~ered by filtration and dried.
The peptide was desalted on Sephadex G-lO in 10% aqueous acetic acid as the eluent. Peptide fractions were lyophilized to yield 1.2 grams (0.79 mmoles) of:

H-isoGln-Lys(NE-Z-Nle)-His-Ile-Gly- Pro-Gly-Arg(Tos)-Ala-Phe Two batches (0.55 gm, 0.36 mmoles) of the peptide were separately dissolved in lO00 mL ice cold DMF and DIEA (0.16 mL, 0.9 mmoles) and DPPA (0.12 mL
were added and the solutions were stirred overnight at room temperature. The DME ~as evaporated in vacuo and the residues combined and solubilized in CHCl3.
The organic fraction was washed with 5% aqueous citric acid, then dried over MgSo4 and evaporated to yield 0.78 gm of crude cyclic peptide. This material was treated with liquid ~F (lO mL) containing anisole (l mL) ~or two hours at 0C. The ~F was evaporated and the residue was purified by graidien elution on reveresed phase HPLC (Vydac C l8, 0-50% CH3CN, over 50 minutes using O.l % aqueous TFA as the buffer~ to give 250 mg of pure cPND31 (M~=1204).

H ~ 0 ~-Nle-N(CH2)~ ~C-His-Ile-Gly-Pro-Gly-Arg-Ala-Phe (~) H~C~C~ CH2C~N C=O
~) 2~NOC k 130/G~B - 124 - 18159IA

Pr~paration of MaleimidQPropion~l-cPND15:
10 milligrams of cPND15 trifluoroacetate salt was dissolved in 0.3 ml of a 1:2 mixture of H20:MeCN. The solution was cooled in an ice bath and then 100 ~L of 0.345 M NaHC03 solution, followed by 3.5 mg of maleimidopropionic acid N-hydroxysuccinimide eæter, was added. The reaction was allowed ~o proceed with stirring for one hour, lo followed by quenching with 3 ~L of tri$1uroacetic acid. The reaction mixtur was fil~ered ~hrough a 0.2 micron filter, and the filter was washed with 0.2 ml of wate~. The combined filtrates were injected onto a 2.15 X 25 cm Vydac C18 reverse phase column.
: 15 The column was eluted i~ocratically for 10 minutes at a flow rate of 10 ml/min. with 25% MeCN/0.1% TFA, followed by gradient elution from 25 to 40% MeCN/0.1%
TFA, over 20 minutes. The product eluting between 20 and 32 min was concentrated and lyophilized, yielding 7 mg of the trifluoroacetate salt of maleimidopropionyl-cPND15 aæ a white amorphous powder. FAB-MS revealed a major ion (M+H) at 1262.
Titration for maleimide by Ellman assay ~uenching gave a concentration of 0.54 ~moles per mg o~ the 2s maleimidopropionyl-cPND15.

Preparation of Maleimidopropionyl cPND31:
Following the procedure o~ ~xample 13, 37.6 mg of the trif luoroacetate ~aIt of cPND31 waæ reacted wi~h 8.3 mg of maleimidopropionyl N-hydroxy-,7 succinimide ester in 0.4 ml of a 0.322 M NaHC03 solution and 1.2 ml of 1:2 H20:MeCN, followed by quenching with 10.5 ~1 of TFA. Preparative HPLC (30%
MeCN/0.1% TFA isocratic for 10 minutes followed by gradient elution from 30-50% MeCN over 5 min gave a product peak eluting between 18-25 min. The lyophilized product weighed 26 mg, and the maleimide titer was 0,57 ~M/mg. FAB-MS gave a major ion (M+H) at 1356. Amino acid analy~iæ gave Nle=460, lo ~-alanine=420 and Lys=460 nmoles/mg.
NMR analysis ga~e a singlet at 6.93 ppm (maleimide H).

