U.S. patent application number 10/199290 was filed with the patent office on 2003-02-06 for structure of adenovirus bound to cellular receptor car.
This patent application is currently assigned to Brookhaven Science Associates. Invention is credited to Freimuth, Paul I..
Application Number | 20030027338 10/199290 |
Document ID | / |
Family ID | 22889439 |
Filed Date | 2003-02-06 |
United States Patent
Application |
20030027338 |
Kind Code |
A1 |
Freimuth, Paul I. |
February 6, 2003 |
Structure of adenovirus bound to cellular receptor car
Abstract
The present invention is based on the solving of the crystal
structure of adenovirus fiber protein knob domain bound to domain 1
of the coxsackie-adenovirus receptor. One aspect of the present
invention relates to a mutant adenovirus which has a genome
comprising one or more mutations in sequences which encode the
fiber protein knob domain, the viral particle encoded by the genome
being characterized by a significantly weakened binding affinity
for CARD1 relative to wild-type adenovirus. Such mutations may be
in sequences which encode either the AB loop, or the HI loop of the
fiber protein knob domain. Specific residues and mutations are
described. Another aspect of the present invention is a method for
generating a mutant adenovirus which is characterized by a receptor
binding affinity or specificity which differs substantially from
wild type., from an adenovirus which binds CARD1. In the method,
residues of the adenovirus fiber protein knob domain which are
predicted to alter D1 binding when mutated, are identified from the
crystal structure coordinates of the AD12knob:CAR-D1 complex. A
mutation which alters one or more of the identified residues is
introduced into the genome of the adenovirus, and whether or not
the mutant produced exhibits altered adenovirus-CAR binding
properties is determined. Mutants produced by this method include
those which under physiological conditions, have significantly
weakened binding affinity for CARD1 relative to wild type
adenovirus and those which bind a receptor other than CARD1,
including an engineered receptor. Introduced mutations may produce
an amino acid insertion, deletion or substitution in the encoded
viral particle, and may serve to alter the conformation of one or
more residues of knob which participate directly in D1 binding.
Such residues include residues of the AB loop, the CD loop, the DE
loop, the FG loop, the E strand and the F strand. Alternatively,
the mutation may be directly introduced in a codon encoding the
residue of knob which participates directly in D1 binding. Specific
residues in the AB loop, the CD loop, the FG loop, the E strand,
the F strand, and the DE loop which participate directly in binding
are identified. Another aspect of the present invention is a method
for identifying an inhibitor of adenovirus binding to CAR. In the
method, a three-dimensional structure derived by X-ray diffraction
from a crystal of adenovirus knob trimer bound to CARD1 is provided
and then employed to design or select a potential inhibitor. The
potential inhibitor is synthesized and then whether or not the
potential inhibitor inhibits adenovirus binding to CAR is
determined. Preferred crystal structures and space group symmetry
is listed. A set of atomic coordinates which define the three
dimensional structure is provided. The potential inhibitor may be
designed to interact non-covalently with one or more residues of
the adenovirus fiber knob protein domain. Alternatively, the
potential inhibitor is designed to interact non-covalently with one
or more residues of CARD1. Specific residues for covalent and
non-covalent interaction are listed. The potential inhibitor may
also be designed to interact non-covalently with residues which
line a cavity formed during adenovirus knob trimer/CARD1 binding.
The potential inhibitor can be designed by identifying chemical
entities or fragments capable of associating with the adenovirus
knob trimer, and assembling the identified chemical entities or
fragments into a single molecule to provide the structure of said
potential inhibitor. Such an inhibitor may be designed de novo or
from a known inhibitor. Methods of inhibition include competitive
inhibition, non-competitive inhibition and uncompetitive
inhibition.
Inventors: |
Freimuth, Paul I.; (East
Setauket, NY) |
Correspondence
Address: |
Margaret C. Bogosian
Patent Counsel
Brookhaven National Laboratory
Bldg. 475D - P.O. Box 5000
Upton
NY
11973-5000
US
|
Assignee: |
Brookhaven Science
Associates
|
Family ID: |
22889439 |
Appl. No.: |
10/199290 |
Filed: |
July 22, 2002 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10199290 |
Jul 22, 2002 |
|
|
|
09389603 |
Sep 3, 1999 |
|
|
|
09389603 |
Sep 3, 1999 |
|
|
|
09236423 |
Jan 25, 1999 |
|
|
|
6395875 |
|
|
|
|
Current U.S.
Class: |
435/456 ;
435/235.1 |
Current CPC
Class: |
C07K 14/705 20130101;
A61K 47/42 20130101; C07K 2319/00 20130101 |
Class at
Publication: |
435/456 ;
435/235.1 |
International
Class: |
C12N 015/861; C12N
007/00 |
Goverment Interests
[0002] This invention was made with Government support under
contract number DE-AC02-98CH10886, awarded by the U.S. Department
of Energy. The Government has certain rights in the invention.
Claims
1. A mutant adenovirus having a genome comprising one or more
mutations in sequences which encode the AB loop of the fiber
protein knob domain, the viral particle encoded by the genome being
characterized by a significantly weakened binding affinity for
CARD1 relative to wild-type adenovirus.
2. The mutant adenovirus of claim 1 which is adenovirus serotype
2.
3. The mutant adenovirus of claim 1 which is adenovirus serotype
5.
4. The mutant adenovirus of claim 1 wherein the mutation results in
an amino acid substitution at one or both positions which
correspond to residue 417 and 418 of Adenovirus serotype 12 in the
encoded viral particle.
5. The mutant adenovirus of claim 4 wherein the mutation results in
amino acid substitutions which correspond to P417E and P418A of
Adenovirus serotype 12 in the encoded viral particle.
6. The mutant adenovirus of claim 1 wherein the mutation results in
an insertion of threonine and isoleucine between residues which
correspond to residue 421 and 422 of Adenovirus serotype 12 in the
encoded viral particle.
7. The mutant adenovirus of claim 1 wherein the mutation results in
a deletion of residues which correspond to E425 and L426 of
Adenovirus serotype 12 in the encoded viral particle.
8. A mutant adenovirus having a genome comprising one or more
mutations in sequences which encode the HI loop of the fiber
protein knob domain, the viral particle encoded by the genome being
characterized by a significantly weakened binding affinity for
CARD1.
9. The mutant adenovirus of claim 8 which is Adenovirus Serotype
2.
10. The mutant adenovirus of claim 8 which is Adenovirus Serotype
5.
11. The mutant adenovirus of claim 8 wherein the mutation results
in a deletion of amino acids which correspond to amino acids G550
and I551 of Adenovirus serotype 12 of the encoded viral
particle.
12. A method for generating a mutant adenovirus, the mutant being
characterized by a receptor binding affinity or specificity which
differs substantially from wild type, comprising: a) providing an
adenovirus which binds CARD1; b) identifying residues of the
adenovirus fiber protein knob domain which when mutated are
predicted to alter D1 binding from the crystal structure
coordinates of the AD12knob:CAR-D1 complex; c) introducing a
mutation into the genome of the adenovirus which alters one or more
of the residues identified in step b); and d) determining that the
mutant produced exhibits altered adenovirus-CAR binding
properties.
13. The method of claim 12 wherein the mutant adenovirus generated
has significantly weakened binding affinity for CARD1 relative to
wild type adenovirus under physiological conditions.
14. The method of claim 12 wherein the mutant adenovirus generated
binds a receptor other than CARD1.
15. The method of claim 14 wherein the receptor to which the mutant
binds is an engineered receptor.
16. The method of claim 12 wherein the introduced mutation results
in an amino acid substitution, an amino acid deletion, or an amino
acid insertion in the encoded viral particle.
17. The method of claim 16 wherein the introduced mutation serves
to alter the conformation of one or more residues of knob which
participate directly in D1 binding.
18. The method of claim 17 wherein the residue which participates
directly in D1 binding is located in a region of knob selected from
the group consisting of the AB loop, the CD loop, the DE loop, the
FG loop, the E strand and the F strand.
19. The method of claim 16 wherein the mutation is introduced in a
codon encoding the residue of knob which participates directly in
D1 binding.
20. The method of claim 19 wherein the residue of knob which
directly participates in D1 binding is in the AB loop.
21. The method of claim 20 wherein the mutation is introduced at
the codon for the residue which corresponds to the Ad12 residue
selected from the group consisting of 409, 415, 417, 418, 419, 426,
and 429.
22. The method of claim 19 wherein the residue which directly
participates in D1 binding is in the CD loop.
23. The method of claim 22 wherein the mutation is introduced at
the codon for the residue which corresponds to the Ad12 residue
selected from the group consisting of 450 and 451.
24. The method of claim 19 wherein the residue which directly
participates in D1 binding is in the FG loop of knob.
25. The method of claim 24 wherein the mutation is introduced at
the codon for the residue which corresponds to the Ad12 residue
selected from the group consisting of 517, 519, 520 and 523.
26. The method of claim 19 wherein the residue which directly
participates in D1 binding is in the E strand of knob.
27. The method of claim 26 wherein the mutation is introduced at
the codon which encodes the residue corresponding to residue 494 of
Adenovirus serotype 12.
28. The method of claim 19 wherein the residue which directly
participates in D1 binding is in the F strand of knob.
29. The method of claim 28 wherein the mutation is introduced at
the codon which encodes a residue corresponding to residues 497 and
498 of Adenovirus serotype 12.
30. The method of claim 19 wherein the residue which directly
participates in D1 binding is in the DE loop of knob.
31. The method of claim 30 wherein the mutation is introduced at
the codon encoding the residue corresponding to residue 487 of
Adenovirus serotype 12.
32. A method for identifying an inhibitor of adenovirus binding to
CAR, comprising: a) providing a three-dimensional structure derived
by X-ray diffraction from a crystal of adenovirus knob trimer bound
to CARD1; b) employing the three-dimensional structure to design or
select a potential inhibitor; c) synthesizing the potential
inhibitor; and d) determining whether the potential inhibitor
inhibits adenovirus binding to CAR.
33. The method of claim 32 wherein the crystal of the
Ad12knob:CARD1 complex has P4.sub.332 space group symmetry with a
cubic unit cell with 167.85 angstroms per side.
34. The method of claim 33 wherein the three dimensional structure
is defined by atomic coordinates listed in the Protein Data Bank
under code 1KAC.
35. The method of claim 32 wherein the atomic coordinates of the
Ad12knob:CARD1 complex are obtained by means of computational
analysis.
36. The method of claim 32 wherein the potential inhibitor is
designed to interact non-covalently with one or more residues of
the adenovirus fiber knob protein domain.
37. The method of claim 36 wherein the adenovirus fiber knob
protein domain residues correspond to Ad12 residues selected from
the group consisting of D415, P417, P418, I426, V450, K451, Q487,
Q494, S497, V498, P517, P519, N520, and E523.
38. The method of claim 32 wherein the potential inhibitor is
designed to interact non-covalently with one or more residues of
CARD1.
39. The method of claim 38 wherein the residues of CARD1 are
selected from the group consisting of P33, D37, L39, V48, D49, V51,
L54, S56, Y61, E62, E63, Y64, K102, K104, A106 and P107.
40. The method of claim 32 wherein the potential inhibitor is
designed to interact non-covalently with residues which line a
cavity formed during adenovirus knob trimer/CARD1 binding.
41. The method of claim 32 wherein the potential inhibitor is
designed by identifying chemical entities or fragments capable of
associating with the adenovirus knob trimer, and assembling the
identified chemical entities or fragments into a single molecule to
provide the structure of said potential inhibitor.
42. The method of claim 41 wherein the potential inhibitor is
designed de novo.
43. The method of claim 41 wherein the potential inhibitor is
designed from a known inhibitor.
44. The method of claim 32 wherein the potential inhibitor is a
competitive inhibitor of adenovirus-CAR binding.
45. The method of claim 32 wherein the potential inhibitor is a
non-competitive or uncompetitive inhibitor of adenovirus-CAR
binding.
Description
[0001] The present application is a continuation-in-part of U.S.
patent application Ser. No. 09/236,423 filed on Jan. 25, 1999,and
is incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0003] Characterization of the molecular basis for virus attachment
to cells has importance both for understanding virus tropism and
for developing agents that inhibit virus binding or alter the
specificity of binding. Recently, a cellular receptor for
adenovirus type 2 and other closely related serotypes was
identified. This receptor, encoded by a single gene on human
chromosome 21 (Mayr et al., J. Virol. 71: 412-8 (1997)), is a 46 kD
glycoprotein which also serves as a receptor for group B
coxsackieviruses (CBV) and thus was termed CAR. CAR mRNA is present
in many human tissues. A broad tissue distribution of CAR protein
expression correlates with the broad tropism of CBV, but subgroup C
adenoviruses that are known to bind CAR have a much more restricted
tropism limited primarily to the upper respiratory tract. Thus,
other factors in addition to receptor availability clearly have
important roles in determining adenovirus tropism. Although
adenovirus binds to CAR with high affinity (Mayr et al., J. Virol.
