CA2138765C - Recombinant haemophilus influenzae protein and nucleotide sequence encoding same - Google Patents

Abstract

Disclosed herein are immunogenic polysaccharide-H. influenzae adhesin protein conjugates, a purified H. influenzae adhesin protein and related proteins and polypeptides, DNA useful for producing the proteins, synthetic polyribosylribotol phosphate (PRP) oligosaccharides and intermediates useful for their synthesis, and methods of making and using these materials. The conjugates comprise a PRP fragment, preferably a synthetic oligosaccharide, coupled to an H. influenzae adhesin protein. The invention further comprises purified H. influenzae adhesin proteins and novel PRP oligosaccharides. The invention also comprises methods of producing these materials and using them in a vaccine to protect humans and other mammals against H. influenzae infection.

Description

RECOMBINANT HAEMOPHILUS INFLUENZAE PROTEIN
AND NUCLEOTIDE SEQUENCE ENCODING SAME
BACKGROUTJD OF THE INVENTION
This invention relates generally to vaccines against Haemophilus influenzae. In particular, it relates to a conjugate vaccine in which a synthetic oligosaccharide corresponding to a fragment of the polysaccharide capsule of H. influenzae type b has been coupled to an H.
influenzae adhesin protein. The vaccine may be used against both invasive and non-invasive H. influenzae infec-tion of humans, particularly very young infants, and other mammals.
H. influenzae (Hi;l are divided into two groups, those strains that possess a polysaccharide capsule and those that do not. The encapsulated strains are typed by a serological reaction of a capsule with reference antisera.
Types a-f have been identified. The non-encapsulated strains, which fail to react with any of the reference antisera, are known as non-typable.

V!',~! 94/00149 PC1'/US93/06016 Hi are a significant health problem worldwide. The type b strain (Hib) is the most virulent of the Hi strains, causing meningitis, acute epiglottitis, and other life-threatening infections in children five years old and younger. The mortality rate from type b meningitis is about 5%, even with the best modern antibiotic treatments, and neurological sequelae are observed in as many as 25-35%
of the survivors. In fact, bacterial meningitis caused by type b strains has been identified as the leading cause of acquired mental retardation in the United States. Thus, World Health Organization has made the development of an effective vaccine against Hib a priority.
Non-typeable Hi also cause various diseases, including pneumonia, bacteremia, meningitis, postpartum sepsis, bronchitis, sinusitis, conjunctivitis, and otitis media. The non-typeable Hi cause about 20-40% of all otitis media in children and young adults. Current therapy for chronic or repeated occurrences of otitis media generally involves antibiotic administration. Children may experience multiple infections because infection does not confer a lasting immunity.
A great deal of time, money, and effort has been spent trying to find a truly effective vaccine to H.
influenzae. The overwhelming focus has been on developing a vaccine for Hib because of its serious threat to very young children. Unfortunately, the approved type b polysaccharide vaccines are not effective for children WO 94/00149 ~ ~ ~ ~ PCT/US93/06016 under 18 months of age, which is the group most threatened by Hib.
It has been known for many years that antibodies directed against the type b capsule will protect individuals against invasive Hib infection, including meningitis. In a randomized, double-blind clinical trial in Finland, a type b polysaccharide vaccine was found to be 90% effective in presenting disease in children immunized between 24 and 72 months of age. However, the vaccine conferred no protective immunity in children younger than 18 months and provided only limited immunity in children aged 18-23 months. Peltola, et al., N. Enql. J. Med., 310:1561-1566 (1984). The type b polysaccharide elicits a T-cell-independent immune response, which probably accounts for the low immunogenicity in young children.
Based on these data, three type b polysaccharide vaccines were licensed in the United States in 1985 and were recommended for use in children aged 24-60 months.
These vaccines obviously have a major problem. They do not adequately protect~children under 24 months of age, the group most succeptible to H. influenzae disease.
There are other problems relating to the fact that the polysaccharide is obtained from natural sources.
Although purified, the polysaccharide fragments are of various lengths and, therefore, not as well characterized as desirable. This creates problems with respect to reproducibility and variable potency. Also, since naturally occurring polysaccharide must be isolated from a pathogen, safety concerns must be addressed with respect to both manufacture and use of the vaccine.
Attempts have been made to make the polysaccharide into a better immunogen. The polysaccharide or fragments thereof have been covalently coupled with various im-munogenic proteins, such as diphtheria or tetanus toxoids.
See, for example, U.S. Patent No. 4,673,574 issued June 16, 1987 to Anderson, U.S. Patent No. 4,808,700 issued Febru-ary 28, 1989 to Anderson, et al., European Patent Office Publication No. 0 245 045 dated November 11, 1987, and European Patent Office Publication No. 0 098 581 dated January 18, 1984.
Several of the conjugate vaccines have been shown to be safe and more immunogenic than the conventional polysaccharide vaccines in children, particularly infants.
The data suggest that the conjugate vaccines are functioning as T-cell dependent antigens. A T-cell-dependent response provides for a better overall immune response in a patient. One of the conjugate vaccines has been approved in the United States for children 15-18 months of age. Two of the conjugate vaccines have been licensed in the United States for infants as young as two months old.
The vaccines currently available to the medical practitioner have several major limitations. First, they do not protect against other Hi infections besides Hib.
The polysaccharide is not found in non-typable H.

WO 94/00149 ~ ~ ~ ~ ~ 6~~ PCT/US93/06016 influenzae: therefore, antibodies to it are non-protective against these strains. Second, they raise problems with respect to reproducibility, potency, and safety.
There are several avenues of on-going research on ways to overcome these limitations. One approach has been to develop procedures for Hib polysaccharide synthesis.
The Hib capsule consists of a linear homopolymer of alternating molecules of ribose and ribitol joined by a phosphodiester linkage represented by the following formula:
HO O ON
oti o OH
O ON O -P
i O n The polymer is known as polyribosylribitol phosphate and abbreviated PRP.
The PRP obtained from natural sources is crude degraded polysaccharide. It varies in molecular weight between 200 KD and 200,000 KD.

A few groups have been able to synthesize small PRP
oligosaccharides. For example, European Patent Office Publication 0 320 942 dated June 21, 1989, discloses the synthesis of synthetic PRP oligosaccharides of 2-20 units and their covalent attachment to immunogenic proteins, specifically tetanus or diphtheria toxins or toxoids. The oligosaccharides are linked to the proteins through a spacer. A phosphate triester synthetic procedure was used for the oligomerization. European Patent Office Publi-cation 0 276 516 dated August 3, 1988, also discloses synthetic PRP oligosaccharides 2-20 monomers in length, their conjugation to carrier proteins, and the use of the conjugates as vaccines against Hib. The oligosaccharides are prepared using the phosphotriester synthetic procedure for oligomerization. Both of these involved solution-type synthetic techniques for the preparation of the PRP oli-gosaccharides.
Elie, et al., Recl. Trav. Chim. Pays-Bas, 108:219-223 (1989), discloses the solid-phase synthesis of a PRP
hexamer. The units were coupled using a phosphate triester method and controlled-pore glass as the solid support.
The use of synthetic PRP fragments should provide several advantages over the PRP obtained from natural sources. Synthetic PRP is chemically well-defined and characterized. It would be of superior quality and less prone to produce side effects in humans. Its use would also obviate problems relating to reproducibility, potency, and safety associated with PRP obtained from natural sources. In addition, while the naturally occurring PRP is generally cross-linked to the protein carrier at random points along its chain, synthetic PRP can be conjugated through a single point, which creates less undesired epitopes.
This research promises improvements to existing vaccines, but there are still drawbacks. First, the PRP
synthesis is complicated and relatively inefficient. Thus, there is a need for improved synthesis procedures. Second, these improvements will be limited to vaccines against Hib.
Another approach has been to focus on the protein.
There are some available data suggesting that the protein and the carbohydrate parts of the conjugate vaccines act as independent immunogens. Therefore, the choice of the protein component becomes important in seeking to enhance immunogenicity. It would be more desirable to have an immunogenic protein or polypeptide derived from H.
influenzae as the protein component rather than a "nonsense" protein.
At least one group has conjugated an Hib outer membrane protein to PRP fragments. See European Patent Office Publication No. 0 338 265, dated October 25, 1989.
This application discloses 38 and 40 KD outer membrane proteins of Hib and their isolation and purification. The two proteins are quite similar. They are known as protein 2 (P2) or protein b/c because they occur as a doublet. The molecular weight depends upon the strain from which they are obtained. They are cross-reactive, have very similar amino acid compositions, and have the same amino and carboxy terminal sequences. The proteins are coupled to PRP fragments by reductive amination. The PRP fragments are obtained from naturally occurring PRP using standard techniques. The application states that the carrier proteins themselves may confer immunity.
This approach also suffers from certain limita-tions. The outer membrane proteins may vary among Hib types or serotypes within a particular type. Granoff, et al., in S.H. Sell and P.F. Wright (ed.), Haemophilus Influenzae: Epidemiolocty, Immunology, and Prevention, (New York: Elsevier Biomedical (1982)). Therefore, a vaccine based upon a particular outer membrane protein may not be effective against the broader spectrum of pathogenic H.
influenzae bacteria and may not even be effective against all strains of Hib.
Others have focused on Hi proteins and peptides alone as vaccine candidates. For example, see PCT Publi-cation No. WO 90/02557, published March 22, 1990. This application discloses two antigenically related Hi outer membrane proteins with a molecular weight of about 16 KD.
It further discloses related fusion proteins and peptide fragments of the outer membrane proteins, methods of puri-fying the proteins, and methods of making them by genetic engineering. All of these are claimed to be useful as g immunogens in vaccines. Such vaccines will also have the drawbacks mentioned immediately above.
Clearly, there is a pressing need for a safe vac-cine that is effective against both invasive and non-invasive H. influenzae, particularly in infants 2-6 months old. This ideal vaccine would also be effective against a wide variety of strains within each of the two categories by eliciting antibodies against a determinant found on the surface of most or all strains of H. influenzae.
The present invention overcomes the limitations of the existing technology and meets that need. It provides a novel synthetic PRP conjugated to newly isolated and purified H. influenzae adhesin proteins.
The ability to use an adhesin protein in a vaccine against H. influenzae is extremely desirable. Because of the way they function, adhesin proteins are believed to be highly conserved among strains of a particular type of bacteria. This is because they are the protein molecules that mediate attachment by bonding bacteria to host cells, the initial step in the infection process. Thus, the adhesins would be expected to be present in all strains (both encapsulated and unencapsulated) of Haemophilus.
Therefore, the present vaccine would be effective against a broad array of types and strains of Hi. In addition, vaccines based upon adhesin proteins should be more ef-fective than those based upon other outer membrane proteins, even for those bacterial strains from which the outer membrane proteins are derived. Antibodies to the 2~~g~ 65 -lo-adhesin protein would prevent adherence of the bacteria to the tissue of the host animal. Adherence is the initial step in Hi infection. Stopping the infection at this point would be the best approach possible.
The novel PRP of the invention also has advantages over the existing technology. It is better defined and characterized, and it is of superior quality when compared to PRP obtained from natural sources. Also, it has been more efficiently produced than the synthetic PRP described above.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an im-munogenic oligosaccharide-H. influenzae adhesin protein conjugate and a method for making it.
Another object of the invention is to provide a vaccine for protecting a mammal against H. influenzae.
Yet another object of the invention is to provide a method of inducing an immune response to H. influenzae in a mammal.
A further object of. the invention to provide purified H. influenzae adhesin proteins.

WO 94/00149 ~ ~ ~ PCT/US93/06016 A still further object of the invention is to provide a purified polypeptide capable of eliciting an antigenic response to H. influenzae in an animal host.
Yet another object of the invention is to provide methods for producing purified H. influenzae adhesin proteins.
A further object of the invention is to provide DNA
coding for the adhesin proteins and derived polypeptides, vectors containing the DNA, microorganisms transformed by such DNA and vectors, and methods for preparing such materials.
A still further object of the invention is to provide a composition of matter consisting essentially of synthetic PRP oligosaccharides having the same number of monomeric units and a method of preparing the synthetic PRP.
Another object of the invention is to provide compounds useful as intermediates in the preparation of synthetic PRP and methods of preparing such compounds.
Additional objects and advantages of the invention will be set forth in part in the description that follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objects and advantages of the invention will be attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

21~g~ 6~

To achieve the objects and in accordance with the purpose of the invention, as embodied and broadly described herein, the present invention provides an immunogenic oligosaccharide-protein conjugate useful in a vaccine for protecting a mammal against H. influenzae. The conjugate is made up of a PRP fragment, preferably a synthetic oligosaccharide, coupled to an H. influenzae adhesin protein. Preferably, the oligosaccharides contain from 2-30 ribosylribitol phosphate monomers, and from 1-30 of such oligosaccharides are attached to the protein. In an alternative embodiment, the oligosaccharide is bound to a polypeptide that is an active site of tha adhesin protein.
Preferably, the conjugate is represented by the following formula:
D/ R - X
HO O O HO O .
OH
OH _ OH
H O ON O-~P O OH
O ~ m where m is 1-30, n is 2-30, R is (CH2)pCH2NH or (CH2CH20)pCH2CH2NHCSNH where p is an integer from 1-3, and 2f 3876 X is an H. influenzae adhesin protein or a fragment thereof containing an active site of the protein.
The vaccine comprises an immunologically effective amount of the conjugate in a pharmaceutically accepLaple carrier. Preferably, the vaccine also contains an adjuvant. The administration of the vaccine or conjugate to a human or other mammal induces a T-cell dependent protective immune response.
The invention further comprises an isolated and a purified H. influenzae adhesin protein and modified proteins and polypeptides derived from the adhesin protein, provided such derived proteins and polypeptides are immunologically cross-reactive with the adhesin protein.
Preferably, such derivatives are one or more epitopes of the adhesin protein. In a particularly preferred embodiment, the epitope is also a receptor binding site.
The proteins and polypeptides may also be used in vaccines without being conjugated to the synthetic PRP.
In one embodiment, the adhesin protein is a minor H. influenzae outer membrane protein with a molecular weight of about 41,000 daltons. In another preferred embodiment, the adhesin protein is an H. influenzae outer membrane protein with a molecular weight of about 47,000 daltons.
In one embodiment, the adhesin protein is purified from H. influenzae bacteria. Hi membranes are solubilized.
The solubilized material contains the adhesin protein.
This material is separated from the insoluble material and WO 94/00149 PCf/US93/06016 r contacted with receptors for the adhesin protein for period of time sufficient for the protein molecules to bind to the receptors. The receptors are attached to an insoluble solid support. As a result, the protein is separated from the solubilized material. The protein molecules are then removed from the receptors thereby being recovered in purified form.
In another embodiment, the adhesin proteins and related polypeptides of the invention are preferably recombinant proteins and polypeptides that have been produced through genetic engineering techniques. They are produced by an appropriate host cell that has been transformed by DNA that codes for such proteins or polypeptides.
An isolated or substantially pure DNA sequence that codes for the adhesin proteins of the invention is obtained as follows. Adhesin protein receptors or antibodies to the adhesin, preferably monoclonal antibodies, are used to screen a genomic library containing H. influenzae DNA. The library is made of~clones which contain different sequences of the DNA which have been operably and recoverably inserted into a vector, with each of the vectors containing only one sequence of the DNA. The monoclonal antibodies or receptors identify the clones that produce the adhesin.
The clone is then isolated. Preferably, the exogenous DNA
sequences are recovered from the clone.
The invention further comprises isolated or substantially purified DNA derived from this DNA, for example, by single or multiple mutations. Preferably, such DNA hybridizes with the DNA obtained from the genomic library under conditions of high stringency.
The invention further comprises a synthetic PRP
oligosaccharide represented by the following formula:
HO O O OH HO O ~ -' R
OH
OH ~
O OH O'--) ~ O OH
O n where n is an integer from 2 to 30 and R1 is (CH2)pCHO or (CH2CH20)pCH2CH2NH2 where p is an integer from 1 to 3.
In still another embodiment, the invention provides a compound useful 'as an intermediate in the preparation of synthetic PRP of the invention. It is represented by the formula:

