Note: Descriptions are shown in the official language in which they were submitted.
.. ... . . . _ I . . _ ..,_, a , . . _ . . .,..-.,.
PLASMID FOR PRODUCTION OF MEMBRANE
PROTEIN, BACTERIUM CONTAI NI NG SAME,
MONOCLONAL ANTIBODY THEREFORE, AND
METHOD FOR THE IDENTIFICATION OF
HAEMOPHILUS INFLUENZAE
BACKGROUND OF THE INVENTION
Haemophilus influenzae type b has long been
recognized as a frequent pathogen, particularly in infants
and children, but only recently has nontypable H.
influenzae been recognized as an important pathogen. it
is now well established that nontypable H. influenzae
causes pneumonia, bacteremia, meningitis, postpartum
sepsis, and acute febrile tracheobronchitis in adults. In
addition, nontypable H. influenzae causes neonatal sepsis
and is a frequent etiologic agent in acute otitis media in
infants and children. Therefore, the importance of
discovering a method to assay a clinical sample such as
sputum, cerebral spinal fluid, blood and others for the
presence of H. influenzae is clear.
The observation that nontypable Hs,influenzae causes
serious infections in adults and children has stimulated
interest in study of the pathogenesis and potential
virulence factors associated with this bacterium. The
ribitol capsule of H. influenzae type b is a virulence
factor for the organism, and antibody to capsule protects
the host by means of bactericidal and/or opsonizing
actions. These observations have generated
1
,~... ~
';41 5 7 4
much investigation on the role of the oapaular polysaocharide in
infection with H. influenzae type b and protection from these
infections. However, nontypable H. influenzae lacks a
polysaccharide capsule, and, aimilar to the outer membranes of
other gram-negative bacteria, the outer membrane of H. influenzae
is composed of outer membrane proteins (ONPs) and lipopoly-
saccharide (LPS). Therefore, atudies of the relationship between
virulence of nontypable H. influenzae and surface antigens focus
on OMPs and LPS.
Analysis of OMPs of nontypable H. influenzae has shown that
there are marked differences in OMP composition among strqins.
See e.g. Murphy et al, "A Subtyping System For Nontypable
Haemophilus influenzae Based on Outer-membrane Proteins," J.
Infect. Dis, 1983, ~147:838-46f Barenkamp et al, "Outer Hembrane
.
Protein and Biotype Analysia of Pathogenia NontyQable Haemophilus
influenzae," Infect. Immun, 1982, 36s535-40= Loeb et al, "Outer
Membrane Protein Composition in Disease Isolates of Haemophilus
influenzae, Pathogenio and Epidemiological I,aoplications," Infect.
Immun, 198o, 30:709-17=
A subtyping syatem for nontypable H. influenzae based on the
major OMPs has previously been developed. If a surfaoe exposed
antigen (immunogen) which is conserved in all strains could be
found, it would be an important tool in developing a method of
identifying H. influenzae in clinioal apec rmena aa well as a
vaccine against H. influenzae. It is therefore an objeot of this
2
ci
~3 41514
invention to find a surface exposed antigen in both typable and
nontypable H. influenzae which is conserved in all strains
including typable H. influenzae such as type b which ia known to
cause bacterial meningitis. It ia a further object of this
invention to develop a means for predictably identifying such
conserved surface exposed antigen. It is a further oboeot to
develop a monoclonal antibody againat such a surface exposed
antigen. A further objeot of the invention ia to develop a means
for producing large quantities of such antigen and another object
is to isolate and introduce the genetic sequenoe for such antigen
into a novel plasmid and to cause expression -of such sequenoe in
a bacteria such as E. coli to produce such antigen.
Another object of the invention is to construct a nucleic acid probe
through the combination of the aurfaoe exposed antigen in tloth
typable and nontypable H. influenzae which is oonserved in all
strains and the monoolonal antibody which would be a diagnostic
test for detecting H. influenzae.
BRIEF DESCRIPTION OF THE INYENTION
In accordance with the present invention there ie provided a
plasmid containing a genetic code for an immunolkenio portion of a
nontypable Haemophilus influenzae, whioh immunogenio portion is
conserved in many strains of nontypable E. i=nfluenzae. The
invention further includes a bacterium which contains said
plasmid and will cause expression of said genetio sequence and
3
ci
13 4 157'4
includes a monoclonal antibody to the immunogenic portion and
further includes the hybridoma which will produce aaid monoclonal
antibody.
The immunogenic portion may be and preferably is an epitope
on an outer membrane protein of the H. influenzae and
specifically may be and preferably is a 16,600-daltox outer
membrane protein. The DNA sequence for the gene expreasing this
16,600-dalton outer membrane protein is believed to begin at
nucleotide 125 and continues until nucleotide 526 of the cloned
insert. The immunogenic portion may be produced in its pure state
or as a part of a longer chain protein.
A diagnostic test for detecting I. influenzae in clinical
samples comprises-a nucleic acid probe synthesized to correspond to
the nucleic acids which code for the immunogenic portion conserved
in many strains of nontypable I. influenzae. This probe may be
labelled, for example, with a radioactive or any other suitable
diagnostically recognizable marker.
DETAILED DESCRIPTION OF THE INVENTION
"Nontypable Haemophilus influenzae" as used herein, means H.
influenzae which lacks a polysaocharide capsule and which has an
outer membrance comprising outer membrane proteins (OMPs) and
also comprises lipopolysaccharides (LPS).
"immunogenic portion" means that portion which will result
in an immunological antibody response in a host organism. Such
portion may be considered an antigen.
4
- -~M
13415 74
"Epitope" means that limited immunogenic portion which
results in a specific immunological response.
In the drawings which form a part of this specification
Figure 1 shows a Western blot assay with three lanes from
the same gel;
Figure 2 is a Western blot assay depicting another experi-
ment to assess whether the epitope recognized by antibody 7F3
is on the protein or LPS;
Figure 3 shows whole cell preparations of nontypable H.
influenzae strain 3524 assayed on the same gel and transferred
to nitrocellulose paper;
Figure 4 shows the construction of the Haemophilus
influenzae 1479 genomic library;
Figure 5 shows the methods used in screening the library.
Figure 6 shows the restriction map gene coding for the
16,600 dalton outer membrane protein.
5
~.
13 41574
In accordance with the present invention, a mouse monoolonal
antibody that recognizes an epitope on a 16,600-dalton outer
membrane protein P6 was developed to nontypable Haemophilus
influenzae . This epitope was present on all 115 isolates of H.
influenzae tested, including typable and nontypable strains.