71: 412-8 (1997); Wickham et al., Cell. 73: 309-19 (1993)), virus
Liters are significantly reduced on cells with down-regulated CAR
expression (Freimuth, P., J. Virol. 70: 4081-5 (1996)). These
results suggest that adenovirus infection in vivo may be restricted
to cells which express CAR at levels above a minimum threshold
concentration. CAR protein levels are relatively low on the apical
surface of differentiated (ciliated) respiratory epithelial cell
cultures, which may account for the poor efficiency of adenoviral
gene transfer to human lung tissue in vivo.
[0004] Adenovirus binding to CAR results from an interaction
between rod-shaped proteins located at the capsid vertices, called
viral fibers, and the extracellular region of CAR. The monomers of
this homotrimeric fiber protein range in size from 30 to 65 kDa
depending on the serotype (Huang et al., J. Virol. 73: 2798-2802
(1999)). They are composed of a conserved amino terminal tail that
mediates their interaction with the Ad penton base, a
variable-length elongated (shaft) domain, and a carboxyl-terminal
globular domain, termed the knob, which mediates the high-affinity
interaction with its cellular receptor. The knob domain of
adenovirus type 5 (Ad5) was expressed in E. coli as a soluble,
trimeric, biologically active protein, and its 3-dimensional
structure was determined by x-ray crystallography (Xia et al.,
Structure 2: 1259-70 (1994)).
[0005] The predicted amino acid sequence of CAR suggests a
structure consisting of two extracellular domains related to the
immunoglobulin IgV and IgC2 domain folds (Bork et al., J. Mol Biol.
242: 309-20 (1994); Bergelson et al., Science 275: 1320-3 (1997);
Tomko et al., Proc. Natl. Acad. Sci. USA 94: 3352-6 (1997)), a
single membrane-spanning region, and one carboxy-terminal
cytoplasmic domain. Regions of CAR necessary for binding the fiber
knob domain have not yet been determined.
SUMMARY OF THE INVENTION
[0006] The present invention relates to a mutant adenovirus which
has a genome comprising one or more mutations in sequences which
encode the fiber protein knob domain, the viral particle encoded by
the genome being characterized by a significantly weakened binding
affinity for CARD1 relative to wild-type adenovirus. Preferably,
the mutant adenovirus is adenovirus serotype 2 or serotype 5. The
mutation may be in sequences which encode the AB loop of the fiber
protein knob domain. Specific residues and mutations are described.
Alternatively, the mutations which cause significantly weakened
binding affinity for CARD1 may be in sequences which encode the HI
loop of the fiber protein knob domain of the encoded viral
particle. Specific residues and mutations are described.
[0007] Another aspect of the present invention is a method for
generating a mutant adenovirus which is characterized by a receptor
binding affinity or specificity which differs substantially from
wild type. This method is performed on adenoviruses which bind
CARD1. Residues of the adenovirus fiber protein knob domain of the
adenovirus, which are predicted to alter D1 binding when mutated,
are identified from the crystal structure coordinates of the
AD12knob:CAR-D1 complex. A mutation which alters one or more of the
identified residues is introduced into the genome of the
adenovirus, and whether or not the mutant produced exhibits altered
adenovirus-CAR binding properties is determined. This method can be
used to produce a mutant adenovirus which, under physiological
conditions, has significantly weakened binding affinity for CARD1
relative to wild type adenovirus or which binds a receptor other
than CARD1, including an engineered receptor. The introduced
mutation may result in an amino acid substitution, an amino acid
deletion, or an amino acid insertion in the encoded viral particle.
Introduced mutations may serve to alter the conformation of one or
more residues of knob which participate directly in D1 binding.
Such residues include residues of the AB loop, the CD loop, the DE
loop, the FG loop, the E strand and the F strand. Alternatively,
the mutation may be introduced in a codon encoding the residue of
knob which participates directly in D1 binding. Specific residues
in the AB loop, the CD loop, the FG loop, the E strand, the F
strand, and the DE loop which participate directly in binding are
identified.
[0008] Another aspect of the present invention is a method for
identifying an inhibitor of adenovirus binding to CAR. In the
method, a three-dimensional structure derived by X-ray diffraction
from a crystal of adenovirus knob trimer bound to CARD1 is provided
and then employed to design or select a potential inhibitor. The
potential inhibitor is synthesized and then whether or not the
potential inhibitor inhibits adenovirus binding to CAR is
determined. The crystal of the Ad12knob:CARD1 complex which is used
in the method preferably has P4.sub.332 space group symmetry with a
cubic unit cell with 167.85 angstroms per side. Atomic coordinates
are preferably obtained by means of computational analysis. A set
of atomic coordinates which define the three dimensional structure
are provided. In one embodiment, the potential inhibitor is
designed to interact non-covalently with one or more residues of
the adenovirus fiber knob protein domain. In another embodiment,
the potential inhibitor is designed to interact non-covalently with
one or more residues of CARD1. Specific residues for covalent and
non-covalent interaction are listed. In another embodiment, the
potential inhibitor is designed to interact non-covalently with
residues which line a cavity formed during adenovirus knob
trimer/CARD1 binding. The potential inhibitor can be designed by
identifying chemical entities or fragments capable of associating
with the adenovirus knob trimer, and assembling the identified
chemical entities or fragments into a single molecule to provide
the structure of said potential inhibitor. Such an inhibitor may be
designed de novo or from a known inhibitor. Methods of inhibition
include competitive inhibition, non-competitive inhibition and
uncompetitive inhibition.
BRIEF DESCRIPTION OF THE DRAWINGS
[0009] FIG. 1 is a diagrammatic representation of the Ad12 fiber
knob and the extracellular domains of human CAR. a) The Ad12 knob
domain (solid line) begins at a conserved motif (amino acids
409-412) and extends to the fiber protein carboxy terminus (Glu
587) (corresponding to nucleotides 30592-31128 of GenBank Accession
#X73487). A fragment of Ad12 DNA encoding the entire knob domain
and several amino acids from the preceding fiber shaft region
(hatched box) was amplified by PCR using forward primer #1 and
reverse primer #2. The resulting PCR product was cloned between the
NdeI and RamHI sites of pET15b. b) The human CAR protein consists
of a N-terminal signal peptide (open box), two extracellular
Ig-related domains (D1, D2), a membrane spanning region (TM) and a
cytoplasmic domain (CYT). cDNA fragments encoding D1 and D1/D2 were
amplified by PCR using forward primer #3 and reverse primers #4 and
#5. The resulting PCR products were cloned between the NcoI and
XhoI sites of pET20b. Similar D1- and D1/D2-encoding cDNA fragments
were amplified by PCR using forward primer #6 and reverse primers
#7 and #8. The resulting PCR products were cloned between the NdeI
and BamHI sites of pET15b. The NcoI-XhoI fragments were transferred
from pET20b into to pET15b, a manipulation which resulted in the
fusion of the genes in frame to pET15b vector DNA encoding a 22
amino acid extension at the carboxy-terminus. c) pET vectors for
protein expression in E. coli. The open and filled boxes represent
bacterial signal peptides and hexahistidine tags, respectively. The
restriction sites used in this study are shown, and the sequence of
the pET15b-encoded 22 amino acid carboxy-terminal extension of sD1
is indicated in single letter code.
[0010] FIG. 2 is a diagrammatic representation of results from
experiments measuring the ability of sD1 and Ad12 knob to prevent
infection of cells by Ad2. HeLa cell monolayers were infected with
about 200 focus-forming units (FFU) per well of Ad2 virus in the
presence or absence of sD1 or Ad12 knob. The number of infected
cells that resulted is shown (mean.+-.SD). Control cultures were
pretreated with sD1 and knob (pre-sD1, pre-Knob) and then washed
prior to infection.
[0011] FIG. 3 is a diagrammatic representation of data from
scanning transmission electron microscopy (STEM) analysis
experiments measuring the mass of Ad12 knob.
[0012] FIG. 4 is a diagrammatic representation of data from size
exclusion chromatography of knob, CAR domains and knob-CAR complex.
20 .mu.L aliquots of purified 41.7 .mu.M CAR D1/D2 (___), 40.7
.mu.M Ad12 knob ( - - - ), or a mixture containing 32.4 .mu.M knob
and 30 .mu.M D1/D2 ( . . . ) were chromatographed on a Superose 6
gel filtration column at a flow rate of 0.25 mL/min (all of the
concentrations given refer to the monomeric species). The marks
show the elution position of size markers: a, earthworm hemoglobin
(3.8 MDa); b, dodecameric earthworm hemoglobin (200 kDa); c, bovine
serum albumin (67 kDa); and d, cytochrome c (12 kDa).
[0013] FIG. 5 consists of Ribbon diagrams of Ad12 knob, CARD1 and
the Ad12 knob CARD1 complex: a) Ribbon diagram of the Ad12 knob
trimer viewed down the viral fiber. The core of each knob monomer
comprises an eight-stranded antiparallel .beta.-sandwich, although
approximately 65% of the residues are found in ordered surface
exposed loops and turns. The V-sheet, composed of strands J, C, B
& A, is colored magenta. The R-sheet, composed of strands G, H,
D & I and the HI loop, are colored purple. All other regions
are colored in grey. Approximately 2050 .ANG. of primarily
hydrophobic surface is buried per monomer upon trimerization. b)
Ribbon diagram of the CARD1 domain from the complex colored in a
rainbow from blue to red. Strands in the foreground are D, E, B
& A and in the background of the diagram C", C', C, F & G
from left to right, respectively. c) Ribbon diagram of the Ad12
knob CAR complex viewed as in a). Each CAR molecule, colored cyan,
binds at the interface of two knob molecules. The three Ad12 knob
molecules are colored red, green and blue, respectively. d) Ribbon
diagram of the Ad12 knob CAR complex in a perpendicular view to c).
The molecules are colored as in c). The N-terminal residues of Ad12
knob are directed away from the membrane-binding surface whereas
the C-terminal residues of the CAR molecule face towards the
membrane. In this orientation, the putative N-glycosylation site
N106 in CARD1 would be located on the face opposite of knob. This
figure was generated in MOLSCRIPT (Kraulis, P. J., J. Appl.
Crystallog. 24: 946-950 (1991)).
[0014] FIG. 6 is a molecular surface representation of the
interface in the Ad12 knob-CARD1 complex: a) Sequence conservation
surface diagram of two knob monomers viewed at the CAR interface.
The molecules are colored on a sliding scale from white (conserved)
to red (no conservation). A white strip of conservation transects
the surface of the molecule. b) shows two knob monomers, colored
red and green, in the same view as a), with an additional CPK
representation of CARD1 also shown. Upon binding, the CARD1
molecule occludes the conserved strip on Ad12 knob. Atoms in CARD1
are colored as follows: carbon-cyan, oxygen-red, nitrogen-blue and
sulfur-green. c) is a surface diagram of two adjacent Ad12 knob
monomers shown in the same view as a). Atoms forming direct
contacts with CARD1 are colored yellow. All other atoms in monomer
1 and 2 are colored red. The atoms in contact with CARD1 come from
both monomers. d) Is a surface diagram of CARD1 shown rotated
180.degree. around the to y-axis relative to CARD1 shown in b).
Atoms involved in direct contact with the Ad12 molecules are
colored magenta and all other atoms are colored cyan. In both c)
and d) the molecules inscribe a ring of contacting residues,
creating two cavities. This figure was generated with GRASP
(Nicholls et al., Proteins 11: 281-296 (1991)).
[0015] FIG. 7 is a representation of the AB loop of the Ad12 knob
CARD1 Interface: a) Is a ribbon diagram of the interface. Residues
in the AB loop are colored yellow, the remainder of that monomer
are red and residues in the other Ad12 knob monomer are green.
Residues in CARD1 are colored cyan. Residues involved in direct
interactions at the interface are shown as CPK representations.
