RECOMBINANT HAEMOPHILUS INFLUENZAE PROTEIN AND NUCLEOTIDE SEQUENCE ENCODING SAME

PROTEINE D'HAEMOPHILUS INFLUENZAE DE RECOMBINAISON ET SEQUENCE NUCLEOTIDE LES CODANT

English 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.

Claims

-101-
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

-102-
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

-103-
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.

Description

CA 02138765 1999-02-16
- 1 -
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
- 2 - 21 387 65
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
- 3 -
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

- 4 - 21 387 65
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
- 5 -
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.

2138765
- 6 -
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,

- 21 387 65
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

- 2138765
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

X138765
WO 94/00149 - PCT/US93/06016
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

WO 94/00149 PCT/US93/06016
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
- 11 -
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.

WO 94/00149 PCT/US93/06016
21~g~ 6~
- 12 -
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

WO 94/00149 PCT/US93/06016
2f 3876
- 13 -
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

WO 94/00149 PCT/US93/06016
2138765
- 15 -
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:

WO 94/00149 PCT/US93/06016
~13$~ 65
- 16 -
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:
0
Dn0 O OBn
OBn
ODn
OMMTr
HO OBn

WO 94/00149 PCT/US93/06016
218765
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:

WO 94/00149 PCT/US93/06016
s
- 18 -
2
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
- 19 -
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.

WO 94/00149 PCT/US93/06016
s
- 20 -
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

PCT/US93/06016
WO 94/00149
- 21 -
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

WO 94/00149 PCT/US93/06016
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

WO 94/00149 PCT/US93/06016
- 23 -
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

WO 94/00149 PCT/US93/06016
~'~'~ ~ ~ _
- 24 -
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
- 25 -
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

- 26 - 2138769
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
- 27 -
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

WO 94/00149 PCT/US93/06016
~,~3g~ 65
- 28 -
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.

- 29 - 21 387 65
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
21 387 65
- 31 -
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.

2138765
- 32 - -
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

PCT/US93/06016
WO 94/00149
- 33 -
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

WO 94/00149 PCT/US93/06016
,~3a~ 65
- 34 -
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

- 36 - 2138765
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
- 39 -
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

- 40 - 2138765
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

2138765
- 42 -
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
- 43 -
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.

WO 94/00149 PCT/US93/06016
'13$7 65 _
- 44 -
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
- 45 -
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.

WO 94/00149 PCT/US93/06016
138't 65
- 46 -
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
- 47 -
2
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:

- 48 -
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

._
WO 94/00149 PCT/US93/06016
- 49 -
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

WO 94/00149 PCT/US93/06016
c~~~ ~~
- 50 -
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

- 51 - 21 387 65
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
- 53 -
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.

WO 94/00149 PCT/US93/06016
1'3~~ b~' _
54 -
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
- 55 -
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.
EXAMPLE 1
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)

WO 94/00149 PCT/US93/06016
2138765
- 56 -
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

WO 94/00149 PCT/US93/06016
2.38765
- 57 -
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
- 58 -
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
- 59 -
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

WO 94/00149 PCT/US93/06016
~~ 387_65
- 60 -
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
- 61 -
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

WO 94/00149 PCT/US93/06016
'~1'~87~
- 62 -
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
- 63 -
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

WO 94/00149 PCT/US93/06016
' - 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
- 65 -
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

WO 94/00149 PCT/US93/06016
- 66 -
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%) .

WO 94/00149 Z PCT/US93/06016
- 67 -
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

WO 94/00149 PCT/US93/06016
2~3g~ 6~
- 68 -
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
- 69 -
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

WO 94/00149 PCT/US93/06016
- 70 -
2138765
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

WO 94/00149 PCT/US93/06016
_ 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.
EXAMPLE 2
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
PCT/US93/06016
21~,g~ ~~
- 72 -
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
- 73 -
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.
EXAMPLE 3
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.
EXAMPLE 4
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

- 75 - 21 387 65
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

- 76 - 21 387 65
[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

- 80 - 21 387 65
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
2138765
- 81 -
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

WO 94/00149 PCT/US93/06016
21 387 65
- 82 -
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).
EXAMPLE 5
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
2138765
- 83 -
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.
EXAMPLE 6
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.