Screening of 89 strains of other bacteria demonstrated that this
epitope is a highly specific marker for H. influenzae because the
epitope was absent in virtually all other baoterial species
tested. Western blot assays were performed with two normal human
serum samples and aonvalescent-phase serum from an adult with
bacteremia due to nontypable H. influenzae. Antibody to the
16,600-dalton outer membrane protein was present in all three
human serum samples.
Prototype strains of nontypable H. influenzae representing
the eight OMP subtypes were obtained from our own collection.
See Murphy et al, supra. Strain 3524 was isolated from the
sputum of a patient with chronic bronchitis at the Erie County
Medical Center (Buffalo, NY). Dr. S. Berk (V.A. Medical Center,
Mountain Home, 76enn) provided 14 strains of nontypable H.
infiuenzae from blood or transtraoheal aspirates. The remaining
strains of nontypable H. influenzae were olinical isolates from
5a
E
13 4~574
the Erie County Medical Center and the Buffalo V.A. Medical
Center.
Dr. J. Ward (University of California at Los Angeles)
provided 54 atrains of H. intluenzae. type b., The remaining
strains of H. influenzae type b were olinioal, isolates from the
Buffalo Children's Hobpital. Reference strains of other capsular
serotypes of H. influenzae were obtained from the Centers for
Disease Control (Atlanta). '
Cultures of Haemophilus paraphrophilua ATCC 29200, Haemophilus
segnis ATCC 10977, HaemoEhilua par ainf:uenzae ATCC 7901 and 9276,
Haemophilus aegyptious ATCC 11116, Hasmophilus parahemolytious
ATCC 10014, nontypable H. influenzae ATCC 19418, Aotinobaoillus
actinomycetemoomitans ATCC 29522, ATCC 29523, ATCC 29524, NCTC
9707, and NCTC 9710, aotinobaoillus e:gul*li ATCC 19392.
Actinobacillus seminia ATCC 15768, and Aotinobaoillua auis ATCC
15557 were provided by Dr. J. Zambon (School of Dentistry, State
University of New York at Buffalo). Iaolates of all other species
were provided by the olinioal microbiology liLboratory at the Erie
County Medical Center.
The identity of strains of H. influenzae was oonfirmed by
colonial morphology and growth requirement for hemin and
nicotinamide adenine dinucleotide. Capsular aerotypes were
determined by CIE with use of reference strains and antiaerum
from the Centers for Disease Control', Hurphy et al, supra.
6
_~ :,
~341574
Strains were stored in Mueller-Hinton broth plus 10% glyoerol at
-70 C.
BALB/c mice were immunized ip with 0.1 ml of 109 oells of
nontypable H. influenzae strain 3524 on days 0 arZd 28. On day 32
after the initial immunization, aeleoted animals were killed with
chloroform, their spleens were removed, and splenio lymphooytes
were harvested by perfusion of splenio pulp with minimal
essential medium.
To achieve hybridoma development by fusion of the donar
spleen cells to the NS 1(nonseoreting varlant of the IgOl BA/c
plasmacytome P3XAg8) plasmacytoma ceila (obtained from the Salk
Institute of Biology [La Jolla, Calif] under National Cancer
Institute contract N01-CB-23886), 35% poly.eti~ylsn~ glycol was
used in a modifi'oa~ion of the prooedure :ot KenneLt, Cell 'Fuslon,
Methods Enzymol, 1979. 58:345-359=, inebriafl F10~ 8p26e6 os11a
were combined with' 106 NS-1 oells in m1a.iotal eesentlal medium
with serum. The ceils were oentrifuge.d dL 170 g.fori 10 min at 25
C. All of the supernatant was removed, and the pellet was tapped
to loosen it. Two-tenths milliliter dl~'j~~-~oD2~!et1l~''ferte;w-.'gtyooi
1,000 (Sigma Chemical Co., St. Louis) in minima2 essential medium
without serum was added and the mixture was atrirred gently and
left at 25 C for 8 min, with the last 3 min consisting of
centrifugation at 500 g to pellet the cells. At the end of the
original 8 min, 5 ml of minimum essential:medium (MEM) with serum
was added and gently pipetted once to res.uspend the pellet. The
.... 7 .
13415 74
mixture was centrifuged at 250 g for 5 min at room temperature
(25 C). All of the supernatant was removed. Five milliliters of
complete minimal essential medium (medium with gluoose I4.5
mg/ml] and 20$ fetal bovine serum) was added eo resuspend the
pellet. The mixture was transferred to a 25-m1 Erlenmyer flask
containing the appropriate amount of complete minimal essential
medium to obtain 3 x 105 plasmacytoma oells/ml. The oells were
stirred gently and distributed in 0.05-ntl samples into miorotiter
wells.
At 24 hr after the polyethylene glycol fusion, 0.05 ml of
medium containing hypoxanthine (13.6 pg/ml), aminopterin 40.36
Ng/ml), and thymidine (3.87pg/ml) was added to each well. The
microtiter plates were placed in a tissue attiture inoubator at
85x humidity in an4atmosphere of 5x C02 and 95x-room air. Freah
medium containing hypoxanthine, aminopter.in,.a=nd thymidine was
added on day 7, and plates w.ere oheokeds~'br mae~~s+~opia ~]saues
.:,sytv,= ,r,= y .,:=. r~ . .
: . . . .
after day 10. The supernatant from all wslls'was tested for the
presence of antibody with an ELISA (ens.'yme linked minimal
absorbant assay).
ELISAs were performed in polyvinyl 96-well miorotiter plates
(Dynatech, Alexandria, VA); 200 -vu1 volumes were used for each
step. Wells were coated with a Qell envelope preparation (10
ug/mi) of nontypable H. influenzae strain 3524 prepared by the
method of Johnston, "Immunobiology of Neiaseriagonorrhooae",
8 -
13 415 74
American Society for Microbiology, 1978, 121-9. Plates were
incubated at 37 C for 1 hr followed by overnight inoubation at 4
C. Wells were washed three times with PBS (phosphate buffered
saline) plus 0.05% Tween 20R surfaotant between eaoh step.