~13$~ 65 Dn0 O O ODn Dn0 O O - RZ
OBn OBn U_ H O ODn O- P O 08n O n where n is an integer from 2 to 30, Bn is benzyl, and R2 is (CH2)pCH(OR3)2 or (CH2CH20)pCH2CH2R4 where p is an integer from 1 to 3, R3 is an alkyl group 1-4 carbons in length, and R4 is a group that can be converted into an amino group.
This compound is prepared using a solid phase synthesis. The monomer for chain initiation is a compound represented by the following formula:

Dn0 O OBn OBn ODn OMMTr HO OBn where Bn is benzyl and I~lTr is monomethoxytrityl. This monomer is coupled to a solid phase and then detritylated.
The resulting detritylated compound is coupled with a monomer for chain elongation represented by the formula:
o Dn0 O OBn ODn OBn OBn OMMTr O ~F-O
O
where Bn is benzyl and I~iTr is monomethoxytrityl. The resulting compound is then detritylated. The chain elongation and detritylation steps are repeated a sufficient number of times until an oligomer of the desired length is obtained. The chain terminating monomer is then added. The chain terminating monomer is represented by the formula:

s Bn0 O O -R
H
O -p_O 08n O' where Bn is benzyl and R2 is (CH2)pCH(OR3)2 or (CH2CH20)pCH2CH2R4 where p is an integer from 1 to 3, R3 is an alkyl group 1-4 carbons in length, and R4 is a group that can be converted into an amino group. The phosphonate groups of the support-bound oligomer are then oxidized to form phosphate groups. The resulting compound is then removed from the solid support and recovered.
The protective groups on this intermediate are then removed by hydrogenation. Where R2 is (CH2CH20)pCH2CH2R4, this results in the synthetic PRP of the invention. In the case where R2 is (CH2)pCH(OR3)2, the hydrogenated compound is further subjected to selective acid hydrolysis.
The preferred conjugate of the invention is then prepared by coupling the synthetic PRP with the Hi adhesin protein by reductive amination where R1 is (CH2)pCHO or, where R1 is (CH2CH20)pCH2CH2NH2, by preparing the cor-responding isothiocynate and then coupling the isothiocynate with the protein.
,, , , , WO 94/00149 ~ PCT/US93/06016 The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate one embodiment of the invention and, together with the description, serve to explain the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 shows an analysis of outer membrane preparations by SDS-polyacrylamide gel electrophoresis.
Samples included the following (lanes): 1, total outer membrane protein preparation from Haemophilus influenzae type b stained with Coomassie blue; 2, autoradiography of 35S-labeled total outer membrane proteins; 3, autoradiography of 35S-labeled adhesin protein eluted from immobilized receptor asialo-GM1: 4, autoradiography of material eluted from immobilized globoside, a nonsense glycolipid. Arrow indicates the adhesin migrating between P1 and P2 with a molecular weight of about 41 kD.
Figure 2 shows the neutralization of Haemophilus adhesin to the glycolipid receptor asialo-GM1.
~35S~methionine-labeled membranes from Haemonhilus influenza type b were incubated with serial dilutions of mouse sera and then allowed to bind to receptor (0.5 microgram/well). The mouse sera used was obtained from 5 mice, designated M-0 through M5, which had been immunized with Haemophilus membranes. The sera from an unchallenged mouse (NMS) was used as a negative control.

s Figure 3 shows inhibition of Haemophilus membrane binding to asialo-GM1 with selected monoclonal antibodies.
~35g~methionine-labeled membranes from Haemophilus were incubated with supernatants of hybridoma cultures and then allowed to bind to receptor (0.5 microgram/well). A
negative receptor control of Gb4 indicates the specificity of the receptor-ligand interaction. Mouse sera (M-2) (1:500 dilution) used in Figure 2 shows strong, positive inhibition. Media shows no inhibition of binding by membranes to asialo-GM1. Two classes of positively inhibiting hybridomas were found. Hib 10 shows total inhibition of binding. Hib 30 and Hib 43 show partial (about 35%) inhibition. Most hybridoma cultures, such as Hib 2, showed no inhibition. All hybridoma cultures tested for binding reacted positively with membranes in an ELISA.
Error bars are included to demonstrate the variability between duplicate wells.
Figure 4 shows the identification and characterization of the 47 kDa Haemophilus adhesin. The monclonal antibody. which partially inhibited membrane binding, Hib 43, was reacted on Western blot to identify the molecular weight of the protein it recognizes. Whole cells were run after no proteinase K treatment or either treatment with proteinase K prior to lysis in sample buffer or treatment with proteinase K after lysis in sample buffer (reading from left to right). Non-treatment identifies the 47 kDa protein; treatment of whole cells by proteinase K
prior to lysis indicates the sensitivity of the protein to this protease in its native location; and treatment after I II T

lysis by proteinase K demonstrates the general sensitivity to this protease after disruption from that native location. The Escherichia coli XL-1, transformed with pMC101, expresses the 47 kDa Haemophilus protein, which reacts with Hib 43. The 47 kDa protein was also sensitive to proteinase K treatment of XL-1/pMC101 whole cells.
These data suggest a surface location for this protein in both hosts.
Figure 5 shows a restriction map of the region in Haemoghilus influenza type b that encodes the 47 kDa adhesin. A 10.5 kbp Eco R1 fragment that produces the 47 kDa protein which reacts with Hib 43 monclonal antibody was cloned from an Haemophilus lambda ZAPII genebank. The helper phage 8408 was used to induce a plasmid containing this insert in the vector pSK(-).
Figure 6 shows the glycolipid binding phenotype of Escherichia coli that express the Hib 47 kDa protein. The ability of membranes from the E. coli. XL-1, or from XL-1 transformed with pMC101, designated 3, were compared using the standard binding assay. Serial dilutions were made of glycolipids with receptor activity: asialo-GM1, asialo-GM2, sulfatide, or the negative control, Gb4. XL-1/pMC101 binds with high affinity to these receptors, similar to Haemophilus.
Figures 7A and 7B show the nucleotide sequence of hin47 (SEQ ID NO:1) and the deduced Hin47 amino acid sequence (SEQ ID N0:2). The nucleotide sequence is numbered above each line and the deduced Hin47 amino acid sequence is shown below the line. The open reading frame for the hin47 gene is between,nucleotide 115 and 1503. The end of the putative leader sequence and beginning of the putative mature polypeptide is indicated at nucleotide base 189. The predicted molecular weight of the mature polypeptide is 46399 and it has a pI of 5.86.
Figure 8 shows the Haemophilus influenza type B, strain 9795 Hin47 amino acid sequence compared with the sequences of 5 non-typable strains as designated.
Identical amino acids are indicated by ~. Amino acid differences that are conservative with respect to charge are noted by lower case letters. Amino acids that differ with respect to charge are noted in upper case letters.
DETAILED DESCRIPTION OF THE INVENTION
Reference will now be made in detail to the pres-ently preferred embodiments of the invention, which, together with the following examples, serve to explain the principles of the invention.
The invention comprises an immunogenic oligosaccharide-H. influenzae adhesin protein conjugate useful as a vaccine against H. influenzae, purified H.
influenzae adhesin proteins and related proteins and polypeptides, DNA coding for the proteins and polypeptides, host cells containing the DNA and producing the proteins and polypeptides, synthetic PRP oligosaccharides and intermediates useful for their synthesis, and methods of making and using these materials.
n m r ImmunoQenic Conjucrate The conjugate comprises a polyribosylribitol phosphate fragment chemically coupled to a purified H.
influenzae adhesin protein. Preferably, the PRP fragment is a synthetic oligosaccharide. From 1 to about 30 and preferably from about 5 to 20 of the natural fragments or synthetic oligosaccharides are attached to the protein.
The fragments are attached to the protein by known techniques for covalently attaching polysaccharides to proteins or polypeptides, applied to the teachings contained herein. The preferred techniques here are reductive amination or isothiocyanate coupling.
Any adhesin protein may be used. In one embodiment, the purified adhesin protein is a minor Hi outer membrane protein with a molecular weight of about 41,000 daltons, distinct from P1 or P2.
In another~preferred embodiment, the purified adhesin protein is an Hi outer membrane protein with a molecular weight of about 47,000 daltons, distinct from P1-P6.
Alternatively, the protein may be replaced by a polypeptide that is an active site of the adhesin protein.
As used herein, the term "active site" means an epitope (antigenic determinant) or an H. influenzae receptor binding site, which may or may not also be an epitope. As ~'~'~ ~ ~ _ used herein, the term "receptor" is a macromolecule that binds to an Hi adhesin protein. The macromolecule is preferably a glycosphingolipid. Without intending to limit the scope of the invention, it is believed that the binding site is an epitope.
When the PRP fragment is obtained from natural sources, it is of varying lengths, but preferably about 8 to 120 monomers in length. Such fragments are obtained by known techniques, such as those described in the above-referenced European Patent Office Publication No.
0 338 265.
Synthetic PRP is a linear homopolymer of alternating molecules of ribose and ribitol joined by a phosphodiester linkage and represented by the formula:
HO O OH
OH O
OH i O ON O -P
I n O
where n is 2 to 30 and preferably 5-20. Such synthetic PRP's include those known in the art as well as the novel ones of the invention. For example, the previously mentioned European Patent Office Publications 0 320 942 and WO 94/00149 ~ ~ ~ ~ PCT/U593/06016 0 276 516 disclose synthetic PRP's that could be used in the conjugate of the invention.
Preferably, the synthetic PRP is a compound represented by the formula:
HO O O OH HO 0 O -'R
OH
OH ~
O OH O-- P O OH
I I
O n where n is an integer from 2 to 30 and R1 is (CH2)pCHO or (CH2CH20)pCH2CH2NH2 where p is an integer from 1 to 3.
Preferably n is 5-20 and p is 1. The synthetic PRP will be associated with a counter ion. Preferably, the ion is sodium (Na+).
The synthetic oligosaccharides usually contain a chemical spacer or linker by which they are attached to the protein. Such a spacer may be any chemical linkage that serves to connect the PRP and the protein and that has limited or no adverse effect to the animal host when the conjugate is administered. Such spacers may include those known in the art as well as the novel spacers of the inven-tion. Known spacers include those disclosed in the pre-viously mentioned European Patent Office Publications 0 320 942 and 0 276 516 as well as those disclosed in U.S.
Patent 4,830,852 issued May 16, 1989 to Marburg, et al.
Preferably, the chemical spacer is a moiety represented by the formula:
HO O O "' R
..-- O OH
where R is ( CHZ ) PCH2NH or ( CH2CH2 ) PCH2CH2NHCSNH and p is an integer from 1-3, preferably 1.
In the most preferred embodiment of the invention, the conjugate is represented by the following formula:

WO 94/00149 ' ~ PCT/US93/06016 R '_ X
HO O O HO O
OH
OH _ OH
H --- O OH O-P O OH
II
O ~ m where m is 1-30, n is 2-30, R is (CH2)pCH2NH or (CH2CH20)pCH2CH2NHCSNH where p is integer from 1-3, and X
is an H. influenzae adhesin protein or a fragment thereof containing an active site of the protein. Preferably, m is 5-20, n is 5-20, and p is 1. The symbol X in the above-referenced formula may also represent certain derived or modified proteins or polypeptides discussed below. The conjugate will be associated with a counter ion.
Preferably, the ion is Na+.
Adhesin Proteins The invention further comprises an isolated H.
influenzae adhesin protein. As used herein, in this context, the term "isolated" means that the protein is significantly free of other proteins. That is, a composition comprising the isolated protein is between 70%
and 94% pure by weight. Preferably, the protein is purified. As used herein, the term "purified" and related ~,~3g~ 65 terms means that the protein is at least 95% pure by weight, preferably at least 98% pure by weight, and most preferably at least 99% pure by weight. The protein binds to a receptor selected from the group consisting of fucosylasialo-GM1, asialo-GM1, and asialo-GM2, all of which contain the structure N-acetylgalactosamine(beta 1-4)galactose(beta 1-4)glucose-(beta 1-1)ceramide abbreviated GalNAc(betal-4)Gal(betal-4)Glc(betal-1)Cer. The protein also binds to another receptor, phosphatidylethanolamine.
In one embodiment, the protein is a minor outer membrane protein with a molecular weight of about 41 KD as determined by SDS PAGE. It is distinguishable from the various major outer membrane proteins that have been identified for Hi. In particular, the protein appears as a fainter band between the bands on a polyacrylamide gel for the outer membrane proteins known as P1 and P2. See Figure 1.
This purified Hi adhesin protein is prepared preferably from natural sources as follows. Hi bacterial membranes are obtained by standard techniques and solubilized, using a solubilizing compound, such as a detergent. Preferably, the membranes are mixed with the detergent, and the mixture is sonicated. The most preferred solubilizing agent is a solution containing about 1.0% to about 1.5% and preferably about 1.3% octyl-glucopyranoside. The adhesin protein is in the solubilized material. The remaining insoluble material from the membrane is separated, preferably by centrifuging.

The supernatant is contacted with receptors that bind the protein and are attached to an insoluble solid support or matrix, such as a microtiter well or a gel, for a period of time and under conditions sufficient for the protein to bind to the receptors, thus separating the protein from the other material. The preferred receptors for the adhesin protein are fucosylasialo-GM1, asialo-GM1, asialo-GM2, and phosphatidylethanolamine. These receptors can be prepared in accordance with the procedures disclosed in Krivan, et al., Proc. Natl. Acad. Sci. USA, 85:6157-6161 (1988). The most preferred receptor, asialo-GM1, is also commercially available. All of these receptors, except phosphatidylethanolamine, contain the carbohydrate sequence GalNAc(betal-4)Gal(betal-4)Glc, which, accordingly, may also be used as a receptor for the purification of the adhesin protein. This sequence can be prepared using standard carbohydrate synthesis techniques.
The adhesin protein is then eluted using the appropriate agent. This may be free receptor in solution, SDS elution buffer, or a chaotropic agent, such as KSCN, NaCl, or guanidine hydrochloride. The eluted protein is then tested against the receptor to confirm that the protein does bind to it. The purity of the isolated protein is analyzed by SDS-PAGE. Preferably, it will be about 99o pure after affinity purification with the most preferred receptor.
For purification of larger amounts of the adhesin protein, chromatography is preferred. The receptor is - 3~ - 21 387 fi5 immobilized onto a hydrophobic gel support, such as octyl-agarose. This matrix is prepared by adsorbing the receptors to the hydrophobic gel in the presence of salt as described by Hirabayashi, et al. for other glycolipids.
Hirabayashi, et al., J. Biochem., 94:327-330 (1983).
Photoactivatable heterobifunctional crosslinking agents have also been used to prepare glycolipid affinity matrices. Lingwood, C., J. Lipid Res., 25:1010-1012 (1984). In this case, the receptor-active lipid is covalently crosslinked to the gel support. The column is then preferably washed extensively with an appropriate buffer solution, such as TMS-buffer saline, before the protein is eluted.
A more preferred method is to purify the adhesin by affinity chromatography using an anti-adhesin monoclonal or polyclonal antibody prepared by standard techniques. In this case, the antibodies are covalently linked to agarose gels activated by cyanogen bromide or succinamide esters (Affi-GelTM, BioRad Inc.) or by other methods known by those skilled in the art. The sonic extract is loaded on the top of the gel as described above.
In another preferred embodiment, the adhesin proteins comprise an H. influenzae outer membrane protein with a molecular weight of about 47,000 daltons. Figures 7A and 7B show the protein amino acid sequence as well as the designated nucleotide sequence of the open reading frame (ORF) encoding a 49kDa protein. The 49kDa protein comprises 463 amino acids (amino acids 1-463 in Figure 7A

~~'O 94/00149 PCT/US93/06016 and 7B), includes a putative signal sequence of approximately 2.5 kDa and 25 amino acids, thereby resulting in a mature protein of approximately 47 kDa and 438 amino acids (amino acids 26 through 463 on Figure 7A and B), herein designated Hin47. This protein is distinguishable from the known Hi proteins P1-P6 on the basis of molecular weight and the fact that those proteins are integral membrane proteins, while this protein is an outer membrane protein. This protein also binds to the previous mentioned receptors as well as to sulfatide, (S03--galactose(beta 1-1)ceramide) and it is soluble in 1% Sarkosy~(N-lauroylsarcosine).
This protein is preferably prepared in purified form as follows. Hi membranes are extracted with a solution that removes membrane associated proteins, which produces an extract containing the adhesin protein along with other membrane associated proteins. Preferably, this solution is a nonionic detergent, such as Sarkosyl or octylglucopyranoside. The insoluble material is separated from the extract, preferably by centrifugation. This produces a supernatant that contains the adhesin protein.
The supernatant is then brought into contact with a monoclonal antibody which recognizes the adhesin protein.
The antibody is bound to an insoluble solid support. The contact is for a period of time and under standard reaction conditions sufficient for the adhesin protein to bind to the monoclonal antibody. Preferably, the solid support is a material used in a chromatographic column. The adhesin protein is then removed from the antibody, thereby permitting the recovery of protein in purified form.