Residues in Ad12 knob that play a role in the interface are: D415,
P417, P418, I426 (AB-loop), V450, K451 (CD-loop) and Q487, Q494,
S497 & V498 (E and F strands) from one monomer, and P517, P519,
N520 & E523 (FG-loop) from the other monomer. In CARD1, the
interfacial residues are P33, D37, L39, V48, D49, V51, L54, S56,
Y61, E62, E63, Y64 (Strands C, C', and C") and K102, K104, A106 and
P107 (later half of F). b) The AB loop is represented as a CPK
model in yellow, with the molecular surface of CARD1 shown in cyan.
c) Is an amino acid sequence alignment of residues in the AB loop
for all knob subgroups. This includes the sequence of Ad12 (SEQ ID
NO: 10), Ad2 (SEQ ID NO: 11), Ad9 (SEQ ID NO: 12), Ad4 (SEQ ID NO:
13), Ad40 long (SEQ ID NO: 14), Ad40 short (SEQ ID NO: 15), and Ad3
(SEQ ID NO: 16). d) Is a CPK model of the region around the two
cavities viewed in the same orientation as a). The cavities are
shown in magenta. Residues are colored as in a). The three
consecutive proline residues in Ad12 knob partially shape the
cavity. The AB loop shown, in yellow, lines one side of the cavity.
The cavity is lined with atoms from residues D415, P416 (backbone),
K429 (side), V448 (side), G449 (backbone), V450, L455 (side), Q535
(side), P573 (side), and S575 (side) from one Ad12 knob, and S514
(backbone), A515 (backbone), P517 (side), N520 (side), A524 (main),
E523, K525 S526 (side), from the other Ad12 knob and L39 (side),
K47 (backbone), V48 (backbone), D49, Q50, V51 and K102 (side) from
CAR (underlined residues are conserved in all Adenovirus
serotypes).
DETAILED DESCRIPTION OF THE INVENTION
[0016] The present invention is based, in one aspect, on the
discovery that the adenovirus-binding activity of human CAR is
localized in the amino-terminal IgV-related domain. As detailed in
the Exemplification section, the isolated amino-terminal
IgV-related domain of CAR (referred to herein as D1 or CARD1) and
the entire extracellular region (referred to herein as D1/D2 or
CARD1/D2) both have the ability to form complexes with Ad12 knob.
Furthermore, the presence of free D1 in soluble form, inhibits Ad2
virus infection of HeLa cells. Collectively, these observations
indicate that D1 is the component of CAR responsible for the
adenovirus-binding activity.
[0017] One embodiment of the present invention is an isolated
polypeptide that binds adenovirus comprising an amino acid sequence
corresponding to the D1 domain of the human CAR protein. The
preferred embodiment is an isolated polypeptide comprising residues
20-144 of the pre-CAR sequence (GenBank Accession #Y07593), with
the amino acid substitutions of L20M and S21G, generated to
facilitate cloning. The wild type sequence comprising residues
20-144 of pre-CAR also binds adenovirus, as does a polypeptide
sequence comprising residues 20-144 of the pre-CAR sequence which
contains one or more conservative amino acid substitutions.
[0018] Another embodiment of the present invention is an isolated
polypeptide comprising an amino acid sequence corresponding to the
D1 and D2 domains of the human CAR protein. D1 and D2 are IgV and
IgC2 domains and constitute the entire extracellular region of the
human CAR protein. The D1/D2 polypeptide demonstrates analogous
viral binding activity to the D1 polypeptide described above. The
preferred embodiment comprises the amino acid sequence
corresponding to residues 20-237 of the human pre-CAR protein
(GenBank Accession #Y07593), with two amino acid substitutions of
L20M and S21G, generated to facilitate cloning. The wild type
sequence comprising residues 20-237 of pre-CAR also binds
adenovirus, as does a polypeptide sequence comprising residues
20-237 of the pre-CAR sequence which contains one or more
conservative amino acid substitutions.
[0019] Another aspect of the present invention is the recombinant
DNA molecule that encodes the above described polypeptides. One
embodiment is a recombinant DNA molecule comprising a DNA sequence
encoding the adenovirus binding domain, D1, of the CAR protein. In
the preferred embodiment, this sequence corresponds to nucleotide
121 to 493 of the human CAR cDNA (GenBank Accession #Y07593)
Another embodiment is the recombinant DNA molecule comprising the
entire extracellular region, D1 and D2, of the CAR protein. In the
preferred embodiment, this DNA sequence corresponds to nucleotide
121 to 770 of the human CAR cDNA sequence (GenBank Accession
#Y07593).
[0020] The isolated polypeptides of the present invention can be
produced in vitro by inserting the corresponding recombinant DNA
molecules described above into an expression vector (e.g. a
prokaryotic or eukaryotic expression vector). Such vectors contain
all necessary regulatory signals to promote the expression of a DNA
sequence of interest. The use of such vectors is a matter of
routine experimentation for one of average skill in the art. The
expression vector with the inserted DNA sequence of the present
invention is then introduced into an appropriate cell under
conditions favorable for expression. In the preferred embodiment,
the cell is prokaryotic and is a bacteria cell. However, the
proteins can also be expressed in eukaryotic cells by similar
methods, utilizing eukaryotic expression vectors. Such cells can be
used to study the biological properties of the protein in a
controlled cell system or, alternatively, for the purpose of
protein production and purification.
[0021] Isolation of the above proteins from the bacteria is
achieved through routine purification procedures. In one
embodiment, the CAR coding sequences are engineered downstream of
sequences encoding hexahistidine, to produce the CAR fragment with
an N-terminal hexahistidine tag. As described in the
Exemplification section of this application, the D1 and D1/D2
polypeptides produced by this method are insoluble when generated
in E. coli. However, functional products are obtained when the
polypeptides are refolded from urea-solubilized inclusion bodies
and purified by anion exchange chromatography. Following
purification, the tag is optionally cleaved off by digestion with
thrombin to yield the intact CAR fragment.
[0022] In the preferred embodiment, the D1 polypeptide is expressed
in the form of a fusion protein which results in the production of
D1 domain that is soluble and functional when exogenously expressed
in E. coli at 18.degree. C. As described in the Exemplification
section of the present application, D1 engineered to have a short
C-terminal amino acid extension is partially soluble when expressed
in E. coli, and also retains virus binding activity. Without being
bound by theory, the fused extension is thought to enable the IgV
domain to fold into a soluble structure within E. coli cells.
Functional D1 isolated in this manner is preferred for use over D1
which is produced as insoluble in E. coli and resolubilized because
resolubilized proteins can contain non-functional structural
isomers.
[0023] The fusion protein is generated by expression from a
recombinant DNA molecule containing the D1 polypeptide coding
sequence, described above, fused in frame to a DNA sequence
encoding a polypeptide sequence which facilitates folding of the D1
polypeptide into a functional, soluble domain. This recombinant DNA
molecule is then inserted into a prokaryotic expression vector
which is then transformed into a bacteria cell, under conditions
appropriate for expression. In one embodiment the fusion is
downstream, resulting in a C-terminal extension In the preferred
embodiment, the D1 coding sequence is fused in frame to a
downstream DNA sequence encoding the 22 residue polypeptide
LEDPAANKARKEAELAAATAEQ (SEQ ID NO: 1) to generate a C-terminal
extension. The isolated polypeptide that results from expression of
this fused sequence comprises an amino acid sequence corresponding
to amino acids 20-144 of human pre-CAR protein, and is herein
referred to as sD1.
[0024] The present invention is also based, in part, on the
discovery that free sD1 polypeptide functions as an antiviral agent
by inhibiting viral infection of a cell. Results presented in the
Exemplification section of this application indicate that free sD1
functions to inhibit cell infection by viruses that bind D1 of
human CAR. As detailed in the Exemplification, both Ad2 and Ad12
(representative of adenovirus subgroup C and A, respectively) bind
D1. These results, combined with the earlier observation that
adenovirus competes for cell binding sites with a subgroup B
coxsackievirus, indicate that members of coxsackievirus subgroup B
also bind D1 (Lonberg-Holm et al., Nature 259: 679-81 (1976);
Bergelson et al., Science 275: 1320-3 (1997); Tomko et al., Proc.
Natl. Acad. Sci. USA 94: 3352-6 (1997)).
[0025] One embodiment of the present invention is a method for the
treatment of a patient infected with a virus characterized as
binding D1 of human CAR protein. The method comprises providing a
therapeutic composition of D1 polypeptide and administering it to
the host. This method can be used to treat any viral infection
involving a viral agent that binds to D1 including, but not limited
to, adenovirus subgroup A, adenovirus subgroup C, and
coxsackievirus subgroup B.
[0026] Effective therapeutic compositions will provide a sufficient
amount of D1 polypeptide to affect binding of the virus to the
extent that progression or spread of the infection is inhibited. In
the preferred embodiment, the therapeutic composition comprises the
D1 polypeptide in a stable, soluble form. For effective therapy,
administration of the composition is targeted to the infected area,
preferably through topical administration to a localized infection.
Adenoviruses commonly infect the upper respiratory tract, the
ocular region, and the gastro-intestinal tract, whereas CBV has a
broad tissue tropism. A therapeutic composition may take the form
of eyedrops, an inhalant fluid, or an ingestible composition, for
the treatment of an ocular, upper respiratory, or gastro-intestinal
infection, respectively.
[0027] The present invention also encompasses several other methods
that utilize the above described compositions. In addition to using
D1 polypeptide directly in the therapeutic treatment of viral
infection, the isolated polypeptide can be exploited experimentally
to identify and characterize molecules which bind CAR through the
D1 domain, to study the infection process, and to develop new
therapeutics. One such embodiment is a method of identifying
molecules and portions thereof involved in binding to CAR through
the D1 domain.
[0028] Experiments presented in the Exemplification section of this
application indicate that CAR is bound by adenovirus-encoded
proteins involved in cellular attachment. Evidence indicates that
other viruses also bind CAR at the D1 domain in the infection
process. In addition to this role as a receptor in viral infection,
CAR is likely bound by a natural ligand in a healthy individual.
Identification of D1 as the domain through which viruses attach to
CAR, allows its use in binding assays in the identification and
further characterization of these molecules.
[0029] Various binding assays can be used to identify and
characterize D1 binding molecules. In vitro binding assays yield
highly quantitative binding data, and have the advantage of being
performed under extremely controlled conditions. In vivo binding
assays are performed under physiological conditions and, while
often more qualitative than quantitative, can provide
physiologically relevant data.
[0030] In one embodiment, the molecule which binds D1 is an
adenovirus knob protein. Presumably, the residues of knob that form
the interface with CAR are conserved in adenovirus serotypes which
bind to CAR, and different in serotypes which do not. The Ad2,
representative of subgroup C, and Ad12, representative of subgroup
A, knob amino acid sequences are only 43% identical, yet both
viruses use CAR as the major attachment receptor. The importance of
these conserved amino acids to CAR binding are readily testable
using the polypeptide sequences disclosed in the present
application. Along the same lines, evidence indicates that subgroup
B adenovirus (e.g. types 3 and 7) do not bind the same cell
receptors as subgroup A and C (Defer et al., J Virol. 64: 3661-73
(1990)): Regions of subgroup B and C knob sequences that differ
radically may define receptor binding specificity (Xia et al.,
Structure. 2: 1259-70 (1994)).
[0031] In the preferred embodiment, an in vitro binding assay
system is used to determine if alterations in viral knob proteins
affect binding to D1 or D1/D2 polypeptides. These alterations
include truncations and internal deletions of D1 binding knob
proteins. Additionally, chimeras of D1 binding and non-D1 binding
knob proteins can be tested for D1 binding, to determine the
influence of defined knob protein regions on binding specificity.
As experiments progress, subtle alterations made in the identified
binding regions (e.g. amino acid substitutions) will define the
residues involved in D1 binding. One of average skill in the art
will recognize that the results of such studies can be applied to
the production of recombinant adenovirus knob proteins and/or D1
proteins that possess distinct binding specificities.
[0032] Identification of D1 as the virus binding fragment also
facilitates structural determination of the virus attachment
proteins (e.g. Ad 12 knob, complexed to D1 or D1/D2). Such studies
facilitate the identification of residues that form the
receptor-knob interface.
[0033] The methods used in the study of knob binding are equally
applicable to identifying molecules and portions thereof involved
in the attachment of viruses other than adenovirus (e.g.
coxsackievirus and natural ligands to CAR) One of skill in the art
will recognize that information gathered from binding studies will
usefully contribute to related applications of the present
invention.