Unbound sites on the plastic were blocked with 3% bovine serum
albumin in PBS for 2 hr..at 37 C. Tissue culture supeNqatabts (or
dilutions of mouse asoites fluid in subsequent expeMiments)
containing monoolonal antibody were inoubated.,in the wells
overnight at 4 C. Rabbit antibody to mouse 1g0.and IgM was then
incubated for 2 hr at 37 C followed by p'rofseih A.-peroxi=dase for 2
hr at 37 C. Two hundred microlitera of aubsbrate was then added
.
to each well. Substrate was prepared by dissolving 10 mg of 0-
phenyl-enediamine in 1 ml of methanol and adding this solution to
99 ml of citrate-phosphate buffer, pH 5.0, plus 0.1 1 of 3%
H202. After the substrate was incubated for 45 min in the dark
at room temperature, the reaction was stopped with 50p1 of 4 N
H2SO4. The 0D490 was measured. Each set of ELISAs was performed
with a control in whioh NS-1 tissue culture supernatant or
ascites fluid was used in plaoe of the monoalonal antibody being
tested. On the basis of the results of ELISA sareening, selected
clones were propagated by subsequent transfer to larger tissue
culture wells. Large quantities of antibody were.produoed in
tissue culture and by ip injeotion of. 105 hybridoma cells into
pristane-primed BALB/c mice. The resulting asoitio -fluid was
harvested in three to four weeks and tested forspeoifioity.
~3 4#514
The strains to be assayed were grown on Ohooolate agar (or
other appropriate medium, depending on the speoies) overnight at
37 C in an atmosphere of 95% room air and 5% CO2. Cells from one
plate were harvested by suspension in PBS and oentrifugation at
10,000 g for 20 min. The resulting pellet was suspended in
enough PBS to allow the suspension to be drawn rinto a
micropipette. One-tenth milliliter of the suspension of bacteria
was added to 0.4 ml of sample buffer (-0.06 M Tris, 1.2% SDS, 1%
B-mercaptoethanol, and 11.9% glycerol) and heated for 5 min in a
boiling water bdth. The resulting organisms are referred to as
whole cell preparation. A 10 -
.,pl drop of whole cell preparation was placed on a
nitrocellulose sheet (Sohleic3her and Schve2l:, Ino:o Keeae, NH)
and allowed to air-dry. The sheet waa -then pliaed in 3x :ge2dtin
in buffer A (0.012 M Tris and 0.15 M< NaCl, pH 17=4) for 1 hr.
After the sheet was rinsed with buffer A, it was plaoed in an
appropriate dilution of antibody and allowed to shake at room
temperature overnight. The sheet was >~i;ed:.4-3Lh.bafter. A and
placed in 1:3,000 dilution of protein A peroxidase' (Zymed
Laboratories, San Francisco) and ahaken for 1 hr at room
temperature. The sheet was rinsed and immersed in horseradiah
peroxidase color development aolution (0.015% H20zi Bio-Rad,
Richmond, Calif) for 45 min. Controls assayed on eaeh sheet
included sample buffer (negative control). A negative result was
_
13 41574
recorded when the dot was no different from the background oolor,
and a positive result was recorded when the dot turned purple-
blue. About 90% of dot assays were unequivooally positive or
negative. Those strains that yielded equivooal results in the
dot assay were subjected to Western blot assay,
Preparation=of LPS. Lipopolysacoharlde (LPS) .vas prepared
from nontypable H. influenzae strain 3524 by,two msthodls. The
first method was a modification of the phenol-water extraction
method of Westphal and Jaan, "Baoteri~~";;~''
. =r~'. - . .
Methods in Carbohydrate Chemistry, 1965, 5036091. The second
method was that of Hitchcock and Brown, Journal of Haateriology,
1983, 154:269-77. The latter method uses the enzyme proteinase K
(Boehringer Mannheim t3mbH, Mannheim, Federal Republic of
Germany), which hydrolyzes proteins but has no effect on LPS.
Whole cell and LPS preparations were subjected to SDS-PAGE
(sodium dodecyl sulfate polyacryZamide ge1 eleotrophoresis) with
either 11% .or 13.2% separating .gels, Murphy et,al, supra. When
electrophoresis was oompleted, the g.l was placed with a
nitrocellulose sheet that had been previo-usly-boiled:i-n distilled
water, and the sheet was immersed in Q.3 H aodi.um citrate plus 3
M NaCl. Eleotrophoretic transfer was o-arrfed out in a Trans-
B1otR oell (Bio-Rad) at 50 V for 90 min. The eleotrode buffer
was 0.025 M Tris, pH 8.3, 0.192 M glycine, and 20% methanol. The
nitrocellulose sheet was then treated exactly as described for
the dot assay; it was blocked with 3% gelatin and inoubated
. 11
;3 415 74
sequentially with antibody 7F3, protein A-peroxidase, and
substrate horseradish peroxidase color developer.
I radiolabeling of surface OMPs. Extrinsic labeling of
- - --- ------- surface-exposed OMPs was accomplished with a laotoperoxidase-
catalyzed radioiodination prooedure, Hansen et al, Infect. Immun.
1981, 32:1084-92.
The ELISA with outer membranes of nontypable H. inrluenzae
strain 3524 coated on micr.otiter plates deadoaatr=atbd that the
hybridoma designated 7F3 was produoing antibody 7F3 that
recognized a determinant in the'outer membrane ~of the bacterium.
Gel immunodiffusion indicated that this antibody was of the.Ig(33
isotype. Figure 1 shows a Western blot that indioates that the
determinant recognized by antibody 7F3 was on a pro-tein with a
molecular size of '16,600 daltons; lane A showa molecular weight
standards on the nitrocellulose sheet, and lane B shows the
16,600-dalton protein recognized by antibody 7F3 in a whole cell
preparation of nontypable H. influenzae strain 3524.
Specifically, lane A shows molecular weight-standards transferred
from a 13.2% gel; lane B shows a whole oell preparation of
nontypable H. influenzae strain 3524 incubated with ahtibody 7F3,
protein A-peroxidase, and peroxide subatratei and lane C is an
autoradiograph of a whole cell preparation of nontypable H.
influenzae strain 3524 made from bacteria extrinsiaal2y labeled
w i th 1251. All three lanes were from th ' . a$me ae1. Weatern blot
12
13 41574
assay done by this method in 25 strains of H. influenzae showed
that antibody 7F3 recognized a determinant on this 16,600-dalton
protein in every strain. Because the antibody recognized a
determinant on a protein of identieal molecular size in multiple
strains, we screened larger numbers of straina with use of a dot
assay rather than Weatet~n blot.