Preferably, the nonionic detergent solution is removed from the supernatant before the supernatant is subjected to the affinity chromatography. Such removal is preferably accomplished by dialyzing the supernatant to produce a dialysate that is substantially free of the detergent.
The monoclonal antibodies can be prepared by standard techniques, given the teachings contained herein.
Such techniques are disclosed, for example, in U.S. Patent No. 4,271,145, issued June 2, 1981 to Wands et al. and U.S.
Patent No. 4,196,265, issued April l, 1980 to Koprowski et al. Briefly, mice are immunized with Hi membranes. Hybri-domas are prepared by fusing spleen cells from the mice with myeloma cells. The fusion products are screened for those producing antibodies that bind to the Hi membranes.
The positive clones are then screened to identify those whose binding with the Hi membranes is inhibited by an Hi adhesin receptor. The positive hybridomas clones are isolated, and the monoclonal antibodies are recovered from those clones.
Alternatively, the outer membrane proteins could be separated on a gel. The 47kDa band could be cut out and injected into the mice. The hybridomas could be prepared and screened as described above.
DNA
The adhesin proteins of the invention are preferably produced through genetic engineering techniques.
In this case, they are produced by an appropriate host cell that has been transformed by DNA that codes for the proteins. Preferably, the host cell is a bacterium, and most preferably the bacterium is E. coli, B. subtilis, or Salmonella.
The DNA of the invention is an isolated or substantially purified DNA sequence (i.e., polydeoxyribonucleotide molecule) encoding a protein or polypeptide that binds to the previously mentioned receptors. Preferably, the DNA of the invention includes an open reading frame (ORF) sequence (nucleotides 115 through 1503 in Figures 7A and B), designated hin47, encoding an approximate 49 kDa and 463 amino acid protein, designated Hin47, as shown in Figures 7A and B. Most preferably, the DNA comprises that part of the ORF that does not code for the signal sequence (nucleotides 191 through 1503 in Figures 7A and B). As used herein, the term "isolated" and variations thereof means that the DNA
is in isolation from DNA encoding other proteins or polypeptides normally accompanying the Hi adhesin proteins.
Thus, the DNA of the invention includes DNA encoding the protein or polypeptide when that DNA has been cloned into a microbial vector, such as a plasmid, or into a viral vector that may be harbored by a bacteriophage, provided that such clones are isolated from clones that contain DNA encoding other proteins or polypeptides normally accompanying this one. As used herein, the term "substantially pure" and variants thereof means that the DNA is substantially free of DNA and RNA that does not encode the proteins or polypeptides of the invention. That is, there will be no more than about 1~ by weight of other DNA and RNA and ,~3a~ 65 preferably no more than about 0.2% by weight of other DNA
and RNA in any sample that contains the DNA of the invention.
Preferably, the DNA is obtained by using either the receptors or monoclonal antibodies to the adhesins to screen an appropriate genomic library that contains H.
influenzae DNA. Such a library comprises colonies of a single type of microorganism, generally bacteria like E.
coli K12 (XL-1), into which pieces of the foreign DNA have been inserted, generally by being incorporated into a plasmid, cosmid, or phage vector compatible with the microorganism. More specifically, the library comprises clones of vectors into which different sequences of the DNA
have been operably and recoverably inserted, each of the vectors containing only one sequence of the DNA. The vectors may be plasmids, cosmids, phagemids, or phage genomes. If necessary because of the type of library being used, segments of DNA will have been inserted into the vectors in a manner that they will be expressed under appropriate conditions (i.e., in proper orientation and correct reading frame and with appropriate expression sequences, including an RNA polymerase binding sequence and a ribosomal binding sequence.) The microorganisms will be ones that do not express the adhesin protein, such as E.
coli HB101.
Clones from the library are brought into contact with the receptors or antibodies to identify those clones that bind. The clones are isolated and the exogenous DNA
sequence is recovered from one of the clones. The sequence - 35 - 2~ 3a~ s5 is preferably evaluated to determine if it encodes the protein.
Preferably, the genomic library comprises bacteria, such as E. coli infected by phage, preferably bacteriophage lambda. Plaques produced by the phage infected bacteria are screened by monoclonal antibodies to identify those plaques containing bacteria that produce the adhesin protein. The screening involves contacting the plaques with the monoclonal antibody to determine if binding has occurred, using standard techniques. Preferably, immunoassays are used.
In this preferred embodiment, the positive clones are then isolated by purifying the positive plaques and inducing plasmid formation in the bacteria in the purified plaque with a helper phage according to standard techniques.
In an alternate preferred embodiment, colonies con-taining DNA that encodes an Hi adhesin protein could be de-tected using DYNA Beads according to Olsvick et al., 29th ICAAC, Houston, Tex. 1989. The previously described recep-tors would be crosslinked to tosylated DYNA Beads M280, and these receptor-containing beads would then be used to ad-sorb to colonies expressing the adhesin protein. Colonies not expressing the adhesin would be removed by washing, and this process would be repeated to obtain an appropriate en-richment. Putative adhesin expressing colonies would then be plated and confirmed by metabolically labelling each colony with 35S-methionine and testing the ability of the colony to bind to the receptor as previously described.
The DNA from several adhering clones would be compared to identify shared sequences, and these shared sequences would be further subcloned and characterized.
Alternatively, the receptors could be nonspecifi-cally immobilized to a suitable support, such as silica or CeliteTM resin. This material would then be used to adsorb to colonies expressing the adhesin protein as described in the preceding paragraph.
In another alternate preferred embodiment, the gene for a specific adhesin would be localized and identified by constructing non-adherent mutants of a specific pathogen.
This would be accomplished by creating mutants using a transposable element such as TnPhoA as described in Manoil et al., Proc. Natl. Acad. Sci. USA, 82:81129-81133 (1985).
Alkaline phosphatase positive mutants would indicate mutations within exported proteins. Since the adhesin for each pathogen is located on the outer membrane surface and therefore exported, this set of mutants would contain a much reduced subset of mutants. They would then be screened for a loss in binding activity.
It will be recognized by persons skilled in the art that a DNA sequence for an Hi adhesin protein can be modified by known techniques in view of the teachings disclosed herein. For example, different codons can be substituted that code for the same amino acid as the ..x .

- 3~ - 2138765 original codon. Alternatively, the substitute codons may code for a different amino acid that will not affect the immunogenicity of the protein or which may improve its immunogenicity. For example, oligonucleotide directed, site specific mutagenesis or other techniques to create single or multiple mutations, such as replacements, insertions, deletions, and transpositions, as described in Botstein -and Shortle, "Strategies and Applications of In Vitro Mutagenesis", Science, 229:1193-1210 (1985), can be employed. Since such modified DNA can be obtained by the application of known techniques to the teachings contained herein, such DNA is within the scope of the claimed invention.
Moreover, it will be recognized by those skilled in the art that the DNA sequence (or fragments thereof) of the invention can be used to obtain other DNA sequences that hybridize with it under conditions of moderate to high stringency (including the derived sequences discussed in the preceding paragraph), using general techniques known in the art. That is, the hybridizing sequences are at least 90% homologous and preferably at least 95% homologous to hin47. Accordingly, the DNA of the invention includes such DNA.
The DNA of the invention may be used in accordance with known techniques, appropriately modified in view of the teachings contained herein, to construct an expression vector, which is then used to transform a microorganism for the expression and production of the adhesins of the invention. Such techniques include those disclosed in U.S.

38 _ 2138765 Patent Nos. 4,440,859 issued April 3, 1984 to Rutter et al., 4,530,901 issued July 23, 1985 to Weissman, 4,582,800 issued April 15, 1986 to Crowl, 4,677,063 issued June 30, 1987 to Mark et al., 4,678,751 issued July 7, 1987 to Goeddel, 4,704,362 issued November 3, 1987 to Itakura et al., 4,710,463 issued December 1, 1987 to Murray, 4,757,006 issued July 12, 1988 to Toole, Jr., et al., 4,766,075 issued August 23, 1988 to Goeddel, et al., and 4,810,648 issued March 7, 1989 to Stalker.
The DNA of the invention may be joined to a wide variety of other DNA sequences for introduction into an appropriate host cell. The companion DNA would depend upon the nature of the host cell, the manner of the introduction of the DNA into the host cell, and whether episomal maintenance or integration is desired.
Generally, the DNA is inserted into an expression vector, such as a plasmid, in proper orientation and correct reading frame for expression. If necessary, the DNA may be linked to the appropriate transcriptional and translational regulatory control nucleotide sequences recognized by the desired host, although such controls are generally available in the expression vector. The vector is then introduced into the host through standard techniques.
Generally, not all of the hosts will be transformed by the vector. Therefore, it will be necessary to select for transformed host cells. Once selection technique ~c WO 94/00149 _ ~ ~ ~ PCT/US93/06016 involves incorporating into the expression vector a DNA
sequence, with any necessary control elements, that codes for a selectable trait in the transformed cell, such as antibiotic resistance. Alternatively, the gene for such selectable trait can be on another vector, which is used to co-transform the desired host cell. The preferred expression vector for use in the invention is the plasmid pMC101. The preferred host cell is E. coli.
The transformed host cells express the proteins or polypeptides of the invention. Such cells are cultured by known techniques, and the proteins or polypeptides are recovered by known techniques. Depending upon the host and expression system used, the recombinant proteins and polypeptides of the invention may be part of a fusion protein produced by the transformed host cells. Such proteins are recovered by known techniques, and the undesired part may be removed by known techniques.
Alternatively, the fusion protein itself may be more immunogenic than the recombinant protein or polypeptide alone and, therefore, may itself be used in a vaccine.
If desirable, the adhesins can be further purified by the application of standard protein purification techniques, modified and applied in accordance with the discoveries and teachings described herein. Such techniques include electrophoresis, centrifugation, gel filtration, precipitation, dialysis, chromatography (including ion exchange chromatography, affinity chromatography, immunoadsorbent affinity chromatography, reverse-phase high performance liquid chromatography, and gel permeation high performance liquid chromatography), isoelectric focusing, and variations and combinations thereof .
One or more of these techniques are employed sequentially in a procedure designed to separate molecules according to their physical or chemical characteristics.
These characteristics include the hydrophobicity, charge, binding capability, and molecular weight of the protein.
The various fractions of materials obtained after each technique are tested for their ability to react with the adhesin receptors. Those fractions showing such activity are then subjected to the next technique in the sequential procedure, and the new fractions are tested again. The process is repeated until only one fraction reactive with the receptors remains and that fraction produces only a single band when subjected to polyacrylamide gel electrophoresis.
The preferred techniques include those identified and described in U.S. Patent No. 4,446,122 issued May 1, 1984 to Chu, et al. Preferably, the adhesins are purified by receptor affinity chromatography or antibody affinity chromatography.
Modified Adhesins The adhesins of the invention may be modified by known protein modification techniques. Such modifications include breaking the protein into fragments that contain at _ 41 _ 2138765 least one active site or the addition, substitution, or deletion of one or more amino acids to the protein or a fragment thereof. Preferably, such derived proteins or polypeptides are immunologically cross-reactive with the Hi adhesin proteins, thus being capable of eliciting an anti-genic response to H. influenzae in an animal host. Most preferably, such derived proteins or polypeptides also bind to an H. influenzae receptor selected from the group con-sisting of fucosylasialo-GM1, asialo-GM1, and asialo-GM2.
(As used in this specification, the term "polypeptide" also includes shorter chains of amino acids that are often re-ferred to as peptides.) Such modifications may enhance the immunogenicity of the protein or have no effect on such activity. The modification techniques include those dis-closed in U.S. Patent No. 4,526,716, issued July 2, 1985 to Stevens.
The proteins of the invention may contain one or more amino acid sequences that are not necessary to their immunogenicity. It may be the case, for example, that only the amino acid sequences of a particular epitope of the antigen will be necessary for immunogenic activity. Un-wanted sequences can be removed by techniques well-known in the art. For example, unwanted amino acid sequences can be removed via limited proteolytic digestion using enzymes such as trypsin, papain, or related proteolytic enzymes.
This latter approach is expected to be particularly useful for the adhesin protein of the invention. Since the protein binds to several related receptors having a consensus sequence, the protein should have a well con-served region that acts as the receptor binding site. This site is the particularly preferred polypeptide of the invention.
Alternatively, polypeptides corresponding to various immunogenic epitopes and/or the receptor binding site of the protein may be chemically synthesized by methods well-known in the art, given the teachings contained herein. These include the methods disclosed in U.S. Patent No. 4,290,944, issued September 22, 1981 to Goldberg.
Modified proteins or polypeptides can be prepared that are substantially homologous to the Hi adhesin protein or to the polypeptides discussed above through the use of known techniques and routine experimentation in view of the teachings contained herein. As used herein, the term "substantially homologous" means immunologically cross-reactive. Such a protein or polypeptide may be identified by the fact that it will bind to antibodies that were made to the adhesin protein of the invention, which antibodies can be prepared by standard techniques. Some of such modified proteins or polypeptides may have enhanced immunogenicity compared to the one from which they are derived.
Thus, the invention includes a class of derived proteins and polypeptides, including synthetically derived peptides or fragments of the adhesin protein, having common elements of origin, structure, and mechanism of action, WO 94/00149 ~ PCT/US93/06016 such as immunogenic effect or being able to bind to the previously mentioned receptors, that are within the scope of the present invention because they can be prepared by persons skilled in the art without undue experimentation, once given the teachings of the present invention.
Moreover, since persons skilled in the art can make modifications to or derivatives of epitopes or the receptor binding site on the proteins or polypeptides of the invention, once such epitopes or site are identified, such modifications or derivatives are within the scope of the invention. Such derived proteins and polypeptides are preferably pure as that term was previously defined herein.
The Hi adhesin protein of the invention (as well as the related proteins and polypeptides derived therefrom) has utility not only in the conjugate vaccine but as an immunogen in its own right. Thus, it can~be used in a vaccine for animals, including mammals, rodents, primates, and humans. The preferred use is a vaccine for humans, preferably children, and most preferably young infants.
Such a vaccine can be prepared by techniques known to those skilled in the art and would comprise, for example, the antigen, a pharmaceutically acceptable car-rier, an appropriate adjuvant, and other materials traditionally found in vaccines. An immunologically effec-tive amount of the antigen to be used in the vaccine is determined by means known in the art in view of the teach-ings herein.