[0034] The compositions of the present invention can also be used
in binding assays to screen for antiviral compounds which interfere
specifically with the interaction between the D1 polypeptide of the
human CAR protein and a viral attachment protein which binds to the
D1 polypeptide The demonstrated binding of D1 to viral attachment
proteins, as detailed in the Exemplification, can be used as an
assay for a primary step in virus infection. Compounds found to
interfere with viral binding will have valuable therapeutic and
experimental potential as antiviral agents.
[0035] In one embodiment, purified viral attachment protein is
contacted with purified sD1 in an in vitro binding assay (e.g.
ELISA type) in the presence and absence of a candidate compound.
Decreased binding of D1 to the viral protein indicates the compound
possesses anti-viral properties. Both in vitro and in vivo binding
assays utilizing various detection methods can be used in such a
screen. The performance of such binding assays is within the
capabilities of one of average skill in the art. In the preferred
embodiment; the viral attachment protein used in the screen is an
adenovirus knob protein (e.g. Ad2 or Ad12).
[0036] The present invention also provides methods to specifically
target a cell for viral infection by redirecting viral binding to a
predetermined surface protein of the targeted cell. This can be
accomplished by using a D1 containing adaptor bridge which binds to
the virus and also binds to the surface protein. Because adenovirus
infection in vivo is restricted to cells which express CAR at
levels above a minimum threshold concentration, the use of a highly
expressed surface protein is expected to produce a higher rate of
infection of the target cell than will a less prevalent surface
protein. In one embodiment, the adaptor bridge is generated by
fusing D1 to single chain antibodies directed towards antigens
expressed on the target cells (e.g. tumor cells). The virus is then
contacted with the adaptor bridge under conditions appropriate for
binding of the virus to the D1 portion of the adaptor bridge, to
produce a virus-adaptor bridge complex. The target cell is then
contacted with the virus-adaptor bridge complex under conditions
appropriate for binding of the antibody portion of the adaptor
bridge to the target cell. Contact of the virus-adaptor bridge
complex with the target cell can take place via topical application
or systemic administration. Upon binding of the virus-adaptor
bridge complex to the target protein, the target cell becomes
susceptible to infection by the attached virus. In one embodiment,
the virus is an adenovirus. However, the virus can be any virus
that binds D1, including a virus modified for therapeutic purposes
(e.g. by recombinant engineering).
[0037] Adenovirus-based DNA expression vectors and delivery systems
are highly utilized systems for gene delivery into animal cells,
including in vitro cell culture and in vivo delivery (e.g. gene
therapy). The use of specific targeting of adenoviruses allows the
development of either a wider spectrum of target cells or
conversely a narrower range of delivery, the latter improvement
being beneficial to therapies such as chemotherapy aimed at
specific elimination of diseased tissue. One skilled in the art can
envision how information gathered in the above experiments
regarding the binding sites involved in Adenovirus binding to D1
can be exploited for therapeutic purposes to generate recombinant
adenovirus and D1 with highly specific binding recognition.
[0038] The present invention also provides methods for treating a
patient with an infection caused by a virus that binds to human
CAR. Experiments detailed in the Exemplification indicate that free
Ad12 knob inhibits infection of HeLa cells by Ad2 virus. These
observations indicate that free adenovirus Ad12 knob protein
administered in a therapeutic composition can prevent the spread of
an infection resulting from a virus that binds D1. The preparation
and administration of effective therapeutic compositions comprising
the Ad 12 knob protein are similar to that described above for
D1.
[0039] Other aspects of the present invention relate to the
three-dimensional structure of the fiber knob protein domain of
Adenovirus serotype 12 (referred to herein as Ad12 knob) alone and
in complex with the isolated and purified D1 domain of CAR,
provided by Applicants. The structure coordinates of the Ad12 knob
and the complex of Ad12 knob and CARD1 are deposited in the Protein
Data Bank, access identification numbers 1NOB and 1KAC respectively
The term "structure coordinates" refers to mathematical coordinates
derived from mathematical equations related to the patterns
obtained on diffraction of a monochromatic beam of X-rays by the
atoms (scattering centers) of the molecule or molecular complex in
crystal form. The diffraction data obtained are used to calculate
an electron density map of the repeating unit of the crystal. The
electron density maps are used to establish the positions of the
individual atoms within the unit cell of the crystal. Those of
skill in the art understand that a set of structure coordinates
determined by X-ray crystallography is not without standard
error.
[0040] The structure coordinates provided herein may be modified
from the original set by mathematical manipulation. Such
manipulations include, but are not limited to, crystallographic
permutations of the raw structure coordinates, fractionalization of
the raw structure coordinates, integer additions or subtractions to
sets of the raw structure coordinates, inversion of the raw
structure coordinates, and any combination thereof.
[0041] Structural analysis presented in Section II of the
Exemplification below was used to identify the CAR binding site on
Ad12 knob and specific residues in Ad12 knob and CARD1 which
directly interact. Regions of Ad12 knob identified as directly
involved in CARD1 binding are the AB loop, the carboxyl ends of the
DE loop, the F strand and the FG loop. The Ad12knob:CARD1 complex
forms without a significant conformational change in the knob
domain, and buries a large mixed hydrophilic/hydrophobic surface
area involving several non-contiguous loops of Ad12 knob and the
.beta.-strands on one face of CARD1, creating two buried cavities
totaling .sup..about.120 .ANG..sup.3. Structure based mutational
analysis presented below supports the structural model deduced from
this analysis.
[0042] One aspect of the present invention is a mutant adenovirus
having a genome comprising one or more mutations in sequences which
encode regions of the fiber protein knob domain specifically
involved in binding of adenovirus to CARD1. The mutant viral
particle encoded by the mutated genome is characterized by a
significantly weakened binding affinity for CARD1 relative to the
CARD1 binding affinity of wild-type adenovirus. A significantly
weakened binding affinity for CARD1 is defined herein as a
reproducible lower binding affinity for CARD1 compared to wild type
as determined by binding assays known in the art. A mutant
adenovirus with undetectable binding affinity for CARD1 is also
considered to have a weakened binding affinity for CARD1, the
affinity being so minimal as to be undetectable. weakened binding
affinity as used herein may also include a coincident gain of
binding affinity for another receptor molecule, as compared to
wild-type adenovirus. The term receptor, as used herein, refers to
any molecule which a virus binds to gain entry into a cell.
[0043] Using the structural data provided in the Exemplification
section below several mutants of Ad12 which have significantly
weakened binding affinity for CARD1 have been generated. One of
skill in the art will recognize that mutations analogous to those
identified in Ad12 can be generated in any serotype which binds
CARD1 to produce a similar weakened binding affinity for CARD1 in
the encoded virion relative to wild type. As such, the present
invention is intended to encompass all adenovirus serotypes which
bind CARD1. The specific mutation in another CARD1 binding serotype
can be generated by one of skill in the art through no more than
routine experimentation by examination of the amino acid sequence
homology of Ad12 knob versus other CARD1-binding serotypes.
[0044] Many of the mutants produced have one or more mutations in
sequences which encode the AB loop The AB loop of knob contributes
over 50% of the interfacial interactions with D1, and is one of the
identified candidate regions for mutagenesis affecting receptor
binding. Specific positions identified for mutagenesis include
position 417 and 418 of Ad12. Mutations at one or both of these
positions results in weakened binding affinity for CARD1. Examples
of specific point mutations at these positions identified as
inhibiting D1 binding are the substitution mutations P417E and
P418A. In addition to substitution mutations, insertion and
deletions at sites within the AB loop also lead to altered receptor
binding affinity. For example, insertion of two amino acids,
threonine and isoleucine, between residue 421 and 422 of Ad12
results in reduced CARD binding. Deletion of E425 and L426 of AD12
also reduces CARD1 binding.
[0045] Other mutant adenoviruses with reduced CARD1 binding have
been generated from mutations in the HI loop of knob. The
concurrent deletion of amino acid G550 and I551 of AD12 results in
reduced CARD1 binding.
[0046] Mutants of adenovirus serotypes currently being used for
gene transfer vectors have particular value as a therapeutic
tool.
[0047] CARD1 binding mutants of the present invention can be
engineered to function as vectors for gene therapy Recently, human
adenovirus serotypes 2 and 5 have been adapted for use as vectors
for efficient introduction of genes in vivo (Trapnell et al.,
Current Opinion Biotech. 5: 617-625 (1994); Acsadi et al., J. Mol.
Med. 73: 165-180 (1995)). The adapted viruses are usually disabled
(e.g. unable to propagate under natural conditions) in order to
reduce the risk of transmission of the gene therapy vector from
patients to the general population. Adenovirus vectors currently
available for use in gene therapy and methods of producing the
vectors are described by Massie et al., U.S. Pat. No. 5,891,690
(1999); Armentano et al., U.S. Pat. No. 5,824,544 (1998); Armentano
et al., U.S. Pat. No. 5,707,618 (1998); Hammond et al., U.S. Pat.
No. 5,792,453 (1998); the contents of which are incorporated herein
by reference. Despite these advances, concern remains regarding the
possible arisal of revertants which become replication competent
through recombination with adenoviral sequences naturally present
in a host recipient. Generation and use of a CARD1 binding mutant
of the present invention in the form of a vector for gene therapy
will provide an extra measure of control against possible
transmission of a therapeutic mutant from patients to the general
population.
[0048] Because the introduced mutations will also block the
efficiency of gene transfer through CAR-mediated infection, it may
be beneficial to introduce the gene delivery vectors of the present
invention into the host cell by an alternate method. For instance,
the viral vector can be further engineered to bind receptors other
than CAR for cellular entry. Currently research is underway to
develop laboratory strains of adenoviruses which infect through
novel receptors other than CAR to direct therapeutic adenovirus
infection (e.g. a gene delivery vector) to specific target cells.
Successful results have been achieved using bi-specific antibodies,
where one arm of the antibody binds some component of the virus
surface and the other arm binds to a molecule on the cell surface.
Recombinant adenoviral gene delivery vectors which are similarly
adapted to infect via alternate receptors to confer tissue
specificity and methods for their generation have been described by
Wickham et al., U.S. Pat. No. 5,559,099 (1996); Wickham et al.,
U.S. Pat. No. 5,770,442 (1998); Wickam et al., U.S. Pat. No.
5,731,190 (1998); the contents of which are incorporated herein by
reference. Adenovirus mutants of the present invention can be
generated by one of skill in the art to produce similar tissue or
cell type specific delivery vectors.
[0049] Ideally, an adenovirus mutant of the present invention has a
significantly weakened binding affinity for CARD1 such that it can
be propagated efficiently under laboratory conditions, but not in
natural infections in humans. Such a mutant would be particularly
useful as a gene therapy vector.
[0050] Another aspect of the present invention is a method for
generating a mutant adenovirus which is characterized by a receptor
binding affinity or specificity which differs substantially from
wild type. This method is applicable to adenovirus serotypes and
mutants thereof which bind CARD1. Generally the method will be used
to produce a mutant which has a reduced affinity for CARD1 relative
to that of wild type. Similarly, a mutant which binds a receptor
other than CARD1 may also be produced from a wild type virus.
[0051] The Exemplification below demonstrates by example that the
findings regarding the structural interactions of Ad12 knob with
CARD1 are readily applicable for generating adenovirus mutants
which have altered receptor binding properties. Specifically, the
information regarding the Ad12knob:CARD1 complex provided in the
Exemplification section below, allows identification of residues of
the adenovirus fiber protein knob domain which when mutated are
predicted to alter D1 binding. Upon identification of one or more
of these residues, a mutation is introduced into the genome of the
adenovirus which alters one or more of the identified residues. The
mutant which is produced from expression of the mutated genome is
then tested in order to to determine that the mutant exhibits
altered adenovirus-CAR binding properties.
[0052] Residues of knob which when mutated are predicted to alter
the binding properties of the virus when mutated include residues
which are involved, either directly or indirectly, in D1 binding.
Examples of amino acids which are directly involved with binding
include, without limitation, residues which make contact with the
receptor, and also residues which line or define binding sites.
Residues directly involved in D1 binding also include residues
which contribute directly to the topological mismatches in the
interface with D1 which result in the formation of cavities upon
binding. Because these cavities are thought to stabilize the
complex, alteration of one or more residues which line or define
these cavities is expected to alter knob/D1 binding.