Protocol for Inoculation of Animals with the MIEP-cPND15 and MIEP-cPND31 conju~ate of thi~ InventiQn:
Alum was used as an adjuvant during the inoculation series. The inoculum was prepared by dissolving the conjugate in phy~iolo~ic saline at a final conjugate concentration of 300 ~g/ml.
Preformed alum ~aluminum hydroxide gel) was added to the solution to a flnal level o$ 500 ~g/ml aluminum.
The conjugate was allowed to adsorb onto the alum gel for two hours at room temperature. Follo~i~g adsorption, the gel with the conjugate was washed twice with physiologic saline and resuspended in saline tG a protein concentration of 300 ~g/ml.
A~rican green monkeys were individually inoculated with three 300 ~g doses or three 100 ~
doses of the conjugate adsorbed onto alum. Each dose was injected intramuscularly. The doses were delivered one month apart ~week 0, 4, 8, 28). The ~, ~" ,,, " ,,-, ~

animals ~-ere bled at intervals of two weeks. Serum samples were prepared from each bleed to assay for the development of specific antibodies as described in the subsequen~ examples.

~ XAMPLE 21 Analysis of Sera.for ~nti-Peptide IgG Anti~odies:
Each serum sample is analyzed by enzyme-linked immunoadsorbent assay (ELISA).
Polystyrene microtiter plates were coated ~ith 0.5 ~g per well of the synthetic peptide (not conjugated to MIEP) in phosphate-buffered physiological saline (PBS) at 4C. Each well was then washed with PBS
containing O.05% TWEEN-20 (PBS-T). Test serum, diluted serially in PBS-T, was added to the peptide-containing wellæ and allowed to react with the adsorbed peptide for one hour at 36C. After ~ ~ashing with PBS-T, alkaline pho~phatase-conjugated goat anti-human IgG was added to the test well~ and was allowed to react ~or one hour at 36C. The ~ells were then washed extensively in PBS-T. Each well received 0.1% p-nitrophenyl phosphate in 10%
- diethanolamine, p~ 9.8, containing 0.~ mM
MgC1~6H20. The ensuing reaction was allowed to proceed at room temperature ~or 30 minute~, at ~hich time it was terminated by the addition of 3.0 N NaOH.
The greater the inteIaction of antibodies in ~he test serum with the peptide substrate, the greater is the amount of alkaline phosphatase bound onto the well. The pho~phatase enzyme mediates the breakdown of p-nitrophenyl phosphate into a molecular substance which a~sorbs light at a wavelength of 40 nm. Hence, there exists a direct relationship between the absorbance at 405 nm of light at the end of the ELISA reaction and the amount of peptide-bound antibody.
All the monkeys inoculated with the maleimidopropionyl-cPND15-MIEP and malemidopropinyl cPND31-MIEP conjugates developed anti~odies specifically capable of binding the peptide.

Analysis of Sera for Acti~ity which Specifically Neutralizes ~IV Inf~Gtivity: _ -Virus-neu~ralizing activity is determined with an assay described by Robertson et al., J.
Virol. Methods 20: 195-202 (1988). The assay mea~ures specific HIV-neutralizing activity in test serum. The assay is based on the observation that MT-4 cells, a human T~lymphoid cell line, are readily suscepti~le to in~ec~ion with ~I~ and, after a period of virus replication, are killed as a reæult of the infection.
The test serum is treated at 56C for 60 minutes prior to the aæsay. This treatment is required to eliminate non-specific inhibitors of HIV
replication. Heat treated serum, serially diluted in RPMI-1640 cell culture medium, i8 mi~ed with a standard infection dose of HIV. The dose is determined priox to the assay as containing the smallest quantity of virus required to kill all the MT-4 cells in the assay culture after a period of 7-8 days. The ~erum-virus mi~ture is allowed to interact ~ r~ ~
130/GHB - 128 - i8159IA

for one hour at 37C. It then is added to l.O x 105 MT-4 cells suspended in RPMI-1640 growth medium supplemented with 10% fetal bovine serum. The eultures are incubated at 37C in a 5% CO2 atmosphere ~or 7 days.
At the end of the incubation period, a metabolic dye, DTT, iB added to each culture. This dye is yellow in color upon visual inspection. In the presence of live cells, the dye is metabolically processed to a molecular species which yields a blue visual color. Neutralized HIV cannot replicate in the target MT-4 cells and there~ore does not kill the cells. Hence, positi~e neutralization is asseseed by the development of blue color following addition of the metabolic dye.
E~ LE 23 Preparation of a cyclic disulfide for conjugation: ~