'13$7 65 _ Synthetic PRP
The invention further comprises novel synthetic PRP
represented by the formula:
t HO O O HO O O ~ R
OH
OH
OH ~
O OH ~- P O OH
II
O
where n is an integer from 2 to 30, preferably 5-20, and R1 is (CH2)pCHO or (CH2CH20)pCH2CH2NH2 where p is an integer from 1 to 3, preferably 1. The ability to prepare this novel synthetic PRP permits the preparation of compositions where all of the PRP oligosaccharides are of the same length (i.e., have the same number of monomeric units), in contrast to PRP obtained from natural sources, where the fragments vary tremendously in length.
The PRP of the invention is prepared by a combina-tion of solid phase synthesis and the highly efficient H-phosphonate method for the construction of the phosphodiester linkage. It also involves the use of gels with higher levels of functionalization, which are better suited for commercial scale operations.

WO 94/00149 ~ ~ ~ ~ PCT/US93/06016 The general approach is to prepare a protected oligomeric ribosylribitol phosphate derivative by the following steps. First, the monomer for chain initiation is coupled to a solid phase. The monomer is represented by the formula:
O
Bn0 O 08n OBn Odn OMMTr HO 08n where Bn is benzyl and MMTr is monomethoxytrityl. See Compound 7, Table 1. The preferred solid phase is a Merrifield-type amino resin. The chain initiation monomer (Compound 7) is coupled to the solid phase by known techniques, such as reaction with succinic anhydride, followed by coupling of the obtained succinate of Compound 7 to amino groups of the solid phase. The loading is determined by colorimetric quantification of the trityl cation released on acid treatment. The coupled compound is then detritylated, such as by treatment with trifluoroacetic acid in dichloromethane.

138't 65 Chain elongation is accomplished by coupling the detritylated chain initiation monomer with a compound represented by the formula:
o Dn0 O 08n ODn H OBn OBn OMMTr O ap-O
O
where Bn is benzyl and MMTr is monomethoxytrityl. See Compound 8, Table 1. (The compound will be associated with a counter ion. Preferably, the ion is an organic cation, such as triethyl ammonium.) The coupling is accomplished by using a condensing reagent, such as pivaloyl chloride.
The resulting compound is then detritylated. The chain elongation-detritylation steps are repeated a sufficient number of times to prepare an oligosaccharide of the desired length. Thus, if n represents the desired number of PRP monomers in the oligosaccharide, the chain elongation-detritylation cycles are repeated n-2 times after the coupling of the chain initiation monomer and the first chain elongation monomer.
The chain is terminated by coupling it with a chain termination monomer represented by the following formula:

WO 94/00149 ~ ~ PCT/US93/06016 O O -R
Bn0 H
O -P--O OBn O-where Bn is benzyl and R2 is (CH2)pCH(OR3)2 or (CH2CH20)pCH2CH2R4 where p is 1-3, R3 is an alkyl group 1-4 carbons in length, and R4 is a group that can be converted into an amino group. See Compounds 10 and 12, Table 2.
(The compound will be associated with a counter ion.
Preferably, the ion is an organic cation, such as triethyl ammonium.) Preferably, p is 1, and R3 is methyl or ethyl.
Preferably, R4 is Ng, trifluoroacetyl, benzyloxycarbonyl, or fluorenylmethoxycarbonyl.
The phosphonate groups of the solid-bound oligomer are then oxidized to form phosphate groups. Preferably, this is accomplished by treatment with iodine in aqueous pyridine.
The resulting compound is then removed from this solid support, preferably through cleavage by methanolysis.
The recovered compound is represented by the formula:

2~ 387 65 Dn0 0 O OOn On0 0 0 - RI
OBn _ OBn ~
O OOn O~p O OBn O n where n is an integer from 2 to 30, preferably 5-20, Bn is benzyl, and Rz is defined as above. See Compounds 13 and 15, Table 3. (The compound will be associated with a counter ion. Preferably, the ion is ammonium or substi-tuted ammonium.) The resulting compound is then deprotected by hydrogenation with palladium on charcoal. In the case where RZ is (CHz) PCH (OR3) z, the hydrogenated compound is further subjected to selective acid hydrolysis, such as by treatment with aqueous trifluoroacetic acid. The resulting PRP oligomers are purified by standard techniques, prefer-ably by ion-exchange chromatography, HPLC or gel filtra-tion. See Compounds 14 and 16, Table 3.
Table 1 shows the synthesis of the chain initi-ation monomer, Compound 7, and the chain elongation monomer, Compound 8. The readily available methyl 2,3-isopropylidene-beta-D-ribofuranoside (Compound 1) (Leonard, et al., J. Het. Chem. 3:485 (1966) is used as starting material. Allylation of ._ Compound 1 with allyl bromide/sodium hydroxide in N,N-dimethylformamide gives the expected 5-O-allyl Compound 2 as an oil that can be distilled. This compound is subjected to a sequence of reactions comprising hydrolysis with aqueous formic acid, sodium borohydride reduction, tritylation with triphenylmethylchloride/pyridine, benzylation with benzyl chloride/sodium hydroxide in N,N-dimethylformamide, and hydrolysis with aqueous acetic acid.
The resulting Compound 3 is purified by silica gel chromatography.
Benzylation of Compound 1 with benzyl chloride/
sodium hydroxide in N,N-dimethylformamide gives the expected 5-O-benzyl compound 4 as an oil that can be distilled. This compound is subjected to a sequence of reactions comprising hydrolysis with aqueous formic acid and benzoylation with benzoyl chloride in pyridine, giving Compound 5, which is purified by chromatography and crystallization. Compound 5 is subjected to a further sequence of reactions comprising treatment with hydrogen bromide in dichloromethane to prepare the glycosyl bromide, followed by treatment with methanol and collidine. The resulting orthoester is then debenzoylated with sodium methoxide in methanol. The resulting product is allylated with allyl bromide/sodium hydroxide in N,N-dimethylformamide to give, after purification by silica gel chromatography, Compound 6.
Glycosylation can be accomplished by several methods. In the preferred method (A), Compound 6 is c~~~ ~~

treated with trimethylsilyl chloride to give the cor-responding glycosyl chloride, which, when treated with Compound 3 in the presence of molecular sieves, gives a ribitol glycoside. Alternatively (B), Compound 6 is transesterified in the presence of Compound 3. The resulting ribitol orthoester is then rearranged in situ to give the ribitol glycoside.
The ribitol glycoside is then subjected to debenzoylation with sodium methoxide in methanol and benzylation with benzyl chloride/sodium hydroxide in N,N-dimethylformamide. The resulting 5-O-allyl-2,3,4-tri-O-benzyl-1-O-(3-O-allyl-2,5-di-O-benzyl-beta-D-ribofuranosyl)-D-ribitol is deallylated by treatment with, successively, tris-(triphenylphosphine)rhodium(I)chloride and aqueous acetic acid and monomethoxytritylated with monomethoxytrityl chloride. The resulting chain initiation monomer (Compound 7) is purified by chromatography.
The condensation reaction of Compound 7 with phosphorous acid/5,5-dimethyl-2-oxo-2-chloro-1,3,2-dioxaphosphorinane.gives the chain elongation monomer (Compound 8).
Table 2 shows the synthesis of the monomers for chain termination. Compound 6 is reacted with trimethylsilyl chloride to give the corresponding chloride, which is reacted with the appropriate alcohols in the pres-ence of molecular sieves to give beta-glycosides of the alcohols. Preferably, the alcohols are 2-(2-azidoethoxy)-ethanol, 2-[2-benzyloxycarbonylamido)ethoxy]ethanol, or 2,2-diethoxyethanol. The beta-glycosides are subjected to the reaction sequence debenzoylation, benzylation, and deallylation, as in the preparation of Compound 7, which gives Compounds 9 or 11. Condensation with phosphorous acid/5-5-dimethyl-2-oxo-2-chloro-1,2,3-dioxa-phosphorinane according to the same procedure used to prepare Compound 8 gives the desired spacer-containing monomers (Compounds 10 or 12).
Table 3 shows the specific PRP oligomers obtained after solid phase synthesis employing Compounds 7, 8, and 10 or 12. Compounds 13 and 15 are the protected oligomers after removal from the solid support, and Compounds 14 and 16 are the final oligomers after deprotection.
The preferred use of the novel PRP is in the pre-paration of the novel immunogenic conjugates. The oligomer is coupled to one of the proteins or polypeptides of the invention by standard techniques applied to the teachings contained herein. When the spacer terminates in an alde-hyde group, the preferred technique is reductive amination using sodium borohydride as described in Roy, et al., J.
Carbohydr. Chem. 6:161-165 (1987) and Lee, et al., Carbo-hydr. Res., 77:149-156 (1979). When the spacer terminates with an aminogroup, the PRP is converted into the iso-thiocynate by treatment with an activated thiocarbonic acid derivative, such as thiophosgene, and then coupled to the - 52 - 2~38~s5 protein at a pH of 9-10 in accordance with the procedures described in Kallin, et al., Glycoconiuaate J., 3:311-319 (1986) and Zopf, et al., Methods Enzymol., 50:171-175 (1978). The ratio of protein/carbohydrate is determined by a combination of Lowry protein determination and ribose determination. The ratio is primarily a function of the ratio of carbohydrate to protein in the initial reaction mixture and the type of spacer used. As shown in Example 3, the use of a spacer terminating in an amino group (Compound 16) results in a greater number of oligosac-charides being coupled to the protein than the use of a spacer terminating in an aldehyde group (Compound 14).
Table 4 shows the formulas of the final conjugates.
Vaccines The adhesin-oligosaccharide conjugates, as well as their protein components as previously mentioned, may be used in vaccines against both invasive and non-invasive strains of H. influenzae. The conjugate vaccines should have greatest utility against H. influenzae type b.
The vaccines comprise an immunologically effective amount of the immunogen in a pharmaceutically acceptable carrier. The combined immunogen and carrier may be an aqueous solution, emulsion, or suspension. An im-munologically effective amount is determinable by means known in the art without undue experimentation, given the teachings contained herein. In general, the quantity of immunogen will be between 0.1 and 100 micrograms per dose.

WO 94/00149 ~ ,~ ~ ~ 5 PCT/US93/06016 The carriers are known to those skilled in the art and include stabilizers, diluents, and buffers. Suitable stabilizers include carbohydrates, such as sorbitol, lactose, manitol, starch, sucrose, dextran, and glucose and proteins, such as albumin or casein. Suitable diluents include saline, Hanks Balanced Salts, and Ringers solution.
Suitable buffers include an alkali metal phosphate, an alkali metal carbonate, or an alkaline earth metal carbonate. The vaccine may also contain one or more adjuvants to improve immunogenicity. Suitable adjuvents include aluminum hydroxide, aluminum phosphate, or aluminum oxide or a composition that consists of a mineral oil, such as Marcol 52, or a vegetable oil and one or more emulsifying agents.
The vaccine may also contain other immunogens.
Such a cocktail vaccine has the advantage that immunity against several pathogens can be obtained by a single administration. Examples of other immunogens are those used in the known DPT vaccines.
The vaccines of the invention are prepared by techniques known to those skilled in the art, given the teachings contained herein. Generally, the immunogens are mixed with the carrier to form a solution, suspension, or emulsion. One or more of the additives discussed above may be in the carrier or may be added subsequently. The vac-cine preparations may be dessicated, for example, by freeze drying for storage purposes. If so, they may be subsequently reconstituted into liquid vaccines by the ad-dition of an appropriate liquid carrier.

1'3~~ b~' _ The vaccines are administered to humans or other mammals, including rodents and primates. Preferably, they are administered to human children, most preferably children younger than 18 months of age. They can be administered in one or more doses. The vaccines may be administered by known routes of administration for this type of vaccine. The preferred routes are intramuscular or subcutaneous injection. Accordingly, the invention also comprises a method for inducing an immune response to Hi in a mammal in order to protect the mammal against infection by invasive or non-invasive Hi. The method comprises administering an immunologically effective amount of the immunogens of the invention to the host and, preferably, administering the vaccines of the invention to the host.
Reagents The conjugates, protein/polypeptides, and oligomers of the invention are also useful as reagents for scientific research on the properties of pathogenicity, virulence, and infectivity of Hi,~as well as host defense mechanisms. For example, the DNA of the invention can be used in an oligonucleotide probe to identify the DNA of other microorganisms that might encode an adhesin for such organism. The protein of the invention could be used to make a monoclonal antibody that could be used to further purify compositions containing the protein by affinity chromatography. The protein could also be used in standard immunoassays to screen for the presence of antibodies to H.
influenza in a sample. A composition in accordance with 'WO 94/00149 PCT/US93/06016 the present invention useful as an investigational reagent contains an amount of conjugate, protein/polypeptide, or oligomer effective to provide the information or analysis sought. The determination of the amount necessary to accomplish a particular research goal depends upon the specific type of investigation involved and is readily within the routine skill of one engaged in such research, once given the teachings contained herein.
It is to be understood that the application of the teachings of the present invention to a specific problem or environment will be within the capabilities of one having ordinary skill in the art in light of the teachings contained herein. Examples of the products of the present invention and processes for their preparation and use appear in the following examples.