[0053] Specific regions of Ad12 identified as containing residues
which participate directly in D1 binding include without limitation
the AB loop, the CD loop, the FG loop, the E strand and the F
strand. Residues in the AB loop involved in binding include without
limitation residues 409, 415, 417, 418, 419, 426, and 429. Residues
in the CD loop involved in binding include without limitation
residues 450 and 451. Residues in the FG loop involved in binding
include without limitation residues 517, 519, 520 and 523. Residues
in the E strand involved in binding include without limitation
residue 494. Residues in the F strand involved in binding include
without limitation residues 497 and 498. In addition, residue 487,
which is in the DE loop, preceding strand E, was also identified as
participating directly in D1 binding. The analogous residues can be
identified in other adenovirus serotypes which bind CARD1 through
examination of the sequence homology between the respective knob
domains of the relevant viruses.
[0054] Amino acids which have indirect involvement influence
residues which are directly involved. Examples include, without
limitation, residues adjacent or near directly-involved residues.
Proximity need not be limited to primary sequence, but may be from
secondary or tertiary structural relationships. In addition,
residues may be located near directly-involved residues due to the
formation of intra-molecular complexes by knob monomers. The
influence that indirectly-involved amino acids have on
directly-involved amino acids may be steric, chemical, or a
combination effects. Indirectly-involved amino acids also include
residues which participate in defining an element of the protein
structure which is crucial for receptor binding (e.g. the necessary
conformation of a protein or the ability to form intra-molecular
complexes).
[0055] Importantly, amino acids which are not significantly
involved (directly or indirectly) in receptor binding of the
unmutated adenovirus, may assume a role (direct or indirect) in
receptor binding through mutagenesis. The term "unmutated
adenovirus" as used herein refers to the original sequence prior to
a specific round of mutagenesis, the term being inclusive of wild
type sequences as well as previously altered sequences. Mutagenesis
which confers a role in function to a previously uninvolved residue
commonly involves the substitution of another amino acid at that
particular position. However, a new role in function may also be
conferred to a specific position by mutagenesis at another residue
position (e.g. to displace a target residue).
[0056] Two aspects should be considered regarding alteration of a
residue: 1) which of the 20 amino acids occupies that position
(e.g. properties the specific amino acid contributes to the
molecule at that position), and 2) the position of the specific
amino acid within the protein (e.g. how position affects the amino
acid's contributions to function). One way of altering a residue is
by substituting a different amino acid at that position within the
polypeptide chain. Alternatively, the positional location of a
specific amino acid within the polypeptide chain can be displaced.
This can be accomplished by inserting or deleting another residue,
either nearby or via a long range structural mutation. Also, the
specific residue can be deleted from the polypeptide chain without
replacement by another amino acid. Any number or combination of
these alterations can be used to produce a desired binding property
of the mutant produced.
[0057] Since the structural data obtained from adenovirus serotype
12 is readily applicable to other adenoviruses (e.g. different
serotypes and/or adenoviruses with other mutations) which bind
CARD1 this method is applicable to any adenovirus serotype which
binds CARD1. A residue(s) of the adenovirus fiber protein knob
domain which is predicted to alter the binding properties of knob
when mutated is first identified from the crystal structure
coordinates of the Ad12knob:CARD1 complex, and when necessary the
analogous residue(s) is identified through examination of amino
acid homology of relevant adenovirus serotypes. The mutation(s) is
then introduced into the genome of the adenovirus by standard
methods.
[0058] Once the desired mutation(s) is introduced, the encoded
mutant virus is produced and the CAR binding properties of the
mutant virus are analyzed (e.g. by comparison to wild type) to
verify that the mutant produced exhibits altered adenovirus-CAR
binding properties. This analysis is accomplished by performing any
number of standard binding assays known in the art. An altered
binding property includes either a reduction or elevation in CARD1
binding affinity. Often, the objective will be to produce a mutant
which has a reduction in CARD1 binding affinity. In one embodiment,
the mutant generated has significantly weakened binding affinity
for CARD1 relative to wild type adenovirus under physiological
conditions. In addition, the binding specificity can also be
altered, for example to produce a mutant adenovirus which binds a
different molecule, such as a CAR mutant or a receptor other than
CAR. In one embodiment, the receptor for which the mutant
adenovirus binds is an engineered receptor.
[0059] Once residues or regions of knob are identified for
mutagenesis, a random mutagenesis approach may be undertaken,
followed by screening the mutant virions generated for the desired
binding property. Alternatively, a more rational approach to
designing a specific mutant may be undertaken by systematically
introducing specific mutations calculated to produce the desired
binding property alteration. The Exemplification below lists
several Ad12 binding mutants produced by alteration of specific
residues in knob identified on the basis of their structural
location.
[0060] The present invention also provides a method for identifying
an inhibitor of adenovirus binding to CAR through rational drug
design. The method entails providing a three-dimensional structure
derived by X-ray diffraction from a crystal of the adenovirus knob
trimer bound to CARD1, and employing the three-dimensional
structure to design or select a potential inhibitor. Once
identified, the potential inhibitor is synthesized, and then tested
for the ability to inhibit adenovirus binding to CAR. A positive
result indicates that the potential inhibitor is an actual
inhibitor. In a preferred embodiment, the crystal of adenovirus
knob trimer bound to CARD1 has P4.sub.332 space group symmetry with
a cubic unit cell with 167.85 angstroms per side. The term "space
group" refers to the arrangement of symmetry elements of a crystal.
The term "unit cell" refers to the basic parallelepiped shaped
block. The entire volume of a crystal may be constructed by regular
assembly of such blocks. In a preferred embodiment, the atomic
coordinates of the Ad12knob:CARD1 complex are obtained by means of
computational analysis of X-ray diffraction data The
three-dimensional structure provided by Applicants is defined by
atomic coordinates listed in the Protein Data Base under code 1KAC,
and is well suited for use in this method.
[0061] A potential inhibitor may function by one of several
different putative mechanisms. Without limitation, the inhibitor
may function as a competitive inhibitor, a non-competitive or an
uncompetitive inhibitor. A competitive inhibitor is one that
inhibits activity by directly competing with the receptor or virus
for the binding site. Competitive inhibition can be reversed
completely by increasing the virus concentration. An uncompetitive
inhibitor is one that inhibits by binding to the virus or receptor
which is in complex. Uncompetitive inhibition cannot be reversed
completely by increasing virus concentration. A non-competitive
inhibitor can bind to either free or bound virus or CAR.
[0062] An inhibitor may be designed or selected to interact with a
particular component or specific residues of a target molecule(s)
to block binding. The target molecule(s) may be either the
adenovirus fiber knob protein domain or the D1 domain of the CAR
receptor, or both. The type of interaction between the inhibitor
and the target molecule(s) may be any form or combination of
interactions known in the art (e.g. covalent or non-covalent,
especially hydrogen bonding, hydrophobic, Van der walls,
electrostatic). Also, the inhibitor compound must be able to assume
a conformation which allows it to associate with the target
molecule(s). Such conformational requirements include the overall
three-dimensional structure and orientation of the chemical entity
or compound in relation to all or a portion of the binding site or
the spacing between functional groups of the inhibitor compound
comprising several chemical entities that directly interact with
the target molecule(s).
[0063] An inhibitor molecule may be computationally evaluated and
designed by means of a series of steps in which chemical entities
or fragments are screened and selected for their ability to
associate with the individual binding pockets or other areas of the
target molecule(s). The structure coordinates of the present
invention may also be used to screen computationally small molecule
data bases for compounds that bind to one of more of the complex
components.
[0064] Potential residues of knob to be targeted for interaction
with an inhibitor include, without limitation, residues
corresponding to D415, P417, P418, I426, V450, K451, Q487, Q494,
S497 and V498 in one monomer of the Ad12 knob trimer, and P517,
P519, N520, and E523 of the adjacent monomer of the Ad12 knob
trimer Preferably the interaction is non-covalent. Potential
residues of CARD1 to be targeted for inhibitor interaction include,
without limitation, P33, D37, L39, V48, D49, V51, L54, S56, Y61,
E62, E63, Y64, K102, K104, A106 and P107 of human CARD1.
Preferably, the interaction is non-covalent.
[0065] In another embodiment, the potential inhibitor is designed
to interact, preferably non-covalently, with residues of knob
and/or D1 which line one of the two cavities which are formed
during adenovirus/CARD1 binding. Such residues are ideal candidates
for interaction with a potential inhibitor, to block or disrupt
virus-receptor binding. Small molecules which specifically fit into
these cavities during binding and destabilize virus binding can be
designed from the information provided regarding the topological
mismatches of the interfacial components of D1 and knob. For
instance, a small molecule designed to bind to residues lining a
cavity of one component of the complex, but to repel residues
lining a cavity of the other component of the complex would inhibit
or disrupt complex formation and subsequently virus binding.
[0066] Potential inhibitors may be designed or assembled by a
variety of methods known in the art. For instance, a potential
inhibitor may be designed de novo, or alternatively may be designed
from a known molecule (e.g. a known inhibitor). Once suitable
chemical entities or fragments have been selected, they can be
assembled into a single inhibitor compound. Assembly may proceed by
visual inspection of the relationship of the fragments to each
other on the three-dimensional image displayed on a computer screen
in relation to the structure coordinates of the present invention.
This would be followed by manual model building using software such
as Quanta or Sybyl. Useful programs to aid one of skill in the art
in connecting the individual chemical entities or fragments
include:
[0067] 1. CAVEAT (Bartlett et al., "CAVEAT: A Program to Facilitate
the Structure-Derived Design of biologically Active Molecules". In
"Molecular Recognition in Chemical and Biological Problems",
Special Pub., Royal Chem. Soc., 78, p. 182-196 (1989)). CAVEAT is
available from the University of California, Berkeley, Calif.
[0068] 2. 3D Database systems such as MACCS-3D (MDL Information
Systems, San Leandro, Calif.) reviewed in Martin, Y. C., "3D
Database searching in Drug Design", J. Med. Chem., 35, p. 2145-2154
(1992).
[0069] 3. HOOK, available from Molecular Simulations, Burlington,
Mass.
[0070] In addition, inhibitory or other binding compounds may be
designed as a whole or de novo using an empty binding site or
optionally including portions of a known inhibitor(s). These
methods include:
[0071] 1. LUDI (Bohm, H. J., J. Comp. Aid. Molec. Design 6: 61-78
(1992)). LUDI si available from Biosym Technologies, San Diego,
Calif.
[0072] 2. LEGEND (Nishibata and Itai, Tetrahedron 47: 8985 (1991)).
LEGEND is available from Molecular Simulations, Burlington,
Mass.
[0073] 3. LeapFrog, available from Tripos Associates, St. Louis,
Mo.
[0074] Without limitation, other molecule modeling techniques which
may also be employed in accordance with this invention are
described by Cohen et al., J. Med. Chem. 33: 883-894 (1990), and
Navia et al., Current Opinions in Structural Bio. 2-202-210
(1992).
EXEMPLIFICATION
Section I
Expression and Characterization of Car Extracellular Fragments
Expression and Purification of CAR Extracellular Fragments
[0075] To localize the Ad-binding activity of CAR, fragments
corresponding the amino-terminal CAR IgV domain (D1) and the
combined IgV+IgC2 domains (D1/D2) were expressed in E. coli. A cDNA
fragment coding for D1 (FIG. 1b) was cloned into pET20b, an
expression vector designed to export expressed proteins into the E.
coli periplasmic space (FIG. 1c), but synthesis of D1 (expected
molecular weight of about 16 kDa) was undetectable after 3 hours of
induction. No bands corresponding to D1 were detected by SDS-PAGE
analysis of whole cell lysates. When the initial construct was
enlarged to include the downstream IgC2 domain (FIG 1b), however,
the resulting D1/D2 polypeptide was overexpressed and ran as a
closely-spaced doublet on SDS-PAGE, which is characteristic of some
periplasmic proteins such as alkaline phosphatase and results from
partial hydrolysis of the signal peptide. These results imply that
the amino-terminal domain (D1) specified by the initial construct
also entered the secretory pathway, but probably was rapidly
degraded in the periplasmic space. The D1/D2 protein fragment was
not soluble in E. coli cells grown at temperatures ranging from
18-37.degree. C.
[0076] To determine if D1 could be stabilized by restricting its
synthesis to the cytoplasm, the D1-encoding PCR product was
transferred as a NcoI-XhoI restriction fragment from pET20b into
pET15b (FIG. 1c). Because of restriction site differences between
these 2 expression vectors, the CAR protein fragment specified by
this construct (pET15b-sD1) had a vector-encoded 22-amino acid
carboxy-terminal extension and it lacked the amino-terminal
hexahistidine Lag that is normally attached to proteins expressed
from pET15b (FIG. 1c). The resulting polypeptide was expressed at
moderate abundance at 37.degree. C., but was insoluble. When the
cultures were induced at 18.degree. C., however, a significant
amount of D1 was contained in the soluble fraction of cell lysates.