To determine whether the protein recognized by anti0ody 7F3
could be extrinsically labeled, we labeled nontypable H.
influenzae strain 3524 with 1251. The proteins were subjected to
SDS-PAGE and transferred to a nitrocellulose sheet. One lane was
exposed to x-ray film, and one lane was inoubated with,7F3,
protein A-peroxidase eonjugate, and substrate. Fl.gure 1 shows
that the band recognized by antibody 7F3 (lane B) oorresponds to
an 125I-labeled baAd (lane C).
To assess further whether the epitope reQognized by antibody
7F3 was on a protein or on LPS, we performed two additional
experiments. An ELISA was performed as described above in which
some welis were coated with a cell envelope preparation of
nontypable H. influenzae atrain 3524 and other .wells were ooated
with LPS prepared from nontypable H. infZuenzae strain 3524 by
the phenol-water method Westphal et a1,-,;.:su1pra. ; Antibody', 7F3 was
reactive 'with a cell envelope preparation (OD, 0.375) that
contained OMPs and LPS, Johnston et al, suprao. but was
nonreactive with LPS (OD, 0.062). This finding indioates that
the epitope recognized by antibody 7F3 resides on an OMP.
13
41574
Figure 2 is a Western blot assay depicting another
experiment designed to assess whether the epitope recognized by
antibody 7F3 is on a protein or LPS. The lanes marked A contain
LPS prepared by proteinase K lysis of oells of strain 3524,
Hitchcock et al, J. Bacteriol, 1983, 154:269-7T, the lanes marked
D contains LPS of .1train 3524 prepared by the phenol-water
t
method, Westphal et al, supra, and the lanes marked C oontain a
whole cell preparation of strain 3524. All aampled were assayed
on the same gel and tranaferred to the same nitrooellulose sheet.
Figure 2, left, was inoubated with antibody 7Z3 (ascites fluid
dilution, 1:500), and figure 2, right,was Inoubated -iith
antibody 3D2 (asoites fluid dilution,, 1:500), a monoolonal
antibody that recognizes the lipid A portion of H. influenzae
LPS. Antibody 7F3 does not bind to either of the LPS
preparations and binds only to a band 'with a molecular weight of
16,600 in the whole cell preparation. This observation
demonstrates that antibody 7F3 recognizes an epitope on a protein
and not on LPS. Specifically, Figure 2 shows a Western blot assay from a
13.2% gel: (left) inoubation with antibody 7F3 and (right)
incubation with antibody 3D2, which reoognizes an epitope on the
lipid A of H. influenzae. The lanes marked A oontain LPS of.
nontypable H. influenzae strain 3524 prepared by lyais of oella
with proteinase K, the lanes marked B bontain phenol-water
14
13 41574
prepared LPS of strain 3524, and the lanes marked C contain a
whole cell preparation of st'rain 3524. Noleoular weight
standards are noted on the right.
Studies were performed to determine the specieaspeaifioity
of the antigen recognized by antfk:Cidy ::7.F3. Yhole : aell
preparations of 115 isolates of H. ih:lluenzae' were e=tudied by
either dot assay or Western blot assay.: =included 73
.<.., .
type b, 37 nontypable; and 1 each of Lyp:es.;a and.;a-f. All 115
wo,.
strains of H. influenzae oontdined th4=,-.epi''Lo0e recognieed 'by
antibody 7F3, a result indicating that this epitope ia a common
antigen among strains of H. intluenzae.
Sixty isolates of various baoterial apeoies were studied to
determine whether this epitope is present in bacteria other than
H. influenzae. 'AlA 60 of these atraine laaked the detersina-nt
recognized by antibody 7F3 (table 1)..
~3 41574
Table 1
SPECIFICITY OF ANTIBODY 7F3 FOR VARIOUS BACTERIAL SPECIES
Bacterium No. tested No. positive
Gram-negative
Esaherichia coli 10 to
Actinobaeillus speoies 10 0
Proteus species T 0
Pseudomonas species 5 0
Klebsiella speaies 4 0
Serratia speaies 4 0.Enterobacter oloaaae 1 0
Morganella "morganii 1 0
Neisseria gonorrhoeae 6 0
Neisseria species 2 0
Gram-positive
Staphylococcus aureus 5 0
Staphylococcus species 2 0
Viridans streptoooooi 1 0
Streptococous faecalis 1 0
Diphtheroids 1 0
Total 60- 0
16
;3-41514
Twenty-nine strains of Haemophilus species other than H.
influenzae were studied. Twenty-five ot these isolates lacked
the 7F3 epitope (table 2). Two strains of ~ parahemo2;'ticus
contained the determinant. In addition, one strain of H.
paraphrophilus and one of H. aegyptious contained a 20,000-dalton
protein that was recognized by antibody 7F3.
Table 2
- .'
SPECIFICITY OF ANTIBODY 7F3 FOR VARIOUS SPECIES OF HAEMOPHILUS
Species No. tested No, positive
~
H. parainfluenzae 24 0
H. parahemolytiaus 2
H. paraphrophilus 1e
H. s-egnis 0
H. aegypticus 1 1*
Total 21
* In the Western blot assay, antibody 7F3 reoognized a 20,000-
dalton protein in these strains.
17
;3 41574
Human serum antibody. Human serum was tested for the
presence of antibody to the 16,600-dalton OMP by Western blot
assay. Figure 3 shows whole cell preparations of nontypable H.
influenzae strain 3524 that were assayed on the same gel and
transferred to nitrooellulose paperf lane A was inoubated with
7F3 ascitea fluid and shows a single band oorresponding to the
16,600-dalton protein, lanes B and C were inoubated with two
different samples of normal human serum,.and-lane D was incubated
with serum obtained from an.. adult 17 days;~:a~ter;~~aoter,emia;.due- to
nontypable H. influenzae. All three samples af human serum Jhave
antibody to the 16,600-dalton OMP that contains the determinant
recognized by antibody 7F3. The DNA sequenoe for the gene
expressing this 16,600-dalton outer membrane protein is believed
to begin at nucleotide 125 and eontinues until nuoleotide 526.
This amino acid sequence is inoluded as a portion of the insert.
The restriction map of this portion of the sequence is shown
in Figure 6.