Preparation of Synthetic PRP OliQOSaccharide The preparation of the synthetic PRP
oligosaccharides of the invention is illustrated as described herein and as shown in the reaction schemes outlined in Tables 1-3.
Methyl 5-O-allyl-2,3-O-isopropylidene-beta-D
ribofuranoside (Compound 2) A solution of methyl 2,3-O-isopropylidene-beta-D-ribofuranoside (Compound 1, 50.0 g), N,N-dimethyl formamide (250 ml), and powdered sodium hydroxide (55.0 g) was stirred while allyl bromide (50.0 ml) was added dropwise.
After 2h, the excess allyl bromide was destroyed by addi-tion of methanol (50 ml). After being stirred for another hour, the mixture was partitioned between water and toluene. The organic phase was washed with water, dried with magnesium sulfate, and concentrated. Barium carbonate (250 mg) was added and the oil was distilled at 90-95°C, 0.75 mm Hg. The yield of Compound 2 was approximately 90%.
5-O-allyl-2,3,4-tri-O-benzyl-D-ribitol (Compound 3) Methyl 5-O-allyl-2,3-O-isopropylidene-beta-D-ribofuranoside (Compound 2, 1.5 g) in aqueous formic acid (25 ml) was heated on an oil bath at 100°C for 10 hrs and was then concentrated and coevaporated twice with water.
The obtained syrupy material, consisting mainly of 5-O-allyl-D-ribose and residual formic acid, was dissolved in water (25 ml), and the pH was adjusted to 7 with aqueous ammonia. Sodium borohydride (0.5 g) was added, and the mixture was stirred for 3h, then adjusted to pH 7 with acetic acid, and concentrated. After three co-concentrations with acetic acid-methanol (1:1) and two co-concentrations with methanol, the residue was dissolved in water (50 ml), and the solution was slowly passed through a column of Dowex-50~ (H+ form, 50-100 mesh, 2x20 cm) ion exchange resin. The eluate, consisting mainly of 5-o-allyl-D-ribitol, was concentrated, taken up in pyridine, n 2.38765 concentrated, and taken up again in pyridine (25 ml).
Triphenylmethyl chloride (8.0 g) was added, and the mixture was stirred at room temperature for 16h, then methanol (2.0 ml) was added. After 15 min, the mixture was partitioned between dichloromethane and water. The organic layer was washed with water, sulfuric acid, and aqueous sodium hydrogen carbonate, dried (magnesium sulfate), and concentrated. The residue was dissolved in N,N-dimethyl formamide (25 ml). The solution was stirred while powdered sodium hydroxide (3.5 g) was added, followed by benzyl chloride (4.40 ml, dropwise). After 2 hours, methanol (5 ml) was added, and after 15 min the mixture was partitioned between toluene and water. The organic layer was washed with water and concentrated. The residue was dissolved in 90% aqueous acetic acid (50 ml) and heated to 100°C for 2h, then concentrated and co-concentrated with toluene. The residue was purified by chromatography on silica gel. The compound was eluted with toluene-ethyl acetate 9:1. The yield of syrupy Compound 3 was 48%.
Methyl 5-O-benzyl-2,3-O-isopropylidene-beta D-ribofuranoside (Compound 4) A solution of methyl-2,3-O-isopropylidene-beta-D
ribofuranoside (Compound 1, 50 g), N,N-dimethyl formamide (250 ml), and powdered sodium hydroxide (50 g) was stirred while benzyl chloride (64 ml) was added dropwise. After 2h, the excess of benzyl chloride was destroyed by addition of methanol (50 ml). After being stirred for another hour, the mixture was partitioned between water and toluene. The WO 94/00149 ~ PCT/US93/06016 organic phase was washed with water, dried with magnesium sulfate, and concentrated. Barium carbonate (250 mg) was added and the oil was distilled at 115-120°C, 0.4 mm Hg.
The yield of Compound 4 was approximately 90%.
Methyl 5-O-benzyl-2,3-di-O-benzoyl-beta D-ribofuranoside (Compound 5) A solution of methyl 5-O-benzyl-2,3-O-isopropylidene-beta-D-ribofuranoside (Compound 4, 23 g) in 95:5 formic acid-water (200 ml) was kept at room temperature for 30 min, then cooled in ice. The cooled solution was poured into a vigorously stirred mixture of crushed ice, aqueous sodium hydroxide (240 g in 2000 ml), and dichloromethane (1000 ml). The mixture was shaken well in a separatory funnel, the organic layer was separated, and the aqueous layer was extracted four times with 500 ml portions of dichloromethane. The combined organic extracts, containing mainly methyl 5-O-benzyl-beta-D-ribofuranoside were concentrated. Dry pyridine (50 ml) was added, the mixture~was concentrated, then dry pyridine (150 ml) was added again. The mixture was cooled in ice while benzoyl chloride (34 ml) was added dropwise. The mixture was further stirred at room temperature overnight, then water (2 ml) was added to destroy excess benzoyl chloride. The mixture was then partitioned between water (1000 ml) and dichloromethane (500 ml). The organic layer was washed with 2 M aqueous sulfuric acid, then with 1 M
aqueous sodium hydrogen carbonate. Concentration yielded a syrup, which was purified on a column of silica gel. The WO 94/00149 ~ ~ PCT/US93/06016 fractions containing pure material were pooled and concentrated. The material could be crystallized from methanol in the cold, mp 68-69°C. The yield of Compound 5 was 22-41%. The chromatography also gave some starting material (Compound 4) in pure form (5-20%).
3-O-allyl-5-O-benzyl-1,2-O-methoxybenzylidene alpha-D-ribofuranose (6) A solution of hydrogen bromide in dichloromethane was prepared by mixing dichloromethane (150 ml), methanol (3.0 ml), and acetyl bromide (6.0 ml). Then methyl 2,3-di-O-benzoyl-5-O-benzyl-beta-D-ribofuranoside (Compound 5, 4.62 g) was added, and the mixture was stirred at room temperature for 30 min., after which the mixture, contain-ing mainly 2,3-di-O-benzoyl-5-O-benzyl-alpha-D-ribofuranosyl bromide, was cooled in ice while collidine (25 ml) was added dropwise with stirring, followed by methanol (10 ml). The mixture was further stirred for 3h at room temperature, then washed with water, concentrated, and co-concentrated with methanol. The residue, containing mainly 3-O-benzoyl-5-O-benzyl-1,2-O-methoxybenzylidene-alpha-D-ribofuranose, was dissolved in methanol (50 ml), and a solution of sodium methoxide in methanol (0.5 M, 20 ml) was added. After 2h at room temperature, the mixture was neutralized by addition of C02(s), then concentrated and co-concentrated once with N,N-dimethylformamide. The residue was dissolved in N,N-dimethylformamide (50 ml) and stirred at room temperature while powdered sodium hydroxide (3.0 g) was added, followed ~~ 387_65 by allyl bromide (3.0 ml). After 1h, the mixture was partitioned between water and toluene, the organic layer was washed with water, and concentrated. The residue was purified by chromatography on silica gel using toluene-ethyl acetate-pyridine (90:10:1) as the eluant. The ap-propriate fractions were pooled and concentrated to give Compound 6 (1.90 g, 48%) as a colorless syrup.
2,3,4-tri-O-benzyl-1-O-(2,5-di-O-benzyl-beta-D-ribofuranosyl)-5-O-monomethoxytrityl-D-ribitol (Compound 7) Glycosidation Method A
Compound 6 (4.0 g) was dissolved in trimethylsilyl chloride (20 ml). After 20 min. at room temperature, the solution was concentrated, then co-concentrated with dry dichloromethane. The residue was dissolved in dry dichloromethane (25 ml) containing powdered 4A molecular sieves (5.0 g) and Compound 3 (4.6 g). The mixture was stirred at room temperature overnight. The mixture was filtered and concentrated. The residue was purified by column chromatography (toluene-ethyl acetate 15:1 as eluant) and then taken up in 0.04M methanolic sodium methoxide (50 ml). After 1 hr at room temperature, the mixture was neutralized by addition of C02(s), then concentrated and co-concentrated once with N,N-dimethylformamide. The residue was dissolved in N,N-dimethylformamide (50 ml) and stirred at room temperature while powdered sodium hydroxide (3.0 g) was added, followed by benzyl chloride (3.0 ml). After lh, the mixture was partitioned between water and toluene, and the organic WO 94/00149 6 ~ PCT/US93/06016 layer was washed with water and concentrated. The residue was purified by chromatography on a short column of silica gel using toluene-ethyl acetate (9:1) as eluant. The frac-tions containing 5-o-allyi-2,3,4-tri-O-benzyl-1-O-(3-O-allyl-2,5-di-O-benzyl-beta-D-ribofuranosyl)-D-ribitol were pooled and concentrated. The residue was dissolved in 30:12:4 ethanol-toluene-water (75 ml), and the solution was refluxed in the presence of tris(triphenylphosphine)rhodium(I)chloride (200 mg) until thin-layer chromatography showed complete conversion. The mixture was concentrated and taken up in acetic acid-water (30m1, 9:1 by volume) and the mixture was heated to 80°C
for 1 hour, concentrated and the residue was partitioned between diethyl ether and water, dried, and concentrated.
The residue, containing mainly 2,3,4-tri-O-benzyl-1-O-(2,5-di-O-benzyl-beta-D-ribofuranosyl)-D-ribitol, was taken up in dry pyridine (50 ml), and monomethoxytrityl chloride (3.5 g) was added. The mixture was stirred overnight, then methanol was added to destroy the excess chloride. After 30 min, the mixture was partitioned between dichloromethane and water, then washed with aqueous sulfuric acid and aqueous sodium bicarbonate, dried, and concentrated. The residue was purified by chromatography on a column of silica gel using toluene-ethyl acetate (9:1, containing 1%
pyridine) as eluant. The appropriate fractions were pooled and concentrated to give Compound 7 (4.9 g, 50%, calculated from 6) as a colorless syrup.
G ~cosidation Method B
Compounds 3 (4.6 g) and 6 (4.0 g) were dissolved in dry nitromethane (60 ml). Methanol was removed by '~1'~87~

continuous distillation at constant volume with continuous addition of nitromethane until thin-layer chromatography showed complete transesterification of Compound 6. Mercury (II) bromide (500 mg) was added, and solvent was distilled.
off at constant volume with continuous addition of nitromethane until thin-layer chromatography showed the formation of a new product. The mixture was purified by chromatography and treated further as described under method A above.
2,3,4-tri-O-benzyl-1-O-(2,5-di-O-benzyl-beta-D-ribofuranosyl)-5-O-monomethoxytrityl-D-ribitol 3-H-phosphonate (Compound 8) Compound 7 (4.9 g) was taken up in dry pyridine, and concentrated to dryness, then taken up in pyridine (20 ml) and added to a solution of phosphonic acid (4.1 g) in pyridine (20m1). 5,5-dimethyl-2-oxo-2-chloro-1,3,2-dioxaphosphorinane (5.0 g) was added. When thin-layer chromatography showed complete conversion, 1 M aqueous triethylammonium bicarbonate (5 ml) was added, and the mixture was partitioned between dichloromethane (200 ml) and 0.5 M aqueous triethylammonium bicarbonate (130 ml).
The organic layer was concentrated, and the residue was purified by chromatography on a short column of silica gel using a stepwise gradient of methanol in dichloromethane (0-20%, containing 1% pyridine) as eluant. The yield of amorphous Compound 8 was 80-90%.
2,2-Diethoxyethyl 2,5-di-O-benzyl-beta-D-ribofuranoside (Compound 9, p=1, R3=ethyl) 213~'~6~
WO 94/00149 ~ PCT/US93/06016 A mixture of Compound 6 (2.0 g) and trimethylsilyl chloride (15 ml) was kept at room temperature for 20 min., then concentrated, and co-concentrated with dry dichloromethane. The residue was mixed with glycolaldehyde diethylacetal (1.0 g), powdered 4 A molecular sieves (3.0 g), and dry dichloromethane (15 ml) and was stirred at room temperature overnight, then filtered and concentrated. The residue was taken up in 0.04M methanolic sodium methoxide (25 ml). After 1 hr. at room temperature, the mixture was neutralized by addition of C02(s) then concentrated and co-concentrated once with N,N-dimethylformamide. The residue was dissolved in N,N-dimethylformamide (20 ml) and stirred at room temperature while powdered sodium hydroxide (3.0 g) was added, followed by benzyl chloride (3.0 ml). When TLC
indicated complete conversion, methanol (2 ml) was added, and after 15 min. the mixture was partitioned between water and toluene, the organic layer was washed with water and concentrated. The residue was purified by chromatography on a short column of silica gel using toluene-ethyl acetate (8:2) as eluant. The appropriate fractions were collected and concentrated, then taken up in 30:12:4 ethanol-toluene-water (50 ml), and the solution was refluxed in the pres-ence of tris(triphenylphosphine)rhodium(I)chloride (100 mg) until thin-layer chromatography showed complete conversion.
The mixture was then diluted with dichloromethane, washed with saturated aqueous potassium chloride, and concentrated. The residue was dissolved in 10:1 acetone-water (20 ml), and mercuric oxide (2.0 g) followed by mercuric chloride (2.0 g) was added. After stirring at room temperature for 30 min., the solids were removed by ' - 64 -filtration, and the filtrate was partitioned between diethyl ether and water, washed with aqueous potassium iodide, dried, and concentrated. Purification on a short silica gel column, using toluene-ethyl acetate (8:2) as eluant, gave syrupy Compound 9. The yield was 60-65%.
2-[2-(benzyloxycarbonylamido)ethoxy]ethyl 2,5-di-O-benzyl-beta-D-ribofuranoside (Compound 11, p = 1, R4 = NHCOOBn) A mixture of Compound 6 (2.0 g) and trimethylsilyl chloride (15 ml) was kept at room temperature for 20 min., then concentrated, and co-concentrated with dry dichloromethane. The residue was mixed with 2-[2-(benzyloxycarbonylamido)ethoxy]ethanol (1.5 g), powdered 4 A molecular sieves (3.0 g), and dry dichloromethane (15 ml) and was stirred at room temperature overnight, then filtered and concentrated. The residue was taken up in 0.04M methanolic sodium methoxide (25 ml). After 1 hr. at room temperature, the mixture was neutralized by addition of C02(s), then concentrated and co-concentrated once with N,N-dimethylformamide. (The residue was dissolved in N,N-dimethylformamide {20 ml) and stirred at room temperature while freshly prepared silver oxide (3.0 g) was added, fol-lowed by benzyl bromide (3.0 ml). When thin layer chromatography indicated complete conversion, the mixture was filtered. The filtrate was partitioned between water and toluene, the organic layer was washed with water and aqueous sodium thiosulfate, and concentrated. The residue was purified by chromatography on a short column of silica gel using toluene-ethyl acetate (8:2) as eluant. The ap-propriate fractions were collected and concentrated, then WO 94/00149 ~ ~ ~ PCT/US93/06016 treated with selenium dioxide (570 mg) and acetic acid (0.4m1) in dioxane (14 ml) at reflux for 40 min. The mixture was then filtered through Celite. The yield of syrupy Compound 11 after chromatographic purification was 50%.
2-(2-Azidoethoxy)ethyl 2,5-di-O-benzyl-beta-D-ribofuranoside (Compound 11, p = 1, R4 = N3) A mixture of Compound 6 (2.0 g) and trimethylsilyl chloride (15 ml) was kept at room temperature for 20 min., then concentrated, and co-concentrated with dry dichloromethane. The residue was mixed with 2-(2-azidoethoxy)ethanol (2.0 g), powdered 4 A molecular sieves (3.0 g), and dry dichloromethane (15 ml) and was stirred at room temperature overnight, then filtered and concentrated.
The residue was taken up in 0.04M methanolic sodium methoxide (25 ml). After 1 hr. at room temperature, the mixture was neutralized by addition of C02(s), then concentrated and co-concentrated once with N,N-dimethylformamide. The residue was dissolved in N,N-dimethylformamide .(20 ml) and stirred at room temperature while powdered sodium hydroxide (3.0 g) was added, followed by benzyl chloride (3.0 ml). When thin layer chromatography indicated complete conversion, methanol (2 ml) was added, and after 15 min. the mixture was partitioned between water and toluene, the organic layer was washed with water and concentrated. The residue was purified by chromatography on a short column of silica gel using toluene-ethyl acetate (8:2) as eluant. The appropri-ate fractions were collected and concentrated, then treated with, acetic acid (0.4m1), dioxane (14m1) and selenium dioxide (0.57g) at reflux for 40 min. The mixture was filtered and concentrated. The yield of syrupy Compound 11 after chromatograhic purification was 50%.
2,2-Diethoxyethyl 2,5-di-O-benzyl-beta-D-ribofuranoside 3-H-phosphonate (Compound 10, p = 1, R3=ethyl) Compound 9 was treated with phosphonic acid and condensing reagent essentially as described for the preparation of compound 8 to give amorphous Compound 10 (67%).
2-[2-(benzyloxycarbonylamido)ethoxy]ethyl 2,5-di-O-benzyl-beta-D-ribofuranoside 3-H-phosphonate (Compound 12, p = 1, R4 = NHCOOBn) Compound 11 was treated with phosphonic acid and condensing reagent essentially as described for the preparation of Compound 8 to give amorphous Compound 12 (75%) .
2-(2-Azidoethoxy)ethyl 2,5-di-O-benzyl-beta-D-ribofuranoside 3-H-phosphonate (Compound 12, p = 1, R4 =
N3) Compound 11 was treated with phosphonic acid and condensing reagent essentially as described for the preparation of Compound 8 to give amorphous Compound 12 (70%) .