1. PREPARATION OF c~ND3~ (SEQ ID: 22:~:

H Nle Cys Tyr Asn Lys Arg Lys Arg Ile His Ile Gly Pro Gly Arg Ala Phe Tyr Thr Thr Lys Asn Xle 2s Ile Gly Cys-O~ (Cl3s~220N42O33s2 formula wei~ht = ~P2 The 26mer was assembled on the Milligen ~
9050 synthesizer, ~tarting from partially racemised Fmoc-L-Cys(Trt)-OPKA resin (Milligen batch B 090426, 0.081 meq/g), using 2.47 g (0.200 meq). The theoretical yield i6 604 mg. The resin was mixed 130/G~B - 129 ~ IA

with an equal volume of glass beadæ (Sigma 150-212 ~m). The mixture completely filled tWO 1 X 10 cm columns, connected in series. Reagents were Fmoc-Pfp ester (except for threonine, which was d~Bt), usin~
four fold molar excess in N-methyl pyrrolidine solvent. Side chain protection was: Tyr (tert-butyl); Lys (Boc); Arg (Mtr); His (Boc); Thr (tert-butyl); Cys (Trt). The protocol was modified to give double coupling with LYS~; Ile9: Ilell;
Glyl2; prol3; Glyl4; Argl5; Phel7; Tyrl8; Thrl9;
Thr20; Ile23; Ile24. Acylation recycle times were extended from 30 to 60 mminutes for all units, except for Glyl4 and Alal6s and to 90 minutes for Ile9 (2x);
Ilell (2x~, Ile23 (2x) and Ile24 (2x). The derivatized resin was maintained as the free terminal amine which wa~ washed with C~2C12 and air-dried.
The mixture of dry derivatized resin and glass beadæ was resuæpended in 95% TFA, 4% ethane dithiol, 1% CH3SPh (30 mL) at 23C in a sealed flask, 20 with gentle stirring on an oscillating tray for 8 hours. The bright yellow mixture was ~he~ filtered and the insolubles were thoroughly extracted with 100% TFA ~3 x 20 mL). The combined dark orange fil~rate~ ~ere evaporated to give a pale tan, oily gum. On trituration with ether (20 mL~ thi~ material instantly became a colorless solid, which was transferred to a filter by triturating with additional ether (3 x 20 mL). After drying, the crude product was obtained as a fine colorless powder (583 mg)-130/G~B - 130 - 18159IA

Analytical reverse phase HPLC ~aqueous 0.1%
TFA/2~% CH3CN, ~ = 215 nm, A = O.05, 2.0 mL/min.] on a 0.46 x 25.0 cm Vydac C18 column of about a 50 ~g sample, dissolved in 50 ~L aqueous 0.1 % TFA/20~/o CH3CN, 4 ~L injected, revealed a major component (36.29') and a later eluting minor component. These were separately collected after injection of a 30 mg and another 50 mg aliquot of the sample onto two 2.21 x Z5.0 cm preparative Vydac Cl8 columns in series lo ~linear gradient over 60': 0.1% TFA/23-27% CH3CN, ~ =
215 ~m, A = 3.00, 10 mL/min]. A total of 35.2 mg of the earlier eluting material (44.45') and 8.2 mg of the later eluting material was recovered following lyophilization. FAB-MS of the major product gave a ~M+H]~ = 302?.1 and an tM+Na]~ = 3044.2, which i~
consistent with the calculated mass.

2. PR~PARATION OF T~E CYCLI~ DI~LFIDE: (S~Q ID: 22):

H-Nle ~ys Tyr Asn Lys Arg Lys Arg Ile ~is Ile Gly Pro / Gly Arg Ala Phe Tyr Thr Thr Lys Asn C\2 Ile Ile Gly C~s-OH
. S -- S~2 : 25 The linear 26 mer dithiol compound (35.~ mg~
was disæol~ed in degassed distilled water (38 mL) at 23 C to give a clear colorless solution at pH 2.73.
The pH was adjusted to 8.5 with 0.1 N NH40H, and the solution was covered with an atmosphere of nitrogen.
An aliquot of the material was immediately run on : analytical reverse phase ~PLC and found to be undergoing o~idation as evidenced by the appearance of an early peak.