- 85 -
EXAMPLE 7 2 1 3 8 7 6 5
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.

WO 94/00149 PCT/US93/06016
_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)
0 0 0 0
1
1
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

WO 94/00149 PCT/US93/06016
-89_
TABLE 2: PREPARATION OF SPACER-CONTAINING MONOMERS FOR
CHAIN TERMINATION IN THE SOLID PHASE SYNTHESIS
Bno
(6)
o
A110 O-t-OMe
Ph
OR3 4
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
3
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)

WO 94/00149 PCT/US93/06016
-92-
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
0 O O O O H
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
0
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~
WO 94/00149 PCT/US93/06016
-9 3-
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
1
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

WO 94/00149 PCT/US93/06016
~~1~~~'~ G5
-94-
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 +
3
- 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~
WO 94/00149 PCT/US93/06016
-95-
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

WO 94/00149 PCT/US93/06016
-96
~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:
TTGTACTGCT CCGATTTCCT TTTAAACAAG ATAATTTGCT CTCCTCTTAT TGAACATTTT 60
TTTTATTTTT ATTTTGGAGT 117
TTGTCTTACA ATTT
GACCACGTTA ATG
TCTGAAATTT
Met
1
AAAAAA ACACGT TTT GTA TTA AAT AGT CTTGGA TTA GTA 165
ATT GCA AGT
LysLys ThrArg Phe Val Leu Asn Ser LeuGly Leu Val
Ile Ala Ser
5 lp 15
TTAAGC ACATCA TTT GTT GCT CAA GCC CCAAGT TTT TCG 213
ACT TTG GTT
LeuSer ThrSer Phe Val Ala Gln Ala ProSer Phe Ser
Thr Leu Val
20 25 30
GAACAA AACAGT CTT GCA CCG ATG TTA GTACAA CCT GTT 261
GAA AAA GCC
GluGln AsnSer Leu Ala Pro Met Leu ValGln Pro Val
Glu Lys Ala
35 40 45
GTCACT CTTTCC GTT GAA GGA AAA GCT GATTCT CGT CCT 309
AAA GTA TCT
ValThr LeuSer Val Glu Gly Lys Ala AspSer Arg Pro
Lys Val Ser
50 55 60 65
TTCCTA GACGAT ATT CCT GAA GAA TTT TTCTTT GGC CGT 357
AAA TTC GAT
PheLeu AspAsp Ile Pro Glu Glu Phe PhePhe Gly.AspArg
Lys Phe
70 75 80
~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
85 90 95
SUBSTITU1'E ~HEE~