Solid phase synthesis: chain initiation 1. Preparation of the 3-succinate of Compound 7 To a solution of Compound 7 (4 mmol) in dry pyridine (25 ml) containing 4-dimethylaminopyridine (1 mmol) was added succinic anhydride (10 mmol). After stir-ring overnight water (0.5m1) was added. After 3 hrs. the mixture was partitioned between 1:1 toluene-ethyl acetate and aqueous phosphate buffer (pH 6.5). The organic lager was washed with buffer, and concentrated. The obtained 3-succinate of 7 was dried in vacuum over phosphorous pentoxide.
2. Coupling of the 3-succinate to the solid phase The succinate obtained above (10 equivalents over the resin amino group content) was dissolved in dichloromethane (5 ml/g) and mixed with a solution of dicyclohexylcarbodiimide (5 equivalents over the resin amino group content) in a small volume of dichloromethane.
The mixture was stirred for 15 min. at room temperature, then concentrated. The residue was dissolved in N,N-dimethylformamide (5 ml/g) and the solution was filtered, then added to Merrifield-type aminomethyl resin (pre-washed with N,N-dimethylformamide). After 6 h, the resin was washed with N,N-dimethylformamide, then with pyridine. The resin was treated with 9:1 pyridine-acetic anhydride for 2 hr., washed with pyridine, then washed with dichloromethane. The degree of functionalization was determined by treating a dried and weighed amount of resin with 0.5% trifluoroacetic acid in 1,2-dichloroethane, and 2~3g~ 6~

estimating the trityl cation content in the supernatant by spectrophotometry (495 nm). A typical value was 0.5 mmol/
g.
Solid phase synthesis: chain elongation cycle The solid-phase synthetic operations were carried out in a semi-automated apparatus, consisting of a reaction vessel with a glass filter bottom, agitation device (small scale batches were agitated by pressing dry nitrogen through the bottom filter), liquid outlet (bottom), and liquid inlet (top). Liquid was removed from the vessel through the bottom filter by suction, and added at the top by pressing with nitrogen from other vessels through teflon tubing.
1. Trityl deprotection The resin was treated with a 0.5~ solution of trifluoroacetic acid in dichloromethane until no more trityl cation was released (as determined spectrophotometrically), then the resin was washed with dichloromethane, followed by 4:1 dichloromethane-pyridine.
2. Coupling Pivaloyl chloride (4 equivalents over the resin hydroxyl groups) in dichloromethane (2m1/mmol. chloride) was added to a solution of compound 8 (4 equivalents) in 4:1 dichloromethane-pyridine (8m1/mmol. chloride). After 2 min., the mixture was added to the resin. Agitation was continued for 10 min, then the resin was washed with, WO 94/00149 213 g 7 ~ ~ PCT/US93/06016 successively, pyridine and 4:1 dichloromethane pyridine and dichloromethane. The yield in each coupling step was 97%-99%, as determined spectrophotometrically by the amount of the released trityl cation in the deprotection step.
Chain Termination Detritylated resin was treated as under (2) but with compound 9 or 11 instead of 8.
Oxidation The resin was treated with a freshly prepared 1%
solution of iodine in 98% aqueous pyridine for 30 min., then washed with, successively, pyridine and dichloromethane.
Removal from resin The resin was treated with sodium methoxide 1:1 dioxane-methanol (0.05M) for 16 hours at room temperature, acetic acid was added, and the mixture was then filtered and the filtrate was concentrated. The residue, according to NMR analysis, contained compound 13 (if 10 was used for chain termination)~or 15 (if 12 was used for chain termination), together with impurities.
Deprotection 1. Conversion of Compound 13 to Compound 14 The material that was removed from the resin as described above was dissolved in 1:2:2 ethylacctate-ethanol-water(0.1 ml/mg material) containing acetic acid (0.3%), and 10% Pd/C (0.5-2 mg/mg material) was added. The mixture was hydrogenated at 60 °C and atmospheric pressure overnight, then filtered, adjusted to pH 7, and concentrated. The residue was partitioned between diethyl ether and water. The aqueous layer was separated and concentrated. The residue was taken up in 50% aqueous trifluoroacetic acid at 0°C. After 4 h, the mixture was neutralized at 0°C with ammonia to pH 7, then the mixture was concentrated to a volume of approximately 10 mg/ml, and applied to a column of Fractogel TSK HW-50, packed and eluted with lOmM ammonium bicarbonate buffer, pH6.2. The appropriate fractions were collected, concentrated, and redissolved in water (0.1 ml/mg material). This solution was slowly passed through a column of Dowex-50 8 (Na form, packed and eluted with water). The appropriate fractions were collected and lyophilized. NMR spectroscopy in D20 solution showed, inter alia, signals from the anomeric protons in the region 4.9-5.1 ppm and signals from the spacer unit (aldehyde proton, dihydrate form) at 5.1-5.2 ppm. The amount of successful coupling cycles (that is, the value of n in formula for Compound 14) was verified by integration over the anomeric signals and the spacer signals, respectively.
2. Conversion of Compound 15 to Compound 16 The material that was removed from the resin was treated essentially as described above for conversion of Compound 13 to 14, except that the trifluoroacetic acid treatment was omitted. NMR spectroscopy in D20 solution of the lyophilized product showed, ter a ia, signals from the anomeric protons in the region 4.9-5.1 ppm and signals _ 21 387 65 from the spacer unit (CH2N triplet) at 3.2 ppm. The amount of successful coupling cycles (that is, the value of n in the formula for Compound 16) was verified by integration over the anomeric signals and the spacer signals, respectively. The purification of 14 and 16 could also be effected by preparative HPLC on Nucleosil C-18~ using 0.1 M
aqueous triethylammonium acetate (pH5.3) with 2.5%
acetonitrite as eluant.

Purification of an Hib Adhesin Bacteria were grown 24 h in defined media and labeled metabolically with 35S-methionine. Cells were harvested and washed by centrifugation three times in saline and suspended in approximately 20 ml of 10 mM Hepes buffer, pH 7.4, and chilled on ice. The bacterial suspen-sion was then sonicated on ice 6 times for 30 seconds each at a setting of 4 on a Bronson Sonicator. The sonic extract was centrifuged at 10,000 x g for 10 min. at 4° C, and the resulting outer membrane protein (OMP) pellet was stored until use in Hepes buffer containing protease inhibitors (PIC I & PIC II).
OMPs were next centrifuged at 100,000 x g for 30 min. at 4' C and the resulting pellet was suspended in 4 ml of 10 mM Hepes, pH 8.0, containing 1.3% octyl-glucopyranoside (Sigma), sonicated 5 min., and incubated at W",Q 94/00149 21~,g~ ~~

room temperature for 30 min. The resulting solubilized OMPs were centrifuged again at 100,000 x g.for 30 min. at 4° C, and the supernatant containing partially purified adhesin was decanted and saved.
The adhesin was purified by a receptor-affinity solid phase procedure as follows. The supernatant was diluted 1/10 in 50 mM Tris-HC1, pH 7.8, containing 150 mM
NaCL and 1% bovine serum albumin (BSA) and incubated in receptor-coated microtiter wells (0.8 micrograms of gangliotetraosylceramide/well) which had been previously blocked with BSA. Control wells lacking receptor were also used. After a 2 h incubation at room temperature, wells were washed 4 times with cold saline. The receptor-bound adhesin was eluted by incubating the wells for 30 min. at 37° C with 0.05 ml of 10 mM Tris-HC1, pH 7.8, containing 0.1% SDS which had been previously heated to 60° C. The SDS elution buffer was removed from the wells and analyzed for protein by SDS-PAGE and autoradiography.
Alternatively, the adhesin can be purified by using an affinity chromatography column where the lipid receptor is immobilized onto an appropriate gel solid support. The sonic extract is loaded on the top of the gel and the column is washed to remove unbound material. The adhesin is then eluted with SDS elution buffer or a chaotropic agent, such as NaCl or KSCN, and dialyzed and analyzed by SDS-PAGE and autoradiography.
The molecular weight of the purified adhesin protein was determined by SDS-polyacrylamide gel WO 94/00149 PC 1'/US93/06016 21387fi5 electrophoresis. Figure 1 shows the sample analysis in the following lanes: 1, total outer membrane protein preparation from Haemophilus influenzae type b stained with Coomassie blue; 2, autoradiography of 35S-labeled total outer membrane proteins; 3, autoradiography of 35S-labeled adhesin protein eluted from immobilized receptor asialo-GM1; 4, autoradiography of material eluted from immobilized globoside, a nonsense glycolipid. Arrow indicates the adhesin migrating between P1 and P2 with a molecular weight of about 41 kD.

Neutralization of Adhesin Bindincr to Receptor BALB C mice were injected IP with 10 micrograms of partially purified adhesin protein (Hib OMPs) in complete Freunds adjuvant (1:1). After one month, the mice were boosted with a second IP injection (10 micrograms of protein) using incomplete Freunds adjuvant followed by a third injection 10 days later.
Antiserum was then tested for neutralizing activity against 35S-labeled Hib adhesin in a receptor binding assay. In this case, antiserum and normal mouse serum at various dilutions were incubated with 35S-labeled Hib adhesin protein for one hour at room temperature and then added to microtiter wells coated with asialo-GM1 or globoside as a negative control. After incubation of the microtiter plates for 2 hours at room temperature, the microtiter wells were washed, cut from the plates and - 74 - 21 387 fi5 radioactivity was quantified using a Beta-scintillation counter. The results are shown in Figure 2. The results show that the adhesin is immunogenic and that antibodies to the adhesin effectively neutralize the adhesin's receptor binding activity.

Identification and Cloning of an Haemophilus Influenza Adhesin 1. Membrane proteins binding to receptor.
Membrane proteins were prepared as follows. Haemophilus influenzae type b (ATCC 9795) were grown to stationary phase, pelleted, resuspended in saline buffer, and soni-cally disrupted. This material was then centrifuged at 12,000 x g for 15 min, and the supernatant was centrifuged at 100,000 x g for 1 h. The resultant pellet contained Haemophilus membranes, which were resuspended in saline and tested for adhesin activity as described in Krivan, et al.
Proc. Natl. Acad. Sci. USA, 85:6157-6161 (1988). Briefly, membranes were prepared from [35S] methionine metabolically-labelled cells (1 micro-Ci/ml of media). Glycolipids were resuspended in chloroform: methanol (1:1, vol:vol) and seri-ally diluted into 96-well microtiter plates. These plates were allowed to dry, washed 5 times with Tris/BSA (25 mMTris, pH7.5, 1% bovine serum albumin), then 2 X 106 CPM of labelled membranes were added to each well and incubated at room temperature for 2 h. The plates were then washed with Tris/BSA 5 times, and the individual wells cut out and counted on a scintillation counter to determine the amount of CPM bound to each well. This showed that Hi membranes bound similar to Hi whole cells.
2. Production of monoclonal antibodies that inhibit adhesion of Haemophilus. Balb/c mice were immunized with membranes from Haemoghilus influenzae type b (ATCC 9795), and their sera was tested for the development of antibody that inhibited membranes from binding to receptor (Figure 2). Spleens from these mice were used to isolate spleno-cytes for fusion with SP2/o-AG14 (ATCC CRL 8287) mouse myeloma cells according to Harlow, et al., Antibodies: A
Laboratory Manual (Cold Spring Harbor Laboratory, Cold Spring Harbor, NH) (1988). Seven hundred and fifty positive fusion hybridoma cultures from four separate fusions were screened for the production of antibody that reacted on ELISA with membranes. The ELISA was performed as follows.
Membranes containing 1 microgram of protein were used to coat 96-well microtiter plates. The coated wells were washed with PBS (phosphate buffered saline, 10 mM sodium phosphate, pH 7.5, 167 mM sodium chloride), then incubated with 100 microliters of hybridoma culture supernatant. The wells were washed, incubated with 100 microliters of secon-dart' goat anti-mouse antibody conjugated with horseradish peroxidase for 1 h, then bound antibody was detected colorimetrically (Biorad). Seventy-five membrane-reactive hybridoma cultures were then tested for the ability to inhibit membrane binding (Figure 3). Hybridoma culture supernatants were incubated with 4 x 106 CPM of [3sS] methionine labelled membranes for 1 h at room temperature. This mixture was then added to serial dilutions of receptor bound passively to 96-well microtiter plates and assayed for binding. Two classes of inhibiting antibodies were identified. One class, such as the antibodies designated HiblO, completely inhibited binding and were subsequently shown to react with the lipooligo-saccharide component of these membranes. The second class of antibodies, such as those designated Hib30 and Hib43, partially inhibited binding.
3. Identification of the putative adhesin. The hybridoma cultures which produced antibodies that partially inhibited binding were cloned by limiting dilution to obtain stable cell lines according to Harlow, E. and R. Lane (1988) "Antibodies: A Laboratory Manual", pp. 139-244, Cold Spring Harbor, NY. Large amounts of antibody were produced in the ascitis fluid of Balb/c mice, and the class of each antibody was determined according to Harlow et al. The antibodies were then used on Western blot of Haemophilus membranes and whole cells to identify a potential protein adhesin according to Harlow et al. All of these antibodies recognized an approximate 47 kDa protein, Hin47, by this technique (Figure 4). Western blot analysis with these antibodies according to Harlow et al. allowed further characterization of this protein. Several lines of evidence suggested that this protein is located on the surface of Haemophilus, as would be expected for a functional adhesin. First, the ability of whole cells to bind the receptor was inhibited by these antibodies in an 77 - 21 387 fi5 assay as described above for membrane binding inhibition but using radiolabeled whole cells (4 X 106 CPM/well). Second, the Hin47, a immunoreactive protein, was degraded when whole cells were treated with proteinase K (Figure 4). Briefly, whole cells were grown to stationary phase, pelleted by centrifugation (12,000 x g), and resuspended in PBS. Serial dilutions of proteinase K were added to the cells and incubated for 1 h. Cells were then mixed with SDS-PAGE
sample buffer according to Laemmll, Nature (London), 227:680-685 (1970), boiled, and separated on SDS-PAGE. This gel was then Western blotted to detect the presence of an immunoreactive Hin47 protein. Third, iodinated whole cells contained a radiolabeled Hin47 protein that could be immunoprecipitated from solubilized proteins by the anti-adhesin antibodies. Briefly, whole Haemophilus were grown to stationary phase and pelleted by centrifugation. Cells were resuspended in PBS and iodinated with Iodo-GenTM
according to the manufacturer's recommendation. Cells were then solubilized in radioimmune precipitation buffer (RIPA
buffer, 20 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, to Nonidet P-40, 1% deoxcholate, 0.1% SDS, 1 mM PMSF), and then incubated with GammabindTM beads overnight at 4°C.
The beads were then pelleted by centrifugation (2000 x g, 5 min), washed 5 times with PBS containing 0.05% Tween-20, and resuspended in SDS-PAGE sample buffer. This sample was then separated by SDS-PAGE, and the gel was dried and autoradiographed. This showed the Hin47 protein was accessible to iodination. Fourth, whole cells and membranes that were extracted repeatedly with 1% Triton X-100 lost this Hin47 immunoreactive protein. This was _ 78 _ 21 387 65 performed by taking whole cells or membranes and mixing them with the detergent, pelleting the material by centrifugation (12,000 x g for membranes and 2000 x g for whole cells), and taking the supernatant. This material (pellet and supernatant) was separated by SDS-PAGE gel, Western blotted, and the presence of Hin47 protein detected with Hib 43 antibody in the soluble fraction (supernatant).
4. Cloning and sequencing of the gene that encodes the 47 kDa adhesin. Cloning methods were performed by standard procedures as described by Maniatis et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) (1982). Total DNA from Haemophilus influenza type b strain ATCC 9795 was isolated and partially digested with the restriction enzymes Eco R1 according to the manufacturer's recommen-dations (Boerhinger-Manheim). DNA fragments 4-15 kbp in length were isolated on a sucrose gradient and ligated to Eco R1-digested Lambda ZAPII arms as supplied by Strata-gene, Inc. This ligation was then packaged into phage particles and used to transfect the Escherichia coli host strain, XL-1 (according to Statagene protocol) to obtain phage plaques which express Haemo~hilus proteins. These plaques were used in an immunoblot screen with Hib 43 using a Stratagene picoBlueTM detection kit. Positive reacting plaques were purified and used to induce the production of a plasmid through the use of the helper phage 8408 (according to Stratagene protocol). These plasmids carried the Haemophilus insert DNA which encoded the Hin47 immunoreactive protein. The restriction map for one of these plasmids, designated pMC101, is shown in v'VO 94/00149 213 8 7 fi ~ PCT/US93/06016 _ 79 -Figure 5. All plasmids which expressed the Hin47 protein contained the 10.5 kbp DNA from Hi. The location of the gene encoding this protein was determined by deletion analysis of pMC101. Deletion analysis was performed by generation of subclones of pMC101 containing various restriction fragments in the vector pSK(-) (Stratagene).
These subclones are represented on Figure 5 with an indication of whether each expresses a Hib 43 immunoreactive protein. The deletion analysis suggested that the Hin47 was encoded by a gene which was bounded by an approximate 2.4 kbase Pst 1 to BamHl fragment.
Therefore, sequence analysis of this entire region was performed using the dideoxy double stranded sequencing methods of Sanger et al., "Determination of Nucleotide Sequences in DNA," Science, 214:1205-1210 (1981), with Sequenase~ brand of DNA polymerase, (US Biochemicals).
The results of this analysis are represented in Figures 7A
and 7B. An open reading frame (ORF) was identified which would encode an approximate 49 kDa protein, comprising 463 amino acids, located between nucleotide 115 and 1503.
Analysis of the amino acid sequence predicted by this ORF
indicated that this protein contains a putative signal sequence of approximately 2.5 kDa and 25 amino acids. This could result in a mature protein of approximately 47 kDa and 438 amino acids as indicated by prior Western blot analysis. This ORF was designated hin47. The expression of the Hin47 protein was similar irrespective of the orientation of the gene with respect to the beta-galactosidase promoter contained in pSK(-), indicating this protein is expressed in E. coli under its own promoter.
Membranes of E. coli clones that expressed this protein were compared with the membranes of E. coli that did not express this protein (Figure 6). The binding curves for both membranes preparations demonstrate that this protein confers upon E. coli the ability to bind to the receptor with high affinity, like Haemophilus.
5. The Hin47 adhesin is a novel protein. A series of major integral membrane proteins has been characterized by several investigators (Gonzales et al., Infect. Immun., 55:2993-3000 (1987). These include P1, which is approxi-mately 43 kDa, and P6, which is approximately 18 kDa. The Hin47 adhesin was analyzed to insure that it was not any of these previously characterized proteins. Using an E. coli clone that expressed P1 or P6, neither clone reacted with Hib 43, demonstrating that this antibody does not recognize either of these proteins. Additionally, since the P1 protein is similar in size to the Hin47 adhesin, we demonstrated by heat modification that the Hin47 adhesin was not P1. The E. coli which expressed P1 was separated by SDS-PAGE after treatment at room temperature or 100°C.
P1 has previously been shown to be heat modifiable (Gonzales et al.). After treatment at 100°C, the protein migrates at about 43 kDa, while after treatment at room temperature, P1 migrates at about 32 kDa. The Hin47 protein was shown not to be heat modifiable. A comparison of the sequence of the 2.4 kbase Pst 1 to BamHl fragment of pMC102 confirmed that hin47 has no homology with the gene that encodes P1.
6. Purification of the Hin47 adhesin. The Hin47 protein was purified to homogeneity using the monoclonal ~.WO 94/00149 PCT/US93/06016 antibody Hib 43 as an immunoabsorbent according to Krivan et al., Tnf. and Immun., 55:1873-1877 (1987). Briefly, antibody was coupled to cyanogen activated sepharose 4CL
beads (Pharmacia) according to the manufacturer's recommendation. A 4 ml column containing about 8 mg of coupled antibody was used. The Hin47 protein was produced by XL-1/pMC101 grown to stationary phase in a 4 L culture in Luria Broth. The cells were pelleted by centrifugation (12,000 x g, 15 min), resuspended in PBS, and sonicated.
The sonicate was pelleted by centrifugation (12,000 x g. 15 min) and the supernatant pelleted by centrifugation (1000,000 x g, 1 h). The resultant membrane pellet was resuspended in 0.5% octylglucopyranoside (Sigma Chemical) and pelleted by centrifugation (100,000 x g, 1 h). The supernatant was exhaustively dialyzed against 50 mM Tris, pH 8.5 and applied to a DEAF-sepharoseT'' column.
A fraction containing Hin47 was eluted from the column using 125 mM NaCl, 50 mM Tris, pH 8.5. This fraction was dialyzed against PBS, then applied to the antibody column. The column was then washed with PBS, and bound protein was eluted with 100 mM glycine, pH 2.8 and immediately neutralized. This material was dialyzed against PBS and analyzed by separation on SDS-PAGE. The gel was stained by silver (Biorad). The Hin47 protein appeared as a single species, indicating purification to homogeneity.
7. Conservation of the Hin47 adhesin with the Haemophilus influenza serotype. The conservation within the Haemophilus influenza species and genus was analyzed using Western blotting of whole cells and Southern blotting using DNA isolated from whole cells. Table 5 contains the results obtained from this study. Seven non typable H.
influenza strains, three serotype b strains and three clinical H. influenza strains that have not been typed all reacted with a monoclonal antibody (Hib 43) specific for that 47 kDa Hin47. The DNA from all these strains also hybridized with a DNA probe of the entire hin47 gene. This hybridization was found at high stingency levels (less than 5% mismatch) which confirmed that strong conservation of this gene within the H. influenza genus. A second measure of the close relationship between these sequences was demonstrated by PCR analysis. Primers that hybrized with the immediate 5' and 3' regions were able to amplify a DNA
fragment from each strain that was identical in size to the hin47 gene from strain ATCC 9795, the strain that was used to originally clone hin47. The PCR analysis was performed using GeneAmp-PCR kit with AmpliTaq~ brand Taq-polymerase (Perkin-Elmer Cetus).