, With magnetic stirring, a freshly prepared solution of 0.01 M K3Fe(CN)6 was added by power driven hypodermic syringe at 23O C under nitrogen.
Analysis of a small aliquot by ~PLC revealed total conversion of starting material to an earlier elution time. The reaction mixture (pH 8.3) was mixed with 10% aqueous acetic acid and stirred to give a p~ of 4Ø The solution was filtered to remove insoluble material, and the faintly yellow solution was evaporated and then lyophilized to give about 27.9 mg of a pale yellow powder. The material was dissolved in 0.1% TFA/20~/o C~3CN and gradient eluted on a preparative ~PLC. A major early eluting peak and a later eluting peak (4:1) were separately collected and lyophilized to yield 6.1 mg of the early and 1.5 mg of the late eluting material. FAB-MS analysis of the early eluting material: tM~H]+ 3019.7; [M~Na]+
3042.5; FAB-MS analysis of the late eluting material: [M+~]+ 3020.0; [M+Na]~ early material =
3041.5; all of which corresponds to the correct mass for the cyclized cPND33. The later eluting material is the D-cysteine carboxy terminu3 diastereomer.
Amino acid analysis of the products gave the predicted amino acid eompositions for the cyclized products and confirmed that the later eluting material is the D-cysteine containing diaætereomer.

b. AIR OXIDATION:
The linear 26 mer prepared in (1) above (86 mg, 28.4 ~mole~) was dissolved in aqueous 0.1%
TFA/20% acetonitrile (284 mL) at 23 C and the solution was allo~ed to st~nd open to the air.

' rl ~ ~' 6' Cyclization was monitorcd by re~erse phase ~PLC and the sample was found to be almost comple~ely converted to the early eluting material, with almost complete dissappearance of starting linear material, by t = 24 hours. The clear, colorless solution was evaporated to about 8 mL at which point an additional 10 mg sample prepared in the same way as the 86 mg, was added. The combined sample was evaporated to about 9 mL. The cloudy colorless solution uas lo subjected to HPLC separation, in ~wo separatc runs, on two 2.12 x 25.0 cm Vydac C18 column~ in series.
Two fraction~ were separately collected, an early eluting peak and a later eluting peak. Each peak was separately evaporated and lyophilized to yield 30.1 l~ mg and 9.7 mg of the early and late materials respectively. The early eluting mat~rial was combined with other preparatisns of early eluting cyclized material to yield a total of 47.5 mg of a faintly bluish fluffy powder. Analytical EPLC of this material gave a single peak.

3. PR~PARATION OF 3-MALEIMIDOPROPIONIC A~ID AN~YDRIDE

3~Maleimidopropionic acid (226 mg) was covered with ~ mL of acetic anhydride and the mixture was heated at 130C for 3.75 hr, and then aged over night at room temperatue. The solution was concentrated to an oil and the NMR spectrum (CDC13) indicated a mixtuxe of the homoanhydride and the mixed anhydride of acetic and maleimidopropionic acids. The ~tarting acid shows the methylene adjacent to the carbonyl as a triplet centered at 2.68 ppm whereas in the anhydride these resonances appear at 2.81 ppm. Purification was effected by fractional sublimation, first at 70C and O.2 mm and then at 120C and 0.2 mm. The latter fraction was removed from the apparatus by dissolving in CDC13, affording 34 mg of pure homoanhydride on evaporation of the solvent. This was recrystallized from CDC13 and cyclohexane affording material melting at 143-147C.
Calcd. for C14 H12N2O7: C,52,51;H,3.7~;N,~.75-Found: C,51.73;H3.67;N,8.16. 200 M~z NMR
(CDC13):2.83 (2H,t)3.84 (2H,t~,6.73 ~2H,s).

4. "SELECTIVE" ACYLATION OF ~PND33 cPND33 (22.5 mg; at estimated 70% peptide is equivalent to 15.75 mg or 5.212 micromoles) was dissolved in 12.0 mL of a O.lM pH 5.25 morpholinoethane sulfonic acid buffer and cooled in an ice bath. Analysis of this solution and progress of the reaction was ~ollowed by HPLC on a 25 cm ODS
column using 25% aqueous acetonitrile:-0.1%
trifluoroacetic acid (TFA) as eluent.
Maleimidopropionic acid anhydride (2.0 mg, 6.25 micromoles) wa dissolYed in 0.600 mL of dry tetrahydrofuran, and 0.5 mL of this æolution (corresponding to 5.2 micromoles of anhydride) was added to the above peptide solution. After 30 sec., a 7 microliter aliquot was removed and evaluated by HPLC. This assay was repeated at O.25, 0.50, 1.25, 2.25 and 3.0 hr. After 3.5 hr the solution was J r~