The gene is believed to have the following sequence:
18
;3 41574
Amino Acid
DNA Sequence
CCCAAGTAAAATTTnCAGCT[CAGTCrCATA.,~TPAACTAAATAAAAAAC'lC1TTCAOGAGAAATCTA
1 ssssa
c~t .~sn lys p!~ val lys ser leu leu val ala gly ser val ala ala leu
11TG AAC AN1 TTr GPT 11AA TCA TTA 'rTA GTT (>CA GGT 'rCT (,'TA CL~'T lxA TTA
68
= tve
ala ata c-ys ser-ser sQr aan asn asp ala.a2a,gly asn gly sla la gin
;'lt; c3Cr ~rC'r ACr TOC TCT MC AAC GAT w1' GCA Q= AA? a'.'r (X.T OCT rJa
thr ;ihe gly gly tyr ser val ala asp leu'gln-gln arg tyr asn thrival
.~r rrr L= GGA TAC TC'r GTT GCT GAT CCrr CAA G1A CGT TAe AAC ACC CTA
200
tyr plye gly ,?he asp lys tyr asp i,le thr.gly qltu'ty,r val qln ile leu
T,-%T 'P'CT caGT 'PrT GAT AAA TAC GAC A'!C ACC GCP CAA TAC CTT CA11 ATC '!'rA
asp ala his ala, ala tyr leu asn ala thr pro ala ala lys'-val leu,val
GAT C;:G CAC CiCA C~A TAT TTA AAT Cr-A AQG: a:A t3;."1' ClC'r AAA CrA TTA GTA
3e0;
glu gly asn thr asp glu arg gly thr prd"sflu ty;-asn..ile ala leu gly
:AN OGr AAT ACT cAT GU OGT GCT ACh C'J1 GA4 'rAC : uC A7C GrA K"i'A (~GA
gln arg arg alfa asp ala val lys gly tyr leci ala gly lys 91y val asp
CAA OGT aGT C1CA GAT CrA CTT AAA GGTT TAT TTA MA CaGT AAA OCT CTT d-T
. 400
ala gly lys leu gly thr val ser tyr gly glu glu lys pro ala val leu
Wr (2;T AAA TTA = ACA GTA TCr TAC Wr GAA G'AA AAA a.'t' C1CA GTA TTA
gly his asp glu ala ala tyr ser lys asn arg arg ala val leu ala tyr
Cr;T CAC GAT GAA C1G"r WA TAT ZCT AAA AAC CGT dGT (rA GTG TTA C1CG TAC
56A
-termination TAA TZ~'t'T~GTATP1ti.'TAAT TCX'ATZT'PTTA'tTOC,A CITI-sermc
ArCG'ITPfa'rAATTTAACCAATTAC~CT'K'AAAGAATGAATTTA7'l1CT'TT~ATTCTAAI~ATAAA'IC~G'G
60A
'rTATCATTAACTCA'!'GACA+CAC'r00GTG67rT AGA AA'iC0C'T
700
TOGTCGAAwT"POGAA TTT~ ~'~
GCtrPOCr AAT'lCX+I'0(~C'i'CGpiGAGTI'OC'~0 CrA0Ca1T'PCT
800
TACC.'G
19
IE
341574
It is worthy of note that this band is among the most prominent
recognized by antibody in human serum.
Specifically, Figure 3 shows a Western blot assay from a
13.2% gel. All four lanes contain a whole cell preparation of
nontypable H. influenzae strain 3524 fromthe same gel, but each
lane was incubated wi'th a different antiserum: lane A, antibody
7F3; lanes B and C, two different samples of normal human serum
(dilution, 1:500); and lane D, serum obtained 17 days after
bacteremia due to nontypable H. influenzae in an adult (dilution,
1:500). The incubation with antiserum was followed by inoubation
with protein A-peroxidase and peroxide substrate. The arrows
indicate that all three samples of human seruo oontarln sntibody
to the 16,600-dalton OMP that oont-a-tns the fiF3 epi'tope.
molecular weight standards are indioated;~on theJeft.
In accordance with the invention; an 1903 mouse monoalonal
r:>
antibody that rec.ognizes an epitope on a,16,600=dal;ton-:-OMP .on~ the
..., t. ,. .
surface of nontypable H. influenzae has been developed. This
epitope is present in all 115 isolates of ai influenzae tested,
including typable and nontypable strains. Screening of 60
strains of non-Haemophilus species demonstrated that the epitope
is absent in all of'these bacteria. The epitope was absent in 24
strains of H. parainfluenzae but was:preaent in four to five
strains of other Haemo hp ilus speoies (table 2). These species
are unusual pathogens in humans. Therefore, from thb standpoint
3 41 5 74
of clinically relevant isolates, antibody 7F3 is highly specific
for H. influenzae.
This morioclonal antibody recognizing a common epitope that
is highly specific for H. influenzae oan be useful aa a tool in
the clinical microbiology laboratory. A rapid teat to confirm
the identity of a cli'nical isolate as H. ee ;; .~i.n-t$uepgae<~Ctypable .or
--~ . .r.~.,....,. . f
nontypable) could be developed based on auoh a.n antibody. In
order to construct a DNA probe to exploit this speoifio epitope,
the DNA sequence of the gene encoding P6 is determined. Based on
the DNA sequence, the amino acid sequence of the active P6
protein can be deduced. This information can be used to peeform
what is known as epitope mapping.
Epitope mapping involves the construotion of a number of
small peptides and testing these peptides for reactivity wAth
monoclonal antibody 7F3. Since the eptitope reoognized by 7F3 is
specific for H. influenzae, the corresponding peptide recognized
by that antibody represents the speoific determinant on H.
influenzae. Once the amino acid sequenoir of the peptide is
known, the DNA sequence of that segment'aan be deduoed. Since H.
influenzae contains the gene which eodea for this epitope, the
bacterium is known to contain DNA which has a sequence
corresponding to this sequence. A DNA probe oan, therefore, be
constructed to correspond to the nucleic aeids which oode for the
specific epitope on'P6. Once the probe is eonstrueted, it can be
labelled, for example, with a radioactive member. This probe
21
1341574
could then be used to assay a clinioal.sample suoh as aputum,
cerebral spinal fluid, blood and others for the presenoe of H.
influenzae. This will be possible beoause the DNA probe,will
bind to its complementary base pairs which are present in the
genome of H. influenzae. Once this probe is oonstruoted, this
approach would represent an advantage over the aurrent,twidely
used method of demonstrating growth requirements for hemin and
nicotinamide adenine dinuoleotide. An assay with a apeoific
monoclonal antibody would yield resulta 24 hrs sooner.