~'I3~765
WO 94/00149 _9~_ PCT/US93/06016
GGT TTA GGT TCT 453
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
100 105 110
ACC AAT AAT CAT ATT GAT GGA GCT GAT AAA ATT ACC GTG 501
GTT CAA TTA
Thr Asn Asn His Ile Asp Gly Ala Asp Lys Ile Thr Val
Val Gln Leu
115 120 125
GA GAT GGG CGT TTT AAA GG AAA TTA GTG GGT AAA GAT 549
GAA GAA GA
Gln Asp Gly Arg Phe Lys Ala Lys Leu Val Gly Lys Asp
Glu Glu Gln
130 135 140 145
TG GAT ATT GG GTA GG CTT GAA AAA CG AGT AAT TTA 597
TTA AG GAA
Ser Asp Ile Ala Val Gln Leu Glu Lys Pro Ser Asn Leu
Leu Thr Glu
150 155 160
ATC AAA TTT GCT TCC GAC AAA TTA CGC GTA GGC GAT TTC 645
GAT ACT GTT
Ile Lys Phe Ala Ser Asp Lys Leu Arg Val Gly Asp Phe
Asp Thr Val
165 170 175
GG ATC GGT AAT TTT GGT TTA GGT CAA ACT GTG AG TG 693
CG GGT ATT
Ala Ile Gly Asn Phe Gly Leu Gly Gln Thr Val Thr Ser
Pro Gly Ile
180 185 190
GTT TCT GCA TTG CGT TCA AG GGT TCT GAC AGT GGC ACT 741
GGT TAT GAA
Val Ser Ala Leu Arg Ser Thr Gly Ser Asp Ser Gly Thr
Gly Tyr Glu
195 200 205
AAC TAT ATT CAA GAT GG GG GTA AAC CGC GGT AAT TCG 789
ACC GGT GGT
Asn Tyr Ile Gln Asp Ala Ala Val Asn Arg Gly Asn Ser
Thr Gly Gly
210 215 220 225
GG TTA GTC AAT AAT GGC GAA CTT ATT GGA ATT AAT ACC 837
CTA GCA ATT
Ala Leu Val Asn Asn Gly Glu Leu Ile Gly Ile Asn Thr
Leu Ala Ile
230 235 240
ATT TCT CCA AGC GGC AAT GG GGA ATT GCC TTT GCG ATT 885
GGT CG AGT
Ile Ser Pro Ser Gly Asn Ala Gly Ile Ala Phe Ala Ile
Gly Pro Ser
245 250 255
AAT CAA GCG AGC TTA GTG GA CAA ATT TTA GAA TTT GGT 933
AAT GA GTG
Asn Gln Ala Ser Leu Val Gln Gln Ile Leu Glu Phe Gly
Asn Gln Val
260 265 270
CGT CGC GGA TTG GGT ATT AAA GGG GGC GAA CTC AAT GCT 981
CTT GAT TTA
Arg Arg Gly Lev Gly Ile Lys Gly Gly Glu Leu Asn Ala
Leu Asp Leu
275 280 285
GCC AAA GCC TTT GTA AGC GCG GA GA GGT GG TTT GTA AGT 1029
AAT GAA
Ala Lys Ala Phe Val Ser Ala Gln Gln Gly Ala Phe Val
Asn Ser Glu
290 295 300 305