lyophilized. The lyophylizate was dissolved in 2.0 mL of 20% aqueous acetonitrile, filtered through a 0.2 micron filter and preparatively chroma~ographed in three 0.700 mL runs on a 21.2 mm x 25 cm Zorbax C-18 column~ The following elution program was used:
flow rate = 10 mL/min; isocratic elution with 25%
aqueous acetonitrile/0.1% TFA (12 min); gradient to 28% acetonitrile (10 min); gradient to 35%
acetonitrile (8 min). The tail fractions were isolated by co~centration and lyophilization to afford 8.9 mg of recovered starting material (penultimate fraction) and 9.6 mg of a product which had a mass spectrum ~FAB) indicatiing a molecular weight of 3172 (i.e the mono-maleimidopropionyl derivative of cPND33).
The product was further characterized by a sequence analysis looking for the absence of lysine (the absence of any sequence would imply terminal amino acylation). The results indicate that most but not all of the maleimidopropionyl moiety is bonded to the ly~ine closest to the carboxy terminus.
While the ~oregoing specification teaches the principles of the present invention, with examples provided for the purpose of illustration, it will ~e understood that the practice of the invention encompasses all the usual variations, adaptations, modifications, or deletions as come within the scope of the following claims and its equivalents.

~ ~ ~ g r, rJ r~

SEQUENCE LISTING

(1) GENERAL INFORMATION:

li) APPLICANT: Oliff, Allen I
Liu, Margaret A
Friedman, Arthur Tai, Joseph Y
l O Donnelly, John J
(ii) TITLE OF INVENTION: THE CLASS II PROTEIN OF THE OUTER
MEMBRANE OF NEISSERIA MENINGITIDIS HAVING IMMUNOLOGIC
CARRIER AND ENHANCEMENT PROPERTIES, AND VACCINES
l 5 CONTAINING SAME
(;;;~ NUMBER OF SEQUENCES: 24 (;v) CORRESPONDENCE ADORESS:
2 0 (A) ADDRESSEE: Merck & Co., Inc.
(B) STREET: P.O. Box 2000 (C) CITY: Rahway ~D) STATE: New Jersey (E) COUNTRY: USA
2 5 (F) ZIP: 07065 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible 3 0 (C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #l.O, Version #1.25 - ' ~

(vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER:
(B) FILING DATE:
(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: Pfeiffer, Hesna J
(B) REGISTRATION NUMBER: 22640 (C) REFERENCE/DOCKET NUMBER: 18159IA

ix) TELECOMMUNICATION INFORM4TION:
(A) TELEPHONE: 908-594-4251 (B) TELEFAX: 908-594-4720 (2) INFORMATION FOR SEQ ID NO:l:

(;) SEQUENCE CHARACTERISTIES:
(A) LENGTH: 41 amino acids (B) TYPE: amino acid (D) TOPOLOGY: li near (ii) MOLECULE TYPE: peptide - 25 (iii) HYPOTHETICAL: NO

~v) FRAGMENT TYPE: interna1 :

30 (ix) FEATURE
~A) NAME/KEY: Disulfide-bond (B) LOCATION: 3..38 ~ ,~ r~ ~ ~" ~, ~, (xi ) SEQUENCE DESCRIPTION: SEQ ID NO:1:

Ile Asn Cys Thr Arg Pro Asn Asn Asn Thr Arg Lys Ser Ile Arg Ile 1 5 lO 15 Gln Arg Gly Pro Gly Arg Ala Phe Val Thr Ile Gly Lys Ile Gly Asn Met Arg Gln Ala His Cys Asn Ile Ser (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:
(A~ LENGTH: 24 amino acids (~) TYPE: amino acid ( D ) TOPOLOGY: 1 i near ( i i ) MOLECULE TYPE: peptide ~0 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:
:
Tyr Asn Lys Arg Lys Arg Ile His I1e Gly Pro Gly Arg Ala Phe Tyr : l 5 lO 15 Thr Thr Lys Asn Ile Ile Gly Thr .