Coublincr Synthetic PRP to Protein Using the OliQOmers of Compound 14 A solution of human serum albumin (41 mg, 1.0 micro-mol) in phosphate buffer (0.1 M, pH 8.0, 1.5 ml) was mixed with a solution of Compound 14 (40 micro-mol), then, after 1 hr., sodium cyanoborohydride (26 mg, 410 micro-mol) was added. The mixture was gently stirred at 37°C for 4 days, then ultrafiltrated, diluted with water, and ultrafiltrated again. The retained material was lyophilized and purified by gel filtration on Bio-Gel P4~

'~O 94/00149 PCT/US93/06016 The appropriate fractions were collected and lyophilized.
The degree of functionalization (as haptens/protein molecule) was estimated by a combination of Lowry protein determination and orcinol ribose determination. Generally, a value of 5-10 haptens/protein molecule was obtained.
U_sincr Oliaomers of Compound 16 A solution of Compound 16 (100 micro-mol) in a mixture of aqueous sodium hydroxide (0.5 M, 6.0 ml), ethanol (4.0 ml), and acetic acid (180 microliters) was stirred while thiophosgene (30 microliters) was added.
After 10 min., the mixture was partitioned between ethyl acetate and water, the aqueous phase was concentrated to half the volume and added to a solution of human serum albumin (164 mg, 4.0 micro-mol) in borate buffer (O.1M, pH
9.3, 6 ml). The pH was adjusted to 9.5 and the mixture was gently stirred overnight at room temperature, then ultrafiltrated, diluted with water, and ultrafiltrated again. The retained material was lyophilized and purified by gel filtration on Bio-Gel P4~ The appropriate fractions were collected and lyophilized. The degree of functionalization ~(as haptens/protein molecule) was estimated by a combination of Lowry protein determination and orcinol ribose determination. Generally, a value of 10-20 haptens/protein molecule was obtained.

Comparison of the deduced amino acid sectuence of Hin47 from H influenzae type b strain 9795 and five phvloQenetically diverse non-tvpable strains ,~y ;::

~~ 21 387 65 - g4 - _ A candidate subunit vaccine for otitis media must be highly conserved between the commonly isolated strains of non-typable H. influenzae. To assess the conservation of Hin47 in non-typable H. influenzae at the amino acid level, the hin47 genes from five phylogenetically diverse strains (Musser et al., Infection and Immunity, 52:183-191 (1986) were cloned by polymerase chain reaction (PCR) methods. DNA from each strain was used for amplification with a primer 5' and a primer 3' to the structural hin47 gene. See Hinf3 and Hinf4 on Figures 7A and 7B. The amplification was performed using an Amplitaq AmpliTaqTM
(Roche Molecular Systems) DNA amplification kit (Perkin Elmer Cetus, Norwalk, Conn.) by the methods provided by the manufacturer. Each amplified gene was cloned into the PCR
cloning vector PCR II (Invitrogen, San Diego, CA) according to the manufacturer's methods and subsequently subjected to DNA sequence analysis. The sequence analysis was performed using the dideoxy double stranded sequencing method of Sanger et al., Science, 214:1205-1210 (1981) with the Sequenase ° brand of DNA polymerase (U. S. Biochemicals).
The deduced Hin47 amino acid sequences obtained from each hin47 gene are compared in Figure 8. Table 6 summarizes the results from this comparison. All non-typable hin47 genes were extremely highly conserved. These results strongly suggest that immunity against Hin47 expressed by the gene from one strain would protect against challenge from all H. influenzae strains.

Animal studies to evaluate Hin47 as a subunit vaccine A candidate subunit vaccine must be strongly im-munogenic and have the ability to generate a protective immune response. Several animal experiments were performed to assess these properties for Hin47.
Immunogenicity studies were performed in Balb/c mice to assess the IgG response to various doses of purified Hin47 protein. Animals received three injections of antigen in the presence of aluminum phosphate as an adjuvant. Anti-Hin47 IgG titers were determined by an enzyme immunoassay using purified Hin47 as the detecting antigen. These results are summarized in Table 7. The minimum challenge dose of 1 ug per injection of protein elicited a strong IgG response in mice. These data demonstrate the highly immunogenic nature of Hin47.
The protective activity of anti-Hin47 antibodies in an infant rat model of H. influenza-mediated bacteremia was used as one measure of the ability of Hin47 to act as a protective antigen. The methodology was that Moxon et al., J. Infectious Diseases, 129:154-162 (1974) and Loeb, Infection and Immunitv, 55:2612-2618 (1987). Rabbit anti-Hin47 antiserum was generated by immunizing a rabbit with 3 doses of 100 ug of purified Hin47 in the presence of com-plete Freund's adjuvant on day 1 and in the presence of incomplete Freund's adjuvant on day 28 and 42. The resul-tant antisera was shown not to possess significant antibody 21387fi5 against the previously demonstrated protective epitopes expressed by H. influenzae (P1, P2, and P6 proteins and polyribitol phosphate). Groups of five 5-day old infant rats were injected subcutaneously with either rabbit anti-s Hin47 antibody, prebleed serum, or saline. After 24 hours, the infant rats were challenged with 200 colony forming units (CFU) of the virulent H. influenzae Minn A strain (Munson and Grass, Infection and Immunity, 56:2235-2242 (1988) by intraperitoneal injection. The results from these experiments are summarized in Table 8. The results are expressed as the average CFU/ 0.1 ml of blood for each group of animals, the number of animals with bacteremia in each group, and the percentage of bacteremia compared to the saline control. Animals receiving Hin47-specific antibodies were significantly reduced in bacteremia, and 3 of 5 animals had no detectable organisms in their blood.
The second animal disease model used to assess the efficacy of Hin47 as a vaccine candidate was the chincilla otitis media model. The methodology was that Bakaletz et al., Infection and Immunity, 57:3226-3229 (1989). Chincilla were immunized with three doses (50 ug per dose) of Hin47 protein in the presence of com-plete Freund's adjuvant on day 1, and in the presence of incomplete Freund's adjuvant on day 28 and 42. Two control animals were injected with saline. On day 47, the chincillas were challenged with 2,000 CFU of a non-typable H. influenzae strain designated strain 12. Ear infection was monitored by otoscopic examination and tympanometry on day 1, 2, and 6 post-infection. Fluid was WO 94/00149 ~ ~ ~ ~ ~ ~ PCT/US93/06016 _ 87 _ collected through epitympanic bulla by injecting 0.2 ml of saline into the middle ear cavity and then aspirating the fluid. The fluid was plated on chocolate agar plates and incubated at 37°C overnight. Positive control protection animals were either an animal that had recovered from an ear infection or animals that were immunized with heat killed strain 12 whole cells. The results are summarized in Table 9. Two of four Hin47-immunized animals were negative by typanogram analysis and had significantly reduced bacteremia at day 2. These data suggest that Hin47 has protective value in the active protection chincilla model for human otitus media.

_88_ TABLE 1: PREPARATION OF MONOMERS FOR SOLID PHASE SYNTHESIS
OF PRP FRAGMENTS
HO ~ OMe ' (1) O O
BnO ~ OMe ABO O OMe (4) (2) OMe Bn0 (5) HO
OBz OBz OBn OBn (3) OBn OAllvl Bn0 (6) ,quo o (7) Bn0 O OBn Bn0 O OBn (8) OBn ~ OBn OBn H OBn OMMTr OMMTr HO OBn O - p - O OBn O' Monomer for chain initiation Monomer for chain elongation in solid phase synthesis in solid phase synthesis -89_ TABLE 2: PREPARATION OF SPACER-CONTAINING MONOMERS FOR
CHAIN TERMINATION IN THE SOLID PHASE SYNTHESIS
Bno (6) o A110 O-t-OMe Ph p~ O~R
Bn0 O (CH2)p OR Bn0 O r _ ~ -P
(9) (11) OH OBn OH OBn 'OR3 R4 Bn0 O O'(CH2)p OR3 Bn0 O
(10) i (12) O =P-O OBn O =P-O OBn O' O' Monomers for chain termination in solid phase synthesis WO 94/00149 _ _ PCT/US93/06016 9~
TABLE 3: OLIGOMERS OBTAINED AFTER COMPLETED SOLID-PHASE SYNTHESIS
(CH2)p Bn0 O O Bn0 O O~ ~ OR
OBn OBn OR
OBn O (13) H O OHn O- P O OBn I I
O n (CH2)p O
HO O O HO O O
OH
off _ (14) OH
H O OH O- P O OH
I I
O n OR, ALTERNATIVELY:
Bn0 O O OBn Bn0 O O ~O~ 4 J - R
P
OBn _ OBn O (15) H O OBn O- P O OBn I I
O n HO O O OH HO O O ~ O~ ~H~
JP
OH
off ~ (16) H O OH O-P O OH
O n WO 94/00149 _ 91- PCT/ US93/06016 TABLE 4: STRUCTURE OF CONJUGATES BETWEEN SYNTHETIC PRP FRAGMENT
AND ADHESIN PROTEIN.
~(CH2)pVNH - X
HO O O OH HO O
OH
OH O
I
H O OH O-I ~ O OH
O n m (17) HO O O OH HO O O~O~y~ 'NH- X
OH _ r _ J \'C
OH
H O OH O- p O OH
II
O n m (18) N
'O
~i .Q i~
x 3 + + + + + + + + + + + + + +
d .i r~ .