OMPs and LPS are olosely associated on the outer membranes
.
of gram-negative bacteria. This fact and the observation that
the determinant recognized by antibody 7F3 is in the molecular
weight range wherp LPS separates lead one to question whether
this determinant is on a protein or.on LPS. Several lines of
evidence indicate that the epitope recognized by antibody 7F3 is
on a protein. First, staining with Coomaasie blue of SDS gels
demonstrated the presence of a band reoognizad tiy antibody 7F3 at
16,600 daltons in all strains of H. influenzae. Because
Coomassie blue stains protein but not LPS, this observation is
presumptive evidence that antibody 7F3 reoognizes a protein
determinant. Second, the configuration of the band on SDS-PAGE
and Western blot was typioal for proteini.LPS showed multiple
bands that were generally less distinot than the band at whioh
the antibody 7F3 epitope resides. 'This point is further
22
341574
corroborated by the observation that monoolonal antibodies that
recognize LPS determinants showed the typical "LPS" pattern in
Western blot assays of whole cell preparations, in contrast to
the well-defined single band recognized by antibody M. Third,
by ELISA, antibody 7F3 showed reaotivity with oell envelope
preparations that oon.tain OMPs plus LPS, but the antibody showed
no reactivity with isolated LPS. Finally, in the Weste6 blot
assay (figure 2), antibody 7F3 recognized a band in a whole cell
preparation but failed to reoognize determinants on LPS that was
prepared by using two different methods. Taken together, these
observations indicate that the epitope reoognized by antibody=7F3
resides on an OMP.
To assess whether the OMP oontaining the.antibody 7F3
epitope was surfacoe exposed, OMPs were labeled by using a
lactoperoxidase-catalyzed radioiodination prooedure, Hansen et
al, supra. Figure 1 shows that the protein containing the
antibody 7F3 epitope is radiolabeled. Thia observation suggests
that this 16,600-dalton OMP is surface expos.ed. For the purposes
herein "surface exposed" or "outer membrane" means available for
antibody binding.
The OMPs- of nontypable H. influenzae show substantial
strain-to-strain variability, as demonstrated by SDS-PAaE
analysis. This variability in the major OMPs in the 32,000-
42,000-dalton range is the basis of the subtyping system for
nontypable H. influenzae, Murphy et al, supra. It is of interest
23
13 4 15 74
that studies of OMPs of H. influenzae from three laboratories
have independently noted the presence of a"16,000"-dalton OMP in
all strains of H. influenzae studieds Murphy et .al, Barenkamp et
al and Loeb et al, supra. It is this protein that contains the
antigenic determinant recognized by antibody TF3. The present
study indicates that the epitope reaog.bized by,antibo.dyOp'3 on
this low-molecular-weight OMP is an antigen common to all strains
of H. influenzae. Identifying common surface antigens among
strains is useful from the point of view of vaccine development
because immunization with a single common antigen might induce
protection from diaease due to many strains. In addition, the
observation that this 16,600-dalton protein has varied far less
than other OMP-s in the oourse of evo=lution leads to the
speculation that this protein serves an important tcihotion 7or
the bacterium and that its function is closely related to
conservation of its structure.
The outer membranes of gram-negative bacteria are
im munologically important structures -'beoause of their
accessibility to host defense meohaniams. Indeed, antibody to
OMPs of H. influenzae type b are widely prevalent in adults and
are detected in the serum of infants who are oonvalesoing from
infections with H. influenzae. It has now been demonstrated that
antibody to a 16,600-dalton OMP (p6) is present in human serum
(figure 3). The presence of antibody to this OMP in normal human
2u
1341574
serum suggests that the OMP is important with regard to the human
antibody response to H. influenzae.
Several observations suggest that p6 is an important target
in immunity to Haemophilus influenzaes
1) Antibody raised from p6 isolated from a type b strain
protects in an infant rat model.
2) A monoolonal antibody, 7F3, directed against p6 blocks
human bactericidal aotivity against Haemophilus influenzae
(NtHi).
3) Depleting normal human eera of p6 by affinity
chromatography resulted in reduced bacterioidal activity of~that
sera for Haemophilus influenzae.
u) And, i~mmunopurified antibody to-p6 from human sera was
~
bactericidal.
This invention therefore includes the molecular cloning of
p6 using H. influenzae as a source of bacterial chromosomal DNA,
lambda gtll bacteriophage as the vector in oonstruotion of the
genoinic library, pUC18 plasmid as the veotor u3ed in subaloning
the gene to facilitate sequencing, and E. coli as the host
strains. The results allow further analysis of the moieoular
basis of both experimental and human immunity to p6 and permits
large quantities of p6 to be produoed once it is approved for use
in vaccine against Haemophilus influenzae.
Molecular Cloning of p6:
41574
The 16,600 dalton protein, designated herein as p6, is
therefore present in the outer membranes ot both typable and
nontypable straina of Haemophilus influenzae and may be an
important target in im munity to Haemophilus influenzae. The DNA
sequence for the gene expressing this 16,600-dalton outer
membrane protein is'believed to begin at nuoleotide 125 and
continues until nucleotide 526. In acoordance with this
t
invention p6 is cloned molecularly using a nontypable strain of
Haemophilus intluenzae as a source of bacterial chromosomal DNA,
lambda gtll bacteriophage as the vector in construction of the
genomic library, pUC 18 plasmid as the vector in auboloning the
gene to facilitate sequenoing, and E. coli as the host atrain.
The monoclonal ;Lntibodies previously disoussed and a polyolonal
antiserum were used to soreen for expression of p6. A portion-of
the genomic library was soreened re3iu2ting #n the detection of
four positive reoombinants. One, olone e, appears to produoe a
full length geneproduct expressed in high frequenay. The DNA
insert of this olone was used to subaloc-e the gene into a plasmid
vector. An E. coli transformant, 7-9B, also appears to express a
full length gene product. It is likely that transcription is
initiated from the actual promoter of the p6 gene, since both
clone 0 and transformant 7-9B express the gene product in both
the uninduced and induced states. Isolating and sequencing the
gene for p6 allows for further analysis of the molecular basis of
both experimental and human immunity to p6.
26
Z3 415 74
More specifically, recombinent DNA teohology was used to
clone the gene for the 16,600 dalton surface protein, p6, of
nontypable Haemophilus influenzae (NtHi) into Esaheriahia coli.