SDS-PAGE of the lysate and fractions revealed a band corresponding
to the 16 kDa molecular weight of D1 present in all fractions. The
larger CAR cDNA fragment encoding D1/D2 also was transferred from
pET20b into pET15b, but none of the expressed protein was detected
in the soluble fraction of cell lysates. SDS-PAGE analysis of cell
lysate and fractions revealed bands corresponding to the molecular
weight of about 27 kDa in both the whole cell lysate and insoluble
fraction, but absent in the soluble fraction. Soluble D1 (sD1) was
partially purified by ammonium sulfate precipitation and
ion-exchange chromatography.
[0077] To determine if removal of the vector-encoded
carboxy-terminal extension would increase the yields of soluble CAR
fragments produced in E. coli, cDNA fragments encoding D1 and D1/D2
were amplified with new primer sets (primers 6-8, FIG. 1b) that
introduced downstream stop codons and also fused the proteins to
the vector-encoded amino-terminal hexahistidine tag. Both CAR
fragments were overexpressed, but were insoluble at culture growth
temperatures between 18-37.degree. C., suggesting that the
carboxy-terminal extension specified by the initial pET15b-sD1
construct may enable the IgV domain to fold into a soluble
structure within E. coli cells. The insoluble his-tagged CAR
fragments were both refolded from urea-solubilized inclusion bodies
and were purified to apparent homogeneity by anion exchange
chromatography. To confirm that D1 solubility within intact E. coli
cells depends on the presence of the 22 amino acid C-terminal
extension rather than the absence of the N-terminal hexahistidine
leader, the D1-encoding insert (PCR product of primers #6 and #7,
FIG. 1b) was transferred from pET15b into pET11a as an NdeI-BamHI
fragment (FIG. 1c). D1 was overexpressed in pET11a-D1-transformed
cells, but was completely insoluble, as determined by comparison of
whole cell lysate to soluble cell fractions by SDS-PAGE analysis,
confirming that the C-terminal 22 amino acid extension specified by
pET15b increases D1 solubility.
Biological Activity of CAR Extracellular Fragments
[0078] Refolded D1 and D1/D2 CAR fragments were examined for the
ability to form specific complexes with recombinant fiber knob from
Ad12. It was previously reported that infection of HeLa cells by
Ad12 virus is inhibited by purified native fiber protein from Ad2,
suggesting that CAR serves as the major attachment receptor for
both Ad2 and Ad12. A fragment of Ad12 DNA coding for the fiber knob
domain (FIG. 1a) was cloned in pET15b. Ad12 knob was abundantly
expressed following IPTG induction of cultures at 37.degree. C.,
but accumulated entirely within the insoluble fraction of cell
lysates. When cultures were induced at 24.degree. C., however, the
majority of knob was in the soluble fraction. The knob was purified
by ammonium sulfate precipitation and anion exchange
chromatography, and the his-tag was removed by digestion with
thrombin. A sample of purified Ad12 knob, visualized by SDS-PAGE,
displayed as a single band at the expected molecular weight. Ad12
knob was then incubated with the his-tagged D1 or D1/D2 in the
presence of purified Ad2 hexon protein (included as a specificity
control). The mixtures were then adsorbed to Ni-NTA beads to
capture the his-tagged CAR fragments. In control incubations
lacking the CAR fragments, Ad12 knob and Ad2 hexon both failed to
bind to Ni-NTA beads, demonstrated by an absence of bands upon
SDS-PAGE analysis of bead eluate. In the presence of either D1 or
D1/D2, however, Ad12 knob bound to Ni-NTA beads and could be easily
detected in bead eluate by SDS-PAGE analysis, whereas Ad2 hexon did
not. This suggested that the CAR IgV domain (D1) specifically binds
the Ad12 knob. This conclusion was supported by the results of an
experiment to test whether his-tagged Ad12 knob and sD1 form
specific complexes. Purified, his-tagged Ad12 knob was mixed with a
partially purified preparation of sD1 and incubated briefly to
allow protein complexes to form. The mixture was then applied to a
column of Ni-NTA beads, unbound proteins were washed from the
column, and the bound fraction was eluted with EDTA. SDS-PAGE
analysis of the bead eluate revealed the presence of Ad12 knob and
sD1, seen as two distinct bands at their expected molecular
weights. Thus, D1 alone is sufficient for binding to the Ad12
knob.
[0079] To determine if the binding activities of the recombinant
Ad12 knob and the CAR IgV domain have the same specificities as
their native fiber and CAR counterparts, the ability of Ad12 knob
and sD1 to inhibit Ad2 infection of HeLa cells was tested. As shown
in FIG. 2, Ad2 infectivity was significantly inhibited when either
sD1 or Ad12 knob was included in the virus inoculum during virus
adsorption. No inhibition of infection was observed in cell
cultures that were pretreated with sD1 and then washed prior to
virus adsorption, indicating that the inhibitory activity of sD1
does not result from a cytotoxic effect on cells. Cells similarly
pretreated with Ad12 knob, however, were still partially refractory
to infection by Ad2 virus. This most likely results from incomplete
dissociation of knob from the CAR receptors on cells rather than a
cytotoxic effect. Thus, the binding specificity of native fiber and
CAR is to reconstituted in recombinant Ad12 knob and sD1.
Physical Characteristics of Ad12 Knob and CAR Domains
[0080] Analysis of boiled and untreated samples of Ad12 knob by SDS
PAGE showed bands of 20 and 60 kDa, respectively, suggesting that,
like the Ad5 fiber knob, the Ad12 knob is trimeric. To confirm this
result, a sample of Ad12 knob was examined in the Brookhaven
scanning transmission electron microscope (STEM), which measures
the mass/unit length of macromolecules. In good agreement with the
PAGE results, STEM analysis showed the Ad12 knob has a mass of 60.6
kDa (FIG. 3).
[0081] The Ad12 knob and the refolded D1 and D1/D2 CAR domains were
subjected to gel filtration chromatography to determine their
native sizes (FIG. 4). In all three cases, the proteins eluted as
symmetric peaks in an elution volume that was independent of the
protein concentration (1-500 .mu.M monomer). D1 consistently eluted
as a .sup..about.30 kDa species while both D1/D2 and knob eluted as
.sup..about.60 kDa species. Based upon the primary amino acid
sequences of these proteins, the gel filtration data suggest that
both D1 and D1/D2 are dimers while Ad-12 knob is a trimer. When
D1/D2 and knob were mixed together at equimolar (monomer:monomer)
ratios two peaks were observed in the elution profiles, a low
molecular weight species eluting at a position corresponding to
free Ad12 knob or D1/D2 and a higher molecular weight species
eluting at a molecular mass of 100 kDa. Fractions from the two
peaks were analyzed by SDS-PAGE, which revealed that the high
molecular weight peak corresponded to the knob-D1/D2 complex, while
the lower molecular weight species was free D1/D2. Similar results
were observed for complexes of knob and D1, which eluted at
.sup..about.80 kDa.
Methods of the Invention
Section I
[0082] Expression and purification of Ad12 knob. A DNA fragment
encoding the entire Ad12 fiber knob domain and several flanking
amino acids from the fiber shaft (amino acids 401-587)
(corresponding to nucleotides 30571-31128 of GenBank Accession
#X73487) was amplified from viral DNA by PCR (30 cycles of
94.degree. C./30 sec, 55.degree. C./40 sec, 72.degree. C./60 sec)
using primers #1, CATATGAGCAACACTCCATACG (SEQ ID NO: 2), and #2,
GGATCCTTATTCTTGGGTAATGT (SEQ ID NO: 3), (FIG. 1a). The resulting
PCR product was cloned between the NdeI and BamHI sites of pET15b
(Novagen) and transformed into strain BL21-DE3 (Novagen) for
protein expression. Overnight cultures grown in LB broth containing
150 mg/L penicillin G (Sigma) were diluted 100-fold into fresh
LB-penicillin broth and grown at 37.degree. C. until midlog phase
(OD of 0.8 at 600 nm) at which time they were chilled to 24.degree.
C. and adjusted to 50 .mu.M IPTG (isopropyl .beta.-D
thiogalactopyranoside) to induce knob expression. After shaking
(250 rpm) overnight at 24.degree. C., the cells were collected by
centrifugation, resuspended in 10% of the original culture volume
of STE (10 mM Tris-HCl pH 8.0, 100 mM NaCl, 1 mM EDTA
(ethylenediaminetetraaceti- c acid)) containing 100 .mu.g/ml
lysozyme, and subjected to 3 cycles of freezing and thawing. The
viscous cell lysate was then sonicated and cleared by
centrifugation at 25,000.times.g for 10 min. Knob was precipitated
from the supernatant by addition of solid ammonium sulfate to 35%
saturation (25.degree. C.), dialyzed against several changes of 10
mM Tris-HCl (pH 7.5) and passed over a column of DEAE-cellulose
(DE52, Whatman) equilibrated in the same buffer. Knob was recovered
from the flow-through fractions essentially free of contaminating
E. coli proteins and nucleic acids, and was further purified by
Ni-NTA affinity chromatography according to the manufacturer's
instructions (Qiagen). About 100 mg of purified Ad12 knob was
obtained from one liter of bacterial culture.
[0083] Expression and purification of CAR protein fragments. cDNA
fragments encoding the human CAR extracellular domains (D1 and
D1/D2, FIG. 1b) were amplified by RT-PCR of total RNA from a mouse
A9 cell line transformed with multiple copies of the cloned human
CAR gene. First strand cDNA synthesis was primed by oligo-dT.
Primers #3: CCATGGGTATCACTACTCCTGAAGAGA (SEQ ID NO: 4) (the first 6
nucleotides add two upstream codons, encoding M20 and G21. The
remaining nucleotides correspond to nucleotides 121-141 of GenBank
Accession #Y07593), #4: CTCGAGCGCACCTGAAGGCTTA (SEQ ID NO: 5)
(complementary to GenBank Accession #Y07593 nucleotides 476-491)
#5: CTCGAGTGAAGGAGGGACAAC (SEQ ID NO: 6) (complementary to GenBank
Accession #Y07593 nucleotides 744-758) (FIG. 1b) were designed for
cloning D1- and D1/D2-encoding PCR products between the NcoI and
XhoI sites of expression vector pET20b (Novagen). The PCR cycling
program was identical to that used for Ad12 knob. These same PCR
products were also cloned into pET15b as NcoI-XhoI restriction
fragments, and thus lacked the vector-encoded hexahistidine tag,
and each had an additional 22 amino acid-long carboxy-terminal
extension encoded by vector sequences downstream of the XhoI site
(FIG. 1c). Another set of primers (#6-#8, FIG. 1b) was designed to
adapt the CAR PCR products for cloning between the pET15b NdeI and
BamHI restriction sites, which provides for attachment of the
amino-terminal hexahistidine tag to the expressed proteins. Primer
#6: CATATGGGTATCACTACTC (SEQ ID NO: 7)(the first 7 nucleotides add
two upstream codons, encoding M20 and G21. The remaining
nucleotides correspond to nucleotides 121-132 of GenBank Accession
#Y07593), #7: GGATCCTACGCACCTGAAGGCT (SEQ ID NO: 8) (complementary
to nucleotides 478-493 of GenBank Accession #Y07593) and #8:
GGATCCTATCCAGCTTTATTTGAAG (SEQ ID NO: 9)(complementary to
nucleotides 754-770 of GenBank Accession #Y07593). Stop codons were
built into the reverse primers to avoid synthesis of the CAR
fragments with the vector-encoded carboxy-terminal extensions.
[0084] The procedure used for expression of the initial pET15b-D1
construct (PCR product from primers #3 and #4) was similar to that
described above for Ad12 knob except that the culture was induced
at 18.degree. C. Soluble D1 (sD1), produced from the pET15b
NcoI/XhoI construct, was precipitated from cleared cell lysates by
ammonium sulfate precipitation (35-60% cut, 25.degree. C.) and was
partially purified by anion exchange chromatography (DE52) in 10 mM
Tris-HCl buffer (pH 7.5). About 5 mg of partially purified sD1 was
recovered from one liter of bacterial culture. The
hexahistidine-tagged CAR fragments expressed from the second set of
pET15b constructs (using primers #6-#8, FIG. 1b) were insoluble,
but were recovered from inclusion bodies. Cultures were induced at
37.degree. C., and cleared lysates were prepared as described
above. After centrifugation, the supernatant was discarded, the
pellet was washed several times in STE containing 0.1% NP40,
dissolved in 8 M urea/50 mM .beta.-mercaptoethanol/50 mM Tris-HCl
(pH 9.2) (20 ml per liter of initial culture), and then diluted
with 15 volumes of 20 mM Tris-HCl (pH 8.0). The slightly hazy
solution was passed through a 10 ml bed volume of DEAE-Sepharose
Fast Flow (Pharmacia) equilibrated in 20 mM Tris-HCl (pH 8.0).