U

ro dl o x c x 3 + + + + + + + + + + + + + +

~i .Ql~ ~1~2 ~t~i ro roro roro ro~

a~ w a o, o,a. a a w >~ >,~, >~>. >,>, >, ~ +~~ +~~ v 1 O O O N Er E-i .~

a~ .a ~ ~az z z z z z z z z z z C1O tl1r'1N r1 tC1 f"1O G~ stO 01C1 e-i0~rl rlri c O~ lf1N O O C1 r'1!1 ~'~i'ro roN

r ~ ~ r~o en r 11 chc~ a w-IU U U 41 d ro ov r~~ ~t ~ r, c~W ~ oo-~ ~ -~ C

O ~ U U U U U U U U U U c c c -~

U U U U U U U U U U ~ ~ w >

cn En H E-~N EaEn N N EiE~.-i.~~ O

a a ~ a a a a W 2 a U U U .tt W

O b ri N

i~ c d O

d W

N c ~~' o b roro roro roro rob ro rob U N N N N N N N N N N N N N

c c c c O c c c c c c C

r1 01CiCl d Cl C1O d Cl4l d 'U N

~ ri O O O O O O O O O O O

IC1N .c r-1r-Irl .-1~ r-1r-1r~1rirl r1r-1c rl Oa W W W W W W W W W W W W E..

d c O c c c c c c c c c c c C O

rl ro ~ W ~.'~~'"~'~ ''~W ~-~W W W -a N

o' y _ ~ ,_, , , H o x x x x x x x x x x x x x x ;~~~g76~

Table 6: Conversation of Hin47 among Haemophilus influenzae Strain Type Nucleotide Amino acids homology homology 9795 b 100 100 1161 NT 99.8 100 3 690 NT 99.8 100 1636 NT 99.7 99.8 9333 NT 95.0 98.0 3 63 9 NT 94. 6 98.0 Table 7: Anti-Hin47 IgG titers in murine sera determined by EIAs.
Dose of Hin47 (pg) REACTIVE
TITERS

Prebleed Bleed Bleed 2 Bleed 3 Bleed 4 p <200 <200 <200 <200 <200 1 <200 12,800 43,520 696,320 819,200 3 <200 28,160 81,920 696,320 819,200 <200 20,480 112,640 1,187,840 2,293,760 <200 74,240 143,360 1,187,840 2,785,280 ~~1~~~'~ G5 Table 8: Protective activity of anti-Hin47 antibodies in the infant rat model.
IMMLTNOGEN cfu/0.1 ml blood % OF CONTROL
(# of animals witli bacteremia/Total # of aamals) Anti-Hin47 1,560 (2/S) 10 Ab Prebleed 17,040 (S/5) 110 Saline 15,440 (5/5) 100 Table 9: Protective Ability of Hin47 Against Non-Typeable H.
influenzae Infection in Chinchillas Chinchilla Immunogen Middle ear infection Day 2 Day 6 (#) TympanogramBacteria Tympanogram cfu/0.1 mL

8 Recovered from - 0 -Hi stain 12 infection 9 Hin47 + 10 x 10 +

- 240 +

11 + 6 x 105 +

12 - 0 +

13 Saline + 4 x 10 5 +

14 + 6x 105 +

Heat-inactivated- 0 -16 strain 12 - 0 -w ~~387G~

SEQUENCE LISTING
(1) GENERAL INFORMATION:
(i) APPLICANT: MicroCarb II1C.
(ii) TITLE OF INVENTION: Adhesin-Oligosaccharide Conjugate Vaccine for Haemophilus Influenzae (iii) NUMBER OF SEQUENCES: 2 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Pennie & Edmunds ( B ) STREET : 1155 Avenue of the Americas (C) CITY: New York (D) STATE: New York (E) COUNTRY: VSA
(F) ZIP: 10036 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release X1.0, Version 41.25 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER: US 07/903,079 (B) FILING DATE: 22-JUN-1992 (C) CLASSIFICATION:
(vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 07/810,966 (B) FILING DATE: 20-DEC-1991 (vii) PRIOR APPLICATION DATA:
(A) APPLICATION NUMBER: US 07/631,698 (B) FILING DATE: 21-DEC-1990 (viii) ATTORNEY/AGENT INFORMATION:
( A ) NAME : Geraldine F. &~ldwin (B) REGISTRATION HUMBER: 3J,232 (C) REFERENCE/DOCKET NUMBER: 7969-402 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (212) 790-9090 (B) TELEFAx: (212) 869-X864/9741 SUBSTITUTE SHEET

~l:~g~ 6~ _ - (2) INFORMATION FOR SEQ ID NO:1:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1611 base pairs (H) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (11) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(ix) FEATURE:
(A) NAME/KEY: CDS
(B) IACATION: 115..1503 (xi) SEQUENCE DESCRIPTION: SEQ ID N0:1:

TTGTCTTACA ATTT
GACCACGTTA ATG
TCTGAAATTT

Met ATT GCA AGT

LysLys ThrArg Phe Val Leu Asn Ser LeuGly Leu Val Ile Ala Ser 5 lp 15 ACT TTG GTT

LeuSer ThrSer Phe Val Ala Gln Ala ProSer Phe Ser Thr Leu Val GAA AAA GCC

GluGln AsnSer Leu Ala Pro Met Leu ValGln Pro Val Glu Lys Ala AAA GTA TCT

ValThr LeuSer Val Glu Gly Lys Ala AspSer Arg Pro Lys Val Ser AAA TTC GAT

PheLeu AspAsp Ile Pro Glu Glu Phe PhePhe Gly.AspArg Lys Phe ~TTTGCC GAACAA TTT GGT GGA CGT GGA AAGCGT AAC CGT 405 GAG TCA TTC

PheAla GluGln Phe Gly Gly Arg Gly LysArg Asn Arg Glu Ser Phe SUBSTITU1'E ~HEE~

~'I3~765 WO 94/00149 _9~_ PCT/US93/06016 GGT GTC ATT
ATT AAT GG AGC
AAA GGC TAT
GTT TTA

Gly Leu Gly Ser Gly Val Ile Ile Asn Ala Ser Lys Gly Tyr Val Leu GTT CAA TTA

Thr Asn Asn His Ile Asp Gly Ala Asp Lys Ile Thr Val Val Gln Leu GAA GAA GA

Gln Asp Gly Arg Phe Lys Ala Lys Leu Val Gly Lys Asp Glu Glu Gln TTA AG GAA

Ser Asp Ile Ala Val Gln Leu Glu Lys Pro Ser Asn Leu Leu Thr Glu GAT ACT GTT

Ile Lys Phe Ala Ser Asp Lys Leu Arg Val Gly Asp Phe Asp Thr Val CG GGT ATT

Ala Ile Gly Asn Phe Gly Leu Gly Gln Thr Val Thr Ser Pro Gly Ile GGT TAT GAA

Val Ser Ala Leu Arg Ser Thr Gly Ser Asp Ser Gly Thr Gly Tyr Glu ACC GGT GGT

Asn Tyr Ile Gln Asp Ala Ala Val Asn Arg Gly Asn Ser Thr Gly Gly CTA GCA ATT

Ala Leu Val Asn Asn Gly Glu Leu Ile Gly Ile Asn Thr Leu Ala Ile GGT CG AGT

Ile Ser Pro Ser Gly Asn Ala Gly Ile Ala Phe Ala Ile Gly Pro Ser AAT GA GTG

Asn Gln Ala Ser Leu Val Gln Gln Ile Leu Glu Phe Gly Asn Gln Val CTT GAT TTA

Arg Arg Gly Lev Gly Ile Lys Gly Gly Glu Leu Asn Ala Leu Asp Leu AAT GAA

Ala Lys Ala Phe Val Ser Ala Gln Gln Gly Ala Phe Val Asn Ser Glu _98_ TCT AAA CTT

Val Val Pro Lys AlaAlaGlu AlaGly LysAlaGlyAsp Ser Lys Leu ATG AAA AGT

Ile Ile Thr Ala AsnGlyGln IleSer PheAlaGluIle Met Lys Ser GCA GCA GAG

Arg Ala Lys Ile ThrThrGly GlyLys IleSerLeuThr Ala Ala Glu GGC GAC ATG

Tyr Leu Arg Asp LysSerHis ValLys LysLeuGlnAla Gly Asp Met CAA AAA TTG

Asp Asp Gly Ser LeuSerSer ThrGlu ProAlaLeuAsp Gln Lys Leu AAA GCT GTT

Gly Ala Thr Leu AspTyrAsp LysGly LysGlyIleGlu Lys Ala Val ATC ACA AAA ATT CCT7~ATTCG GCTGCA CGTGGTTTAAAA 1365 CAA CTG CAA

Ile Thr Lys Ile ProAsnSer AlaAla ArgGlyLeuLys Gln Leu Gln ATT AAT ATG

Ser Gly Asp Ile IleGlyIle ArgGln IleGluAsnIle Ile Asn Met AAA ACT TCA

Arg Glu Leu Asn ValLeuGlu GluPro AlaValAlaLeu Lys Thr Ser Asn Ile Leu Arg Gly Asp Ser Asn Phe Tyr Leu Leu Val Gln AGCAATCGTT C'I'T'AACAAAT CCACCACAAA TTCTAACCGC ACTTTGTT 1611 (2) INFORMATION FOR SEQ ID N0:2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 463 amino acids (8) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein z~~~7s~
WO 94/00149 _9 9 _ PC1'/US93/06016 (xi) SEQUENCE DESCRIPTION: SEQ ID N0:2:
Met Lys Lys Thr Arg Phe Val Leu Asn Ser Ile Ala Leu Gly Leu Ser Val Leu Ser Thr Ser Phe Val Ala Gln Ala Thr Leu Pro Ser Phe Val Ser Glu Gln Asn Ser Leu Ala Pro Met Leu Glu Lys Val Gln Pro Ala Val Val Thr Leu Ser Val Glu Gly Lys Ala Lys Val Asp Ser Arg Ser Pro Phe Leu Asp Asp Ile Pro Glu Glu Phe Lys Phe Phe Phe Gly Asp Arg Phe Ala Glu Gln Phe Gly Gly Arg Gly Glu Ser Lys Arg Asn Phe Arg Gly Leu Gly Ser Gly Val Ile Ile Asn Ala Ser Lys Gly Tyr Val Leu Thr Asn Asn His Val Ile Asp Gly Ala Asp Lys Ile Thr Val Gln Leu Gln Asp Gly Arg Glu Phe Lys Ala Lys Leu Val Gly Lys Asp Glu Gln Ser Asp Ile Ala Leu Val Gln Leu Glu Lys Pro Ser Asn Leu Thr Glu Ile Lys Phe Ala Asp Ser Asp Lys Leu Arg Val Gly Asp Phe Thr Val Ala Ile Gly Asn Pro Phe Gly Leu Gly Gln Thr Val Thr Ser Gly Ile Val Ser Ala Leu Gly Arg Ser Thr Gly Ser Asp Ser Gly Thr Tyr Glu Asn Tyr Ile Gln Thr Asp Ala Ala Val Asn Arg Gly Asn Ser Gly Gly Ala Leu Val Asn Leu Asn Gly Glu Leu Ile Gly Ile Asn Thr Ala Ile Ile Ser Pro Ser Gly Gly Asn Ala Gly Ile Ala Phe Ala Ile Pro Ser Asn Gln Ala Ser Asn Leu Val Gln Gln Ile Leu Glu Phe Gly Gln Val Arg Arg Gly Leu Leu Gly Ile Lys Gly Gly Glu Leu Asn Ala Asp Leu Ala Lys Ala Phe Asn Val Ser Ala Gln Gln Gly Ala Phe Val Ser Glu Val Val Pro Lys Ser Ala Ala Glu Lys Ala Gly Leu Lys Ala Gly Asp Ile Ile Thr Ala Met Asn Gly Gln Lys Ile Ser Ser Phe Ala Glu Ile Arg Ala Lys Ile Ala Thr Thr Gly Ala Gly Lys Glu Ile Ser Leu Thr Tyr Leu Arg Asp Gly Lys Ser His Asp Val Lys Met Lys Leu Gln Ala Asp Asp Gly Ser Gln Leu Ser Ser Lys Thr Glu Leu Pro Ala Leu Asp Gly Ala Thr Leu Lys Asp Tyr Asp Ala Lys Gly Val Lys Gly Ile Glu Ile Thr Lys Ile Gln Pro Asn Ser Leu Ala Ala Gln Arg Gly Leu Lys Ser Gly Asp Ile Ile Ile Gly Ile Asn Arg Gln Met Ile Glu Asn Ile Arg Glu Leu Asn Lys Val Leu Glu Thr Glu Pro Ser Ala Val Ala Leu Asn Ile Leu Arg Gly Asp Ser Asn Phe Tyr Leu Leu Val Gln

Claims (16)

WE CLAIM:

1. A recombinant protein produced by a host cell transformed with an isolated or substantially purified DNA
comprising nucleotides designated 115 to 1503 in Sequence ID No. 1 encoding an H. influenzae protein operably linked to appropriate regulatory control nucleic acid sequences that are capable of effecting expression of said DNA sequence in said transformed host cell.

2. A recombinant protein produces by a host cell transformed with an isolated or substantially purified DNA
comprising nucleotides designated 190 to 1503 in Sequence ID No. 1 encoding an H. influenzae protein operably linked to appropriate regulatory control nucleic acid sequences that are capable of effecting expression of said DNA in said transformed host cell.

3. The protein of claim 1 or 2 which is a fusion protein.

4. A purified H. influenzae protein comprising Sequence ID No. 2.

5. A purified H. influenzae protein with a molecular weight of about 47,000 daltons which comprises the amino acids designated 26 to 463 in Sequence ID No. 2.

6. A vaccine for protecting a mammal against H.
influenzae comprising an immunologically effective amount of the protein of claim 1-5 in a pharmaceutically acceptable carrier.

7. An immunogenic composition comprising an immunologically effective amount of the purified H. influenzae protein of claim 1-5, which composition induces an immune response to H. influenzae in a mammal and an acceptable carrier.

8. An isolated or substantially purified DNA sequence encoding an H. influenzae protein, wherein said sequence comprises the nucleotides designated 115-1503 or 190-1503 in Sequence ID No. 1.

9. A method for producing an isolated DNA sequence comprising Sequence ID No. 1 encoding an H. influenzae adhesin protein comprising the steps of:
screening a genomic library containing the DNA of H.
influenzae, said library comprising clones which contain different sequences of said DNA which been operably and recoverably inserted into a vector, each of said vectors containing only one sequence of said DNA, each of said clones expressing said DNA, by contacting the clones comprising said library with a monoclonal antibody to said adhesin protein or a receptor for said adhesin protein to identify a clone that binds to said antibodies or said receptor; isolating said clone; and recovering the exogenous DNA sequence from said clone.

10. An isolated or substantially purified DNA
sequence encoding the recombinant protein of claim 1 or 2.

11. An isolated or substantially purified DNA
sequence derived from the DNA sequence of claim 8 by single or multiple mutations, wherein said DNA sequence is at least 99%
homologous to the DNA of claim 8.

12. A recombinant DNA sequence comprising the DNA
sequence of claim 8 operably linked to appropriate regulatory control nucleic acid sequences that are capable of effecting the expression of said DNA in a transformed host cell.

13. An expression vector for expressing DNA that encodes an H. influenzae protein in a compatible host cell comprising an expression vector capable of transforming a procaryotic or eucaryotic cell wherein the DNA of claim 12 has been inserted into said vector in proper orientation and correct reading frame for expression.

14. A host cell transformed with the recombinant DNA sequence of claim 12.

15. The recombinant protein produced by the transformed cell of claim 14.

16. A method for producing a recombinant H.
influenzae protein which comprises the steps of:
culturing host cells transformed by a recombinant DNA sequence comprising a DNA sequence that codes for an H. influenzae protein comprising Sequence ID No. 2 or amino acids designated 26 to 463 in Sequence ID No. 2 operably linked to appropriate regulatory control nucleic acid sequences that are capable of effecting the expression of said DNA sequence in said transformed cells; and recovering the protein.