Chromosomal DNA from a clinical isolate was shearedp ligated to
lambda gtll arms and paokaged into phage heads. Four reoombinant
phages were deteoted by screening with monoolonal antiboAies and
a polyclonal antiserum. One, clone 0, was restricted with EooRt
and ligated to plasmid vector pUC18 to faoi'litate sequencing.
coli carrying reoombinent.plasmf,ds were'sareen'ed resulting in-one
positive, 7-9B. Both clone 0 and 7-9B produce a protein with an
apparent moleoular weight equal to or similar, to native-p6 as
determined by Western blot analyses. In soreening it was
determined thatTtransoription and translat'ion of the Haemophilus
lnfluenzae p6 gene(s) were not dependent on the lao operator 'and
promoter of either vector. Using immunofluoreaoenoe, the
recombinant gene produot's p6 epitopes could be localized on the
surface of these E. coli and accessible to antiVody.
Haemophilus influenzae strain 1479 was grown at 37 degrees C
in brain heart infusion broth supplemented with hem (10yg/ml) and
nicotinamide adenine dinucleotide (1 nag/ml).
The E. coli strain y1090 (r-m+) was used for the lytic
growth of bacteriophage lambda gtll and strain JM83 as the host
for the plamid pUC18. The E. coli strains were grown in L-broth
(LB) or on LB agar with or without 50pg/ml of ampicillin,
27
13 41574
depending on the host strain. A more detailed deaoripti:oh for
the use of the respective host strains oan be found elsewhere.
Young et al, Science 222: 778-782; Mersing Ree. DNA Tech. Bull.
2:43-48.
A pellet of Haemophilus influenzae 1479 cells from a 750m1
culture was resuspended in 10mis of 10mM HEPES butfer, pH 7.4.
To this mixture was added EDTA to a concentration of 5m M6d SDS
to a concentration of 0.5% w/v and then incubated =at 60 degrees
C for 30 minutes. This lysate was then digested with 0.5m1 of
pronase (10mg/ml) at 37 degrees C for 2 hours and then subjected
to two phenol/CIAA extraotions followed by one CIAA extrac~ion.
Sodium chloride was added to a concentration of 0.2M to the
aqueous phase, and DNA was precipitated with 2.5 volumes of
chilled ethanol. Following precipitation in the oold, the DNA
was pelleted by centrifugation, resuspended in Tris-EDTA buffer,
and treated with DNase-free RNase at a concentration of 0.1 mg/mi
at 37 degrees C for 1 hour. Finally, the DNA was extracted with
phenol/CIAA, precipitated with sodium chloride and ethanol, and
pelleted by centrifugation. The DNA was resuspended in Tris-EDTA
buffer, measured for concentration by A260/A280 and =stored at 4
degrees C.
The phage library was soreened with monoolonal antibody 7F3.
Also used in screening was rabbit polyolonal antiserum produced
by immunizations with solubiliz4d p6 preparations of Haemophilus
influenzae strain 1808.
28
~341574
Construction of the Haemophilus influenza 1479 genomio library:
The strategies for the construction of the library, depioted
in Figure 4, were essentially those described by Young and Davis,
1985, Vol. 7, pp 29-41, Genetic Engineering, Plenum Press, N.Y.
Haemophilus influenzae 1479 DNA was sheared by sonioation with
one 10 second burst (output control setting at 2) to an average
length of 2-4 kilobase pairs (kb). The degree of shear was
monitored by agarose gel electrophoresis. The Eco R1 sites of
50ug of this sheared DNA were methylated using Eco R methylase.
=F
The ends of the methylated DNA were made fluah by the additidM of
Klenow polymerase and deoxynualeotide triphosphates. Following
this reaction and the addition of aodium aoetate to a
concentration of' 0.3M, the DNA 'was preoipitated. After
centrifugation, the pellet was resuspended in Tria-EDTA buffer.
The DNA was then ligated to Eoo R1 linkers (Bethesda Research
Laboratories, Bethesda, Maryland) that had been phosphorylated.
The ligation reaction was terminated by heaLting the mixture to 70
degrees C and then the excess Eoo R1 linkera were digested using
an excess of Eco 91. The methylated Haemophilus ieifluenza 1479
DNA blunt end-ligated to Eco R1 linkers was purified from excess
linkers by passage over a gel filtration polumn (Biogel* p60, BIO
BAD laboratories, Richmond, CA) using a column buffer containing
10mM Tris pH 7.5, 100mM NaCl, 1mM EDTA. Fractions were
monitored by A280 and agarose gel eleotrophoresis. Fractions
29
* Trade mark
~3 41574
containing DNA of desired size range were pooled, and
precipitated. The DNA was pelleted by centrifugation and
resuspended in 4u1 of Tris-EDTA buffer. The DNA was ligated to
3ug of dephosphorylated lambda gtll arms (STRATAGENE*Cloning
Systems, San Dieto, CA) in a total reaction volume of lOul. The
ligation mixture was packaged using two packaging extracts
according to the directions of the manufacturer (GigapackTM~
STRATAGENE Cloning Systema). P-aokaged phage wbre plated on E.
coli strain y1090 to determine the titer of plaque forming units
and to determine the non-recombinant baokground by growth with
IPTG and X-gal on LB + AMP plates. The library oontains
approximately 1*.5X106 independent reoombinant olones with a
background of less than or equal to 5%.
Screening the library:
Figure 5 depicts the methods used in screening. A portion
of the y1090 plating stock, 0.2m1 of a y1090 pellet resuspended
in 10 m M MgS04, was infected with 1.5X103 pfus of the lambda gtll
library for each 85mm plate. d Following the adsorption
incubation, the cells were mixed with LB-agarose butter, poured
and spread evenly onto an LB + AMP plate. The plates were then
incubated at 42 degrees C for 3 hours. Each plate was then
overlaid with a dry nitrooellulose filter disk which had been
saturated previously in 10m M IPTO. The plates were then
incubated for 3 hours at 37 degrees C. Before removing the
* Trade mark
341514
filters, the orientation was marked and the filters and
respective plates were labelled. The filters were rinsed briefly
with buffer A(0.01 M Tris, 0.15M NaCl, pH 7.4) and placed in 3%
gelatin in buffer A for 1 hour. After the filters were rinsed
again with buffer A, they were incubated in a screening mixture
of antibodies overnight at room temperature. The screening
mixture was buffer A containing 7F3 ascites fluid, at titers of
1:1000. The monoclonal antibody used shared no crossreactivity
with the E. coli host strains while anti-1808 antiserum required
a working dilution of 1:10,000 to maintain sensitivity and
.
specificity. The filters were rinsed with buffer A and placed in
a 1:3000 dilution of protein A-peroxidase oonjugate and shaken
for 1 hour at room, temperature. The filters were again rinsed
with buffer A, then immersed in horoeradiah peroxidase color
development solution (0.15% H202; BIO RAD, Richmond, CA) for 45
minutes. Plaques that appeared positive were removed from their
respective plates, resuspended in 500).113, of SM buffer, and
rescreened. Plaques that were positive in the rescreening were
then rescreened again but against the individual=antibodies
rather than the screening mixture.