Approximately half of the bound CAR fragments eluted with 50 mM
NaCl and were essentially pure. The remaining bound CAR eluted with
300 mM NaCl along with contaminating E. coli proteins, and was
discarded.
[0085] Export of CAR D1 and D1/D2 into the E. coli periplasm.
Mid-log phase cultures of strain BL21-DE3 cells transformed with
pET20b-D1 and pET20b-D1/D2 were treated with IPTG to induce
synthesis of D1 and D1/D2. After 3 hr of induction, whole cell
lysates were prepared and analyzed by SDS-PAGE.
[0086] D1 and D1/D2 expression and solubility in the E. coli
cytoplasm. BL21-DE3 cells transformed with pET15b-D1 and
pET15b-D1/D2 (PCR products from reactions with primers 3-5, FIG.
1b) were induced with IPTG at 18.degree. C. Protein content of
whole cell lysates and of the soluble and insoluble fractions of
cell sonicates were analyzed by SDS-PAGE.
[0087] Assays for detection of knob-CAR complexes. The
hexahistidine tag was cleaved from Ad12 knob using biotinylated
thrombin and was then passed through Ni-NTA and avidin columns to
remove residual his-tagged proteins and thrombin. The resulting
knob was mixed with purified Ad2 hexon protein and then divided
into 3 equal samples. His-tagged D1 or D1/D2 were then added to 2
of the samples, and an equivalent volume of buffer added to the
third (control) sample. Each sample was then batch-adsorbed to
Ni-NTA beads, washed, and eluted with 100 mM EDTA/25 mM Tris-HCl
(pH 8.0). Samples were then electrophoresed in SDS-polyacrylamide
gels and stained with coomassie blue.
[0088] Inhibition of Ad2 infection of HeLa cells. HeLa monolayer
cultures were grown in 50% Dulbecco's modified Eagle medium (DMEM,
Gibco)/50% Ham's F12 Nutrient Mixture (Gibco) containing 10% calf
serum. Monolayers were seeded in 24-well cluster plates 1 day
before infection. Ad2 virus diluted in binding buffer (50% DMEM,
50% PBS, 0.4% bovine serum albumin) was divided into 3 equal
samples and mixed with an equal volume of Ad12 knob, sD1 (both
approximately 2 mg/ml in PBS) or an equal volume of binding buffer.
Each preparation was adsorbed in triplicate (0.2 ml/well) for 30
min at 4.degree. C., the wells were then washed twice with PBS and
incubated for 2 days at 37.degree. C. in DMEM containing 2% calf
serum. The number of infected cells in each culture was then
determined by immunoassay for the viral hexon antigen as previously
described (Bai et al., J. Virol. 67: 5198-5205 (1993)). To control
for possible cytotoxic effects of the recombinant proteins,
additional sets of cultures were pre-incubated with Ad12 knob or
sD1 (1 mg/ml) in binding buffer for 30 min, washed twice with PBS
and then infected with Ad2 virus.
[0089] Analysis of Ad12 knob by scanning transmission electron
microscopy (STEM). The mass of Ad12 knob (with the His tag removed)
was measured by STEM. Five microliters of the purified protein
(.sup..about.10 mG/ml) was applied to an electron microscope holey
grid covered with thin (.sup..about.2 nm) carbon, and after 1
minute was wicked and washed 10 times with 20 mM ammonium acetate.
The grid was blotted and rapidly frozen in liquid nitrogen slush,
then freeze-dried overnight. Data was collected with the Brookhaven
NIH Biotechnology Resource STEM (Wall, J. S. (1982) in Introduction
to Analytical Electron Microscopy, Plenum, New York, p 333-342) at
scans of 0.512 micron width with a dose of 200 electrons/nm.sup.2
Protein particle masses were measured (Wall et al., Annu. Rev.
Biophys. Biophys. Chem. 15: 355 (1986)) off-line using the
"PC-Mass" program, and statistics and curve fitting were generated
with SigmaPlot. Mass calibration was done using TMV particles
adhered to the grid before the sample was applied.
[0090] Gel filtration analysis of Ad12 knob and CAR D1 and D1/D2.
The native molecular masses of Ad12 knob, the refolded D1 and D1/D2
domains of CAR, and knob complexed to D1 or D1/D2 were determined
by size exclusion chromatography using a Superose 6 gel permeation
column. In brief, 20 .mu.L aliquots of purified proteins or protein
complexes were chromatographed at 0.25 ml/min on the Superose 6
column in 20 mM Tris.cndot.HCl pH 7.8, 200 mM NaCl, 1 mM DTT and
0.1 mM EDTA. Aliquots of the fractions were analyzed by SDS-PAGE.
These experiments were run over a range of concentrations from
1-500 .mu.M monomer.
Section II
Crystal Structure of Adenovirus Fiber Knob Domain and CARD1
[0091] The structure of Ad serotype 12 (Ad12) fiber knob domain
(knob) (fiber protein residues 401-587) alone, is reported here at
2.6 .ANG. resolution (Table 1, FIG. 5a). The structure is
essentially identical to that of the previously determined Ad
serotype 5 (Ad5) knob (RMS deviation on equivalent C.sub..alpha.
positions=0.8 .ANG.) (Xia et al., Structure 2: 1259-70 (1994)).
Knob monomers adopt an eight-stranded antiparallel .beta.-sandwich
fold with strands J, C, B, and A comprising the so-called V-sheet
that provides the majority of contacts in the tightly packed trimer
interface, and strands G, H, D and I form the solvent exposed
R-sheet. The one notable difference between the structures of the
Ad5 and Ad12 knobs is the well-ordered HI loop in Ad12 (residues
548-556), which in Ads (residues 536-549) is five residues longer
and disordered.
[0092] The location of the CAR binding site on Ad12 knob was
determined experimentally by solving the structure of the Ad12 knob
CARD1 complex (Table 1). In this complex, CARD1 (preCAR residues
22-144)) has an Igv-like .beta.-sandwich fold (FIG. 5b), and binds
at the interface between two adjacent Ad12 knob monomers producing
a triskelion shaped complex (FIG. 5c) without any significant
rearrangement in the Ad12 knob structure. The complex is formed by
the AB loop, the carboxyl ends of the DE loop, and the very short F
strand of one knob monomer and the FG loop of an adjacent knob
monomer interacting with a single face of the CARD1 sandwich
(strands C, C', C", and the second half of F). The amino-terminal
end of the knob molecule, which is attached to the viral fiber
shaft domain in vivo, is on the opposite face to the carboxyl
terminus of CARD1 (FIG. 5d). Although the current model does not
provide direct information concerning the location of CARD2
(residues 126-222) comparisons of CARD1 with structures of
homologous proteins, solved with both D1 and D2 domains (CD4:
1cdh.pdb, ICAM-1: 1iam.pdb & licl.pdb, and ICAM2: 1zxq.pdb,
where the names of the molecules are followed by their Protein Data
Bank access identification numbers), suggest that CARD2 will not
make extensive contacts with knob. This observation is consistent
with the biochemical data presented in Section I above, indicating
that D1 alone is necessary and sufficient for knob binding. The
high efficiency of Adenovirus infection may be due, in part, to the
three CAR binding sites on each fiber knob. Thus, each virion
contains 36 high-affinity CAR binding sites ensuring that most
collisions with the cell surface will result in viral attachment,
while at least 12 and possibly as many as 36 antibodies are
required for neutralization of all the CAR binding sites on each
virion.
[0093] Based on the crystal structure of Ad5 knob (Xia et al.,
Structure 2: 1259-70 (1994)) two putative CAR binding sites were
proposed, the first at the trimer interface and the second on the
surface of each of the R-sheets. In addition, viruses with
CAR-binding knobs have been suggested to interact with other cell
surface proteins including class I MHC molecules (Hong et al., EMBO
J 16: 2294-306 (1997)). The knob-CAR-binding site proposed here is
strongly supported by mutational analysis despite these other
proposed virus-receptor-binding schemes. Indeed, when a set of
single site substitutions in the R-sheet, based on one proposed
receptor binding site was made (Xia et al., Structure 2: 1259-70
(1994)), the substitutions did not abrogate CARD1 binding (Table
2). The loss of CARD1 binding activity in the HI loop mutant (Table
2) is thought to result from loss of water-mediated hydrogen bonds
between the N and O of G550 in the HI loop and the O of R518 and N
of A521 in the FG loop which normally stabilize the FG loop in a
conformation where it can make direct contacts with CARD1.
[0094] The results presented here also indicate that relying on
mutational analysis in the absence of a structure to identify the
receptor binding sites on knob can prove misleading, because most
of the trimeric knob surface which is accessible to receptors is
composed of loops whose conformations may be altered by long range
effects.
[0095] In the Ad12 knob trimer and all other CAR-binding serotypes,
there are two sets of solvent exposed regions whose sequences are
conserved. The first is located on the virion-facing surface where
the shaft domain begins and the second is located at the side of
the molecule on a ridge between adjacent monomers (FIG. 6a). This
ridge is comprised primarily of residues from the relatively
conserved AB loop that becomes buried in the complex (FIG. 6b). In
total, complex formation buries 1880 .ANG..sup.2 of mixed
hydrophobic and hydrophilic surfaces at each knob-CAR interface,
970 .ANG..sup.2 contributed from knob and 910 .ANG..sup.2 from
CARD1. The majority of the buried surface in knob is contributed by
one monomer (770 .ANG..sup.2) with the second making a relatively
small contribution (200 .ANG..sup.2) The actual surface area
involved in protein-protein contacts across this interface is
.sup..about.15% smaller due to a mismatch in their surface
topology. The contact residues in each molecule draw out a
serpentine which, when united with its partner, create two adjacent
cavities, discussed below.
[0096] Fourteen residues in knob and sixteen residues in CARD1 make
direct interactions across the interface (FIG. 7a) including 7
hydrogen bonds. The AB loop of knob is a key anchor, contributing
over 50% of interfacial interactions, including the three hydrogen
bonds between conserved residues (D415O/K104N.zeta., L426O/Y64OH,
and K429N.zeta./E37O.epsilon.2 of knob/CARD1, respectively). The AB
loop spans the width of the CARD1, held at one end by D406 of knob
and at the other end by E416 of knob, which changes rotomer
conformation upon CARD1 binding to accommodate the approaching side
chain of Y61 (FIG. 7b). In the middle of the AB loop, the conserved
residue P418 of knob makes contacts with residues E37, V51 &
L54 in CARD1. The importance of this loop is emphasized when
comparing the sequence of non-CAR binding Ad serotypes such as
subgroup B (serotypes 3 and 7), which have evolved to bind a
different receptor (Roelvink et al., J Virol 72: 7909-15 (1998)),
or serotypes 40 and 41 where two types of fiber exist on the same
virus, only one fiber type binds CAR (Yeh et al., Virus Res 33:
179-98 (1994)). The sequences of the AB-loop of knob in the non-CAR
binding serotypes diverge widely from each other and from known CAR
binding serotypes (FIG. 7c). Specifically, the knob domains in
non-CAR binding fibers have either insertions or deletions in the
AB loop relative to the conserved residues, P418 & N419.
[0097] In order to further explore the idea that the AB loop is an
important determinant in CAR binding, a number of knob variants
that mimic non-CAR-binding knobs in this region were constructed
The double substitution P417E/P418A which converts the Ad12 AB loop
to a sequence similar to Ad3, the insertion of TI between S421 and
L422, which lengthens the AB loop and the deletion E425 & L426,
which shortens the AB loop, all abrogated CARD1 binding (Table 2).
These mutants are consistent with the hypothesis that much of the
selectivity of adenovirus serotypes for CAR involves interactions
with the AB loop.