130/GHB - 138 ~ 18159IA

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi~ SEQUENCE DESCRIPTION: SEQ ID NO:3:

Asn Asn Thr Thr Arg Ser Ile His Ile Gly Pro Gly Arg Ala Phe Tyr Ala Thr 61y Asp Ile Ile Gly Asp Ile 2 0 (2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino ac;ds (B) TYPE: amino acid 2 5 ID) TOPOLOGY: linear l;i) MOLECULE TYPE: pept;de (xi) SEQUENCE DESCRIPTION: sEq ID NO:4:

Asn Asn Thr Ary Lys Ser Ile Arg Ile Gln Arg Gly Pro Gly Arg Ala Phe Yal Thr Ile Sly Lys Ile Gly Asn (2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:
tA) LENGTH: lB amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide lxi) SEQUENCE DESCRIPTION: SEQ ID NO:5:
.
Arg Ile Gln Arg Gly Pro 61y Arg Ala Phe Val Thr Ile Gly Lys Ile Gly Asn 2 0 (2) INFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 11 am;no acids (~) TYPE: amino acid 2 5 (D) TOPOLOGY: linear (ii) MGLECULE TYPE: peptide :
: 3 0 (xi) SEQUENCE DESCRIPTION: SEO, ID NO:6:

Arg Ile Gln Arg:61y~Pro Gly Arg Phe Val Thr .

, : .
.:
. , : ` :

J ~ J

(2) INFORMATION FOR SEQ ID NO:7:

(;) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 8 amino atids (B) TYPE: amino acid ~D) TOPOLOGY: li near (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:

His Ile Gly Pro Gly Arg Ala Phe l 5 (2) INFORMATION FOR SEQ ID NO:8:

~i) SEQUENCE CHARACTERISTICS:
2 0 (A) LENSTH: 6 amino acids (B) TYPE: amino acid ~D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide :
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:

3 0 61y Pro Gly Arg Ala Phe , (2) INFORMATION FOR SEQ ID NO:9:

(i) SE~UENCE CHARACTERISTICS:
(A) LENGTH: 9 am;no acids (8) TYPE: amino acid (D) TOPOLOGY: linear (i;) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

Ile Gln Arg Gly Pro Gly Arg Ala Phe (2) INFORMATION FOR SEQ ID NO:10:

(;) sEquENcE CHARACTERISTICS:
~A) LENGTH~ 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (;i) MOLECULE TYPE: peptide ' (%i) SEQUENCE DESCRIPTION: SEO lD NO:10:

3 0 Ile Tyr Ile Gly Pro Gly Arg Ala Phe :..

~ 3 (2) INFORMATION FOR SEQ ID NO~

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide ~xi) SEQUENCE DESCRIPTION: SEQ ID NO:ll:

Ile Ala Ile Gly Pro Gly Arg Thr Leu l 5 (2) INFORMATION FOR SEQ ID NO:12:

(i) SEQUENCE CHARACTERISTICS:
2 0 (A) LENGTH: 9 amins ac;ds (B) TYPE: amino acid (D) TOPOLOGY: linear : (ii) MOLECULE TYPE: peptide 2 ~

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:

3 0 Val Thr Leu Gly Pro Gly Arg Val Trp j r~

(2) INFORMATION FOR SEQ ID NO:13:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids S (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:

Ile Thr Lys Gly Pro Gly Arg Val Ile (2) INFORMATION FOR SEQ ID NO:14:

(i) SEQUENCE CHARACTERISTICS:
2 0 (A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: pept;de (xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:

3 0 Thr Pro Ile Gly Leu Gly Gln Ser Leu (2) INFORMATION FOR SEQ ID NO:15:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids ~B) TYPE: amino acid ~D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (x;) SEQUENCE DESCRIPTION: SEQ ID NO:15:

Thr Pro Ile Gly Leu Gly Gln Ala Leu l 5 (Z) INFORMATION FOR SEQ ID N0~16:

li) SEQUENCE CHARACTERISTICS:
2 0 (A~ LENGTH: 9 amino acids (B) TYPE: amino acid : (D) TOPOLOGY: llnear : (ii) MOLECULE TYPE: peptide : ' (xi) SEQUENCE DESCRIPTION: SEQ ID NO:16:

lle His Phe Gly Pro Gly Gln Ala Leu - 130/GHB - 145 - . 18159IA '' ~2) INFORMATION FOR SEQ ID NO:17:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 9 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (~i) SEQUENCE DESCRIPTION: SEQ ID NO:17:

Ile Arg Ile Gly Pro Gly Lys Val Phe (2) INFORMATION FOR SEQ ID NO:lB:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 4 amino acids (B) TYPE: am;no acid (D) TOPOLOGY: b~th : .
, : : (ii) MOLECULE TYPE::peptide ~ : 2 5 -: :
xi) SEQUENCE DESCRIPTION: SEQ ID no:l8:

Gly Pro Gly Arg . 130/GHB - 146 - 18159IA

(2) INFORMATION FOR SEQ ID NO:l9:

(;) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 4 amino acids (B) TYPE: amino acid (D) TOPOLOGY: both (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l9:

Gly Pro Gly Lys (2) INFORMATION FOR SEQ ID NO:20:

(i) SEQUENCE CHARACTERISTICS:
2 0 (A) LENGTH: 4 amino acids (B) TYPE: amino acid (D) TOPOLOGY: both (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:20:

Gly Pro Gly Gln ,-.

f v of (Z) INFORMATION FOR SEQ ID NO:21:

~i) SEQUENCE CHARACTERIST.CS:
(A) LENGTH: 4 amino acids (B~ TYPE: amino acid (D) TOPOLOGY: both ~ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:21:

Gly Leu Gly Gln (2) INFORMATION FOR SEQ ID NO:22:

(i) SEQUENCE CHARACTERISTICS:
2 0 (A) LENGTH: 26 amino ac;ds (B) TYPE: amino acid (D) TOPOLOGY: both .
(ii) MOLECULE TYPE: peptide .

(ix) FEATURE:~
(A) NAME/KEY: Mod i f i ed-s i te ~ -(B) LOCATION: 1 3 0(D) OTHER INFORMATION: /label= Nle /note= "norleucine"

~ " ~ ~j r~ J

(ix) FEATURE:
(A) NAME/KEY: D;sulfide-bond (B) LOCATION: 2..26 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:Z2:

Leu Cys Tyr Asn Lys Arg Lys Arg Ile His Ile Gly Pro Gly Arg Ala 1 5 ~O 15 Phe Tyr Thr Thr Lys Asn Ile Ile Gly Cys (2) INFORMATION FOR SEQ ID NO:23:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 11 amino ac;ds (B) TYPE: am;no acid (D) TOPOLOGY: both (;;) MOLECULE TYPE: pept;de (ix) FEATURE:
2 5 (A) NAME/KEY: Modified-s;te (B) LOCATION: 1 (D) OTHER INFORMATION: /label- Nle /note= "norleucine"
:
: 3 0 (x;) SEQUENCE DESCRIP7ION: SEQ ID NO:Z3:

Leu Cys His Ile Gly Pro Gly Arg Ala Phe Cys 1 5 lû

130~GHB - 149 - 18159IA

(Z) INFORMATION FOR SEQ ID NO:24:

(;) SEQUENCE CHARACTERISTICS:
(A) LE'JGTH: 4 amino acids (P) TYPE: amino acid (D) TOPOLOGY: both (i;) MOLECULE TYPE: peptide (x; ) SEQUENCE DESCRIPTION: SEQ ID NO:24:

Gly Pro Gly Val

Claims (4)

1. A vaccine for use in mammals, comprising the Class II protein of the outer membrane of Neisseria meningitidis serogroup B coupled to an antigen, which in mammals said vaccine will induce or enhance the formation of antibodies specific for said antigen.

2. The vaccine according to Claim 1 wherein the antigens are derived from bacteria, viruses, mammalian cells, fungi, rickettsia, allergens, poisons or venoms, synthetic peptides, and polypeptide fragments.

3. The vaccine, according to Claim 1, wherein the Class II protein of the outer membrane of Neisseria meningitidis serogroup B is a recombinant protein produced in a recombinant host cell.

4. A composition comprising an effective amount for immunization of mammalian species, of the polysaccharide/protein conjugate which comprises Haemophilus influenzae serotype B polysaccharide and the Class II outer membrane protein of Neisseria meningitidis serogroup B, and a pharmaceutically acceptable carrier.