WO 94/00149 PCT/US93/06016
_98_
TT GTA CCG AAA GCTGCTGAA GCAGGA AAAGCGGGCGAT 1077
TCT AAA CTT
Val Val Pro Lys AlaAlaGlu AlaGly LysAlaGlyAsp
Ser Lys Leu
310 315 320
ATT ATC ACG GCG AACGGTCAA ATCTCA TTCGCTGAAATT 1125
ATG AAA AGT
Ile Ile Thr Ala AsnGlyGln IleSer PheAlaGluIle
Met Lys Ser
325 330 335
CGT GCA AAA ATC ACCACTGGT GGCAAA ATTAGCTTGACT 1173
GCA GCA GAG
Arg Ala Lys Ile ThrThrGly GlyLys IleSerLeuThr
Ala Ala Glu
340 345 350
TAC TTA CGT GAT AAATCCCAC GTTAAA AAATTACAAGCG 1221
GGC GAC ATG
Tyr Leu Arg Asp LysSerHis ValLys LysLeuGlnAla
Gly Asp Met
355 360 365
GAT GAT GGT AGC CTTTCCTCA ACTGAG CCTGCATTAGAT 1269
CAA AAA TTG
Asp Asp Gly Ser LeuSerSer ThrGlu ProAlaLeuAsp
Gln Lys Leu
370 375 380 385
GGC GCA ACA TTG GACTACGAT AAAGGC AAAGGAATTGAA 1317
AAA GCT GTT
Gly Ala Thr Leu AspTyrAsp LysGly LysGlyIleGlu
Lys Ala Val
390 395 400
ATC ACA AAA ATT CCT7~ATTCG GCTGCA CGTGGTTTAAAA 1365
CAA CTG CAA
Ile Thr Lys Ile ProAsnSer AlaAla ArgGlyLeuLys
Gln Leu Gln
405 410 415
TCG GGC GAT ATT ATTGGTATT CGTCAA ATCGAAAACATT 1413
ATT AAT ATG
Ser Gly Asp Ile IleGlyIle ArgGln IleGluAsnIle
Ile Asn Met
420 425 430
CGT GAA TTA AAT GTGCTTGAA GAACCG GCAGTTGCACTT 1461
AAA ACT TCA
Arg Glu Leu Asn ValLeuGlu GluPro AlaValAlaLeu
Lys Thr Ser
435 440 445
AAT ATT TTA CGA GGT GAC AGT AAT TTC TAT TTA TTA GTG CAA 1503
Asn Ile Leu Arg Gly Asp Ser Asn Phe Tyr Leu Leu Val Gln
450 455 460
TAATCTGCTT GATATATTTA AGAAAAAAGT CCGATCACAA TGATCGGCTT CTTTTTATGC 1563
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
1 5 10 15
Val Leu Ser Thr Ser Phe Val Ala Gln Ala Thr Leu Pro Ser Phe Val
20 25 30
Ser Glu Gln Asn Ser Leu Ala Pro Met Leu Glu Lys Val Gln Pro Ala
35 40 45
Val Val Thr Leu Ser Val Glu Gly Lys Ala Lys Val Asp Ser Arg Ser
50 55 60
Pro Phe Leu Asp Asp Ile Pro Glu Glu Phe Lys Phe Phe Phe Gly Asp
65 70 75 80
Arg Phe Ala Glu Gln Phe Gly Gly Arg Gly Glu Ser Lys Arg Asn Phe
85 90 95
Arg Gly Leu Gly Ser Gly Val Ile Ile Asn Ala Ser Lys Gly Tyr Val
100 105 110
Leu Thr Asn Asn His Val Ile Asp Gly Ala Asp Lys Ile Thr Val Gln
115 120 125
Leu Gln Asp Gly Arg Glu Phe Lys Ala Lys Leu Val Gly Lys Asp Glu
130 135 140
Gln Ser Asp Ile Ala Leu Val Gln Leu Glu Lys Pro Ser Asn Leu Thr
145 150 155 160
Glu Ile Lys Phe Ala Asp Ser Asp Lys Leu Arg Val Gly Asp Phe Thr
165 170 175
Val Ala Ile Gly Asn Pro Phe Gly Leu Gly Gln Thr Val Thr Ser Gly
180 185 190
Ile Val Ser Ala Leu Gly Arg Ser Thr Gly Ser Asp Ser Gly Thr Tyr
195 200 205
Glu Asn Tyr Ile Gln Thr Asp Ala Ala Val Asn Arg Gly Asn Ser Gly
210 215 220
Gly Ala Leu Val Asn Leu Asn Gly Glu Leu Ile Gly Ile Asn Thr Ala
225 230 235 240
Ile Ile Ser Pro Ser Gly Gly Asn Ala Gly Ile Ala Phe Ala Ile Pro
245 250 255
Ser Asn Gln Ala Ser Asn Leu Val Gln Gln Ile Leu Glu Phe Gly Gln
260 265 270

WO 94/00149 -10 0 - PCT/US93/06016
Val Arg Arg Gly Leu Leu Gly Ile Lys Gly Gly Glu Leu Asn Ala Asp
275 280 285
Leu Ala Lys Ala Phe Asn Val Ser Ala Gln Gln Gly Ala Phe Val Ser
290 295 300
Glu Val Val Pro Lys Ser Ala Ala Glu Lys Ala Gly Leu Lys Ala Gly
305 310 315 320
Asp Ile Ile Thr Ala Met Asn Gly Gln Lys Ile Ser Ser Phe Ala Glu
325 330 335
Ile Arg Ala Lys Ile Ala Thr Thr Gly Ala Gly Lys Glu Ile Ser Leu
340 345 350
Thr Tyr Leu Arg Asp Gly Lys Ser His Asp Val Lys Met Lys Leu Gln
355 360 365
Ala Asp Asp Gly Ser Gln Leu Ser Ser Lys Thr Glu Leu Pro Ala Leu
370 375 380
Asp Gly Ala Thr Leu Lys Asp Tyr Asp Ala Lys Gly Val Lys Gly Ile
385 390 395 400
Glu Ile Thr Lys Ile Gln Pro Asn Ser Leu Ala Ala Gln Arg Gly Leu
405 410 415
Lys Ser Gly Asp Ile Ile Ile Gly Ile Asn Arg Gln Met Ile Glu Asn
420 425 430
Ile Arg Glu Leu Asn Lys Val Leu Glu Thr Glu Pro Ser Ala Val Ala
435 440 445
Leu Asn Ile Leu Arg Gly Asp Ser Asn Phe Tyr Leu Leu Val Gln
450 455 460