[0098] Topological mismatches over the AB loop and at other
discrete regions of the interface result in the formation of two
adjacent approximately equal-sized cavities totaling
.sup..about.120 .ANG..sup.3 (FIG. 7b). They are separated by the
interaction of P409 with residues in CARD1, pinching off
communication between them. The cavities are lined with a mixture
of hydrophobic and polar groups, more than 60% of which are either
part of the backbone or are conserved in sequence. The two
interfacial cavities are estimated to accommodate a total of
.sup..about.4 water molecules, although in the current structural
model only one well-ordered water molecule was observed. This water
molecule forms a bridging interaction with the conserved/backbone
atoms of E37O.epsilon.1, K102N.zeta. in CAR and D415O, P416O and
K429N.zeta. in Ad12 knob. Weaker electron density was also observed
throughout the two cavities, which may be indicative of additional
mobile water molecules not sufficiently ordered at this resolution
to be included in the current model.
[0099] Since the presence of such interfacial cavities in
protein-protein complexes is atypical, it is striking that similar
cavities exist at the interface between HIV gp120 and CD4 (Kwong et
al., Nature 393: 648-659 (1998)), currently the only other
virion-protein receptor complex whose structure has been determined
by crystallography. In addition, the adhesion proteins of both
viruses use surface loops to interact with the C-C'-C" face of
respective receptors which both belong to the IgV superfamily.
Moreover, in both complexes the viral proteins contribute
approximately the same amount of surface area to the interface,
burying a similar ratio of conserved and non-conserved
residues.
[0100] These results, and those of others, now suggest that viruses
have developed at least two structural means of evading immune
attack. In picornoviruses, the receptor binding sites, including by
extension the CAR binding site on coxsackievirus B, are associated
with deep crevices or canyons on the capsid surface (Muckelbauer et
al., Structure 3: 653-667 (1995)), which have been suggested to act
as an antigenic shield for the conserved residues that define
receptor binding specificity (Rossmann, M. G., J.B.C. 264:
14587-14590 (1989)). By contrast, the structures of the Ad12
knob-CARD1 and the HIV gp120-CD4 (Kwong et al., Nature 393: 648-659
(1998)) complexes show that solvent exposed loops comprise their
receptor binding faces thus exposing them to immunoselective
pressure. Water molecules that become trapped within interfacial
cavities in both virus systems mediate specific bridging hydrogen
bonds and van der Waals contacts across the interface. In contrast
to direct amino acid contacts across the interface, which are very
sensitive to changes in sequence, these indirect water-mediated
contacts may be able to tolerate a higher degree of antigenic drift
while still maintaining receptor-binding specificity.
[0101] The structure coordinates of the Ad12knob:CARD1 complex are
deposited in the Protein Data bank (Abola et al., in
Crystallographic Databases, Information Content, Software Systems,
Scientific Applications (eds. Allen, F. H., Bergerhoff, G. &
Sievers, R.) 107-132 (Data Commission of the International Union of
Crystallography, Bonn/Cambridge/Chester, 1987) under access
identification number 1KAC. The structure coordinates of isolated
Ad12knob are deposited in the same database under access
identification number 1NOB.
Methods of the Invention
Section II
[0102] Protein expression, purification and crystallization. The
knob fiber protein (Ad12 knob) and the N-terminal fragment
(residues 22-125) of the cellular receptor (CARD1) were expressed
in E. coli and purified as described previously (Freimuth et al.,
J. Virol. 73: 1392-8 (1999)). Prior to crystallization, the
purified proteins were proteolysed with 10 .mu.g/ml trypsin, the
1:3 (trimeric knob: CARD1) complex formed and purified by anion
exchange chromatography. Crystals of Ad12 were grown at room
temperature using the sitting drop vapor diffusion method at room
temperature from a protein solution of 20 mg/ml suspended over a
reservoir of 26% PEG3350. Showers of small poorly ordered crystals
grew over the course of a week which were harvested, washed in 30%
PEG3350 and seeded into a drop containing equal volumes of protein
and 26% PEG3350 over a reservoir of 26% PEG3350. Large, rhomohedral
plates grew overnight. They were transferred into a solution
containing 50% PEG3350 and cooled directly into a stream of
nitrogen at 99K. Crystals of the Ad12 knob/CARD1 complex were grown
at room temperature using the sitting drop vapor diffusion method
from a protein solution of 10-12 mg/ml and a reservoir of 0.9 M
ammonium sulfate in 100 mM MES (pH 6.2). A typical crystal had a
cubic habit and grew to 1.0 mm over a period of .sup..about.10
days. They were cooled directly into a stream of nitrogen at 99K,
using 50% ethylene glycol as a cryoprotectant.
[0103] Structure Determination and refinement. The crystal
structure of the knob-CARD1 complex was determined using crystals
which had P4.sub.332 space group symmetry. The unit cell was cubic
(all sides of equal length) with 167.85 angstroms per side.
[0104] Each data set was collected from a single crystal at 99K
using the NSLS beamlines X8C, X12C, and X25 at Brookhaven National
Laboratory, Upton, N.Y. with either a Mar345 imaging plate
detector, or the Brandeis 4-cell CCD detector. Mercury was
introduced into Knob-CARD1 complex by soaking a single CARD1-knob
complex crystal in 10 mM thimerosal for 6 hours. The data was
collected on X12C using the Brandeis 4-cell CCD at a wavelength of
1.0 .ANG.. In all cases, the crystals were cooled using an Oxford
Cryostream. Raw data were reduced and scaled using the HKL program
Suite (Otwinowski, Z. & Minor, W. in Methods in Enzymology
(eds. Carter, C. W. & Sweet, R. M.) 307-326 (Academic Press,
1997)). All further calculations leading to the structure solutions
were performed using programs in the CCP4 program Suite (CCP4. The
SRC (UK) Collaborative Computing Project No. 4: A Suite of Programs
for Protein Crystallography. (Daresbury, UK., 1991)).
[0105] The structure of Ad12 knob was solved by molecular
replacement, using a monomer of Ad5 knob (1KNB.PDB) as a search
model. The Ad12 knob is 48% identical and 78% similar in sequence
to the Ad5 knob which also binds CAR (Bergelson et al., Science
275: 1320-3 (1997)). Two families of peaks were found in the cross
rotation function, relating to two trimers. Based on an estimation
of the Matthews coefficient (Matthews, B. W., J. Mol. Biol. 33:
491-497 (1968)) the asymmetric unit contained 1 or 2 trimers. The
asymmetric unit contained six molecules based on the increasing
correlation coefficient above the level of noise. Attempts to
position of a seventh molecule reduced the correlation coefficient
and we took this as evidence that the molecule contained 6 and not
3 monomers in the asymmetric unit. The structure was refined using
the rigid body and simulated annealing protocols in CNS (Brunger et
al., Acta Crystallogr. D. Biol. Crystallogr. 54: 905-21 (1998))
using tight NCS restraints, punctuated by rounds of model building.
The current model contains residues 394-578 and 47 water molecules
with good geometry (RMS deviation on bond lengths and bond angles
is 0.009 .ANG. and 1.6.degree. respectively). It has an R-factor of
23.9% with a corresponding R.sub.free=29.4%.
[0106] The structure of Ad12 knob-CARD1 complex was determined
using a combination of SIR, solvent flattening and molecular
replacement. The refined structure of the Ad12 knob monomer
described above was used as a search model in molecular
replacement. A single clear solution was found corresponding to a
single monomer in the asymmetric unit close to the crystallographic
three fold axes. Positioned in this way, the biological trimer was
generated by the crystallographic 3-fold of the unit cell. At this
point, no interpretable density corresponding to the CAR-D1
molecule was visible, therefore, a thimerosal derivative data set
was collected. The position of the mercury was identified in a
difference map, calculated between the derivative data set and the
current model. This map revealed the location of a mercury atom at
a single site, close to Cys433. The position of this mercury was
refined using the program MLPHARE to provide phase information for
the crystal to 3.6 .ANG.. The initial phases set had a FOM of 0.24
and provided useful phases to 3.6 .ANG. resolution. Phase
combination using the structure of Ad12 knob and the experimental
SIR phases followed by solvent flattening using DM resulted in a
map with a FOM of 0.74 to 2.6 .ANG. resolution. Continuous density
corresponding to a single domain of CAR was clearly visible in this
map. A polyalanine model was built into the density, followed by a
cycle of simulated annealing to 3000 K using the program CNS. The
resultant SIGMAA-weighted map allowed the side chains to be built
unambiguously. Subsequent rounds of model rebuilding and refinement
reduced the R factor to 22.5 (R.sub.free 24.9%). The current model
contains residues 394-578 of the Knob molecule, residues 22-144 of
the CAR-D1 and 70 water molecules, with good overall geometry (RMS
deviation on bond lengths and bond angles is 0.010 .ANG. and
1.7.degree. respectively). Based on an interrogation of the DALI
database (Holm et al., Science 273: 595-603 (1996)), the structure
of CARD1 most closely resembles that of the extracellular domain of
the myelin adhesion molecule (Shapiro et al., Neuron. 17: 435-449
(1996)), followed by domain 1 of human CD4, a receptor for HIV (Ryu
et al., Structure 2: 59-74 (1994)), and several other cell surface
glycoproteins. Although all of these molecules share a common fold,
there are large differences in strand lengths and loop
conformations when equivalent atoms are superimposed.
[0107] Mutational Analysis. The Ad12 knob mutants listed in Table 2
were constructed by primer-directed PCR mutagenesis and confirmed
by nucleotide sequence analysis. Mutant knob proteins were purified
as described previously (Freimuth et al., J. Virol. 73: 1392-8
(1999)), and were then immobilized on nitrocellulose membranes and
incubated with biotinylated CARD1 to examine the effect of
mutations on knob CARD1-binding activity. Purified mutant or
wild-type his-tagged knob proteins were bound to nitrocellulose
membranes using a dot-blot manifold (5 .mu.g per dot). The membrane
was then fixed in 0.25% glutaraldehyde in phosphate-buffered saline
(PBS), blocked in 5% milk-PBS, and then incubated sequentially with
5 .mu.g/ml biotinylated CARD1 protein, 1:500 horse-radish
peroxidase (HRP)-conjugated mouse-anti-biotin monoclonal antibody
(Sigma Chemical Co.) (both reagents diluted in 0.5% milk-PBS), and
finally with chemiluminescent substrate (SuperSignal, Pierce), with
several PBS washes between each incubation. Chemiluminescence was
detected on x-ray film, and relative signal intensities were
visually assessed. CARD1 was biotinylated with sulfo-NHS-LC biotin
(Pierce) according to the protocol recommended by the manufacturer.
A duplicate membrane containing the same set of knob proteins was
incubated with rabbit anti-Ad12 knob serum and then
HRP-goat-anti-rabbit IgG (Cappel) followed by chemiluminescent
detection to demonstrate that equal amounts of knob protein were
present in each dot. All proteins were assayed in duplicate, and
the same results were obtained in 3 successive experiments.
1TABLE 1 Summary of data collection statistics Knob + CAR Knob Knob
+ CARD1 Thimerosal Resolution (.ANG.) 30-2.6 30-2.6 30-3.4 R merge
(%) 10.0 (22.4) 7.0 (34.6) 9.8 (23.6) Completeness (%) 100.0 (100)
100.0 (100) 99.6 (99.7) I/.sigma.I 10 (3) 20 (6) 12 (5) Phasing
Power 1.1
[0108]
2TABLE 2 The effect of structure-based substitutions on CARD1
binding. SUBSTITUTION POSITION IN KNOB BIND TO CARD1 L430A B
strand, R-Sheet ++ V466Y D strand, R-Sheet +++ H467K D strand,
R-Sheet +++ V469E D strand, R-Sheet +++ T543R H strand, R-Sheet +++
K545E H strand, R-Sheet +++ K545M H strand, R-Sheet +++ T560R I
strand, R-Sheet +++ S564R I strand, R-Sheet +++ P417E/P418A AB loop
- Ins TI 421 AB loop - .DELTA.E425/L426 AB loop - E425S AB loop ++
.DELTA.G550/I551 HI loop +
[0109]
Sequence CWU 1
1
16 1 22 PRT Artificial Sequence Description of Artificial Sequence
Cloning vector encoded sequences 1 Leu Glu Asp Pro Ala Ala Asn Lys
Ala Arg Lys Glu Ala Glu Leu Ala 1 5 10 15 Ala Ala Thr Ala Glu Gln
20 2 22 DNA Artificial Sequence Description of Artificial Sequence
Cloning Primer 2 catatgagca acactccata cg 22 3 23 DNA Artificial
Sequence Description of Artificial