Western blot analysis:
The Haemophilus influenza control, y1090 recombinants, and
the molecular weight standards were prepared by heating at 100
degrees C for 5 minutes in a sample bufter containing 0.06M Tris
31
1341574
pH 6.8, 1.2% SDS, 5% beta-mercaptoethanol, 11.9% glycerol and
0.003% bromophenol blue. The preparations were subjeoted to SDS-
PAGE on a 15% separating gel. Gels were placed on a
nitrocellulose sh,eet which had been previously boiled in
distilled water and immersed in a 0.3M sodium citrate, 3M NaCl
solution. Electrophoretie transfer was done using a TranaphorR
electrophoresis unit (Hoefer Scientific Instruments, San
Francisco, CA) at 50 volts -for 90 minutes, in a buffer of 0.025M
Tris, pH 8.3, 0.192M glycine and 20% methanol. The blooking,
with subsequent additions of antibody, conjugate, and substrate
development was performed in the same manner as described ir! the
plaque screening.
Subcloning into a l~ asmid vector:
A strategy was devised to facilitate sequencing. The DNA
insert (which consists of 867 residues), of a reoombinant phage
expressing p6 epitopes (as determined by soreening), was
subcloned into a plasmid vector. pUC18 was chosen as the plasmid
vec'tor for suboloning for several reasons inoluding a means of
selection, an inducible promotor, and an Eoo RI restriotion aite,
features shared by phage cloning vector lambda gtll. The DNA of
a recombinant phage was restricted with Eoo R1 and then ligated
to pUC18 whioh had been restricted with Eco R1 and
dephosphorylated using calf intestinal alkaline phosphatase. The
ligation mixture was used to transform competent E. eoli atrain
32
3 415 74
JM 83. Transformants were selected for by growth on LB + AMP
plates overlaid with IPTG and X-gal. White colonies, thought to
represent JM 83 containing a plasmid plus insert, were
individually picked and transferred to wells of microtiter plates
containing L-broth + AMP + 10% glycerol. The plates were
incubated overnight at 37 degrees C. A comb device was used to
innoculate from the mierotiter plates onto nitrooellulose sheets,
previously immersed in IPTO, overlaying LB + AMP' plates. The
plates were incubated overnight at 37 degrees C, and then the
nitrocellulose sheets were removed. The nitrocellulose was'hung
for 15 minutes in a chamber containing chloroform vapors to =lyse
the colonies, blocked in 3% gelatin containing 40pg/ml lysozyme
and screened in the same way as the genomie library.
0
Results:
Screening the genomic library and characterization of
recombinants.
Approximately 45,000 plaques were screened, the remainder of
the iunamplified library being frozen in aliquots at -70 degrees C
in 7% DMSO. Four reactive clones, designated as lambda gtll-
Haemophilus influenzae 1479 olones 0, P, 8, and 10, were found.
Efforts focused on clones 0 and P since they appear to express
gene products that are in larger quantities than 8 or 10 and/or
more closely resemble the conformation of the native protein, p6.
Plates containing clones 0 and P were oarefully scraped to
- 3 3
;3 4 1574
harvest protein for Western blot analysis. Western blots show
that both clones 0 and P produce a protein that is the aame or
similar in size to native p6. However, clone 0 produces the
protein in larger quantities when compared to clone P.
Therefore, clone 0 was selected as the recombinant phage for the
source of DNA to sUbalone into aplasmid vector..
Subcloning into a plasmid vector and characterization of the
transformant.
One thousand tranaformants, from the same transformation,
were screened before one, 7-9B, was found to be positive. To
make certain that 7-9B was in pure oulture, positi,ve colonies
were pi.cked, passaged and resoreened. Western.blota.ot 7-9B
grown on platea and in broth show that this reoombinant also
produces a gene product that is of an apparent moleoular weight
equal or si m i lar to native p6. The reooIDbinant pl.asm 3d isolated
from 7-9B, was restricted using Eco R1 and subjeat to agarose gel
electrophoresis to determine the insert size. Restriction
analys is reveals the presence of a 2.5 kb DNA insert
containing the p6 gene.
Discussion
Using molecular cloning techniques, we produced both
recombinant phages and reoombinant plasmid.containing the gene
encoding for Haemophilus influenzae 1479 p6, a 16.6K dalton outer
membrane protein. The protein can be described by the nucleic
34
13 415 74
acid sequence between nucleie acid 125 and nuoleie acid 526,
inclusively. Expression of the gene product by phage alone 0 and
by recombinant plasmid 7-9B is independent of induotion of the B-
galactasidase promoter. This finding, coupled with Western blot
analyses revealing apparently no difference in eleetrophoretic
mobilities between the gene products of olbne 0 and reeombinant
7-9B, provide evidence for the likelihood that the reooJbinant
gene product is initiated from its own, the p6 gene's,
constitutive promoter.
The epitopes of the 16.6K dalton protein are aooessible to
monoclonal antibody 7F3 on the surface of E. ooli transformaqt 7-
9B.
In summary, the gene which codes for p6 was cloned into E.
coli using a phag'e vector and plasmid veotor. The E. ao1i
recombinants express the protein on-the surfaae in a form which
is fully immunogenic.
Except as otherwise indicated all miorobiologioal strai'ns
are generally available. All such strains as described herein
are available from Dr. Timothy Murphy, Division of Infectious
Diseases, State University of New York Cllnioal Center, 462
Grider Street, Buffalo, NY 14215. H. influenzae strains 3524 and
1479; E. coli strain JM83 containing plasmid 7-9B and hybridoma
7F3 are being deposited with the American Type Culture Collection
(ATCC) at 12301 Parklawn Drive, Rockville, Maryland 20852 in
accordance with"the Budapest Treaty.
1341574
The registration date of the deposits with ATCC are
listed below:
ATCC Designation Deposit
Numbers Date
Nontypable H. influenzae, 3524 53599 3-18-87
Nontypable H. irifluenzae, 1479 53600 3-18-87
Escherichia coli, JM83 containing
plasmid 7-9B 67358 3-18-87
Hybridoma cell line, 7F3 HB9625 1-14-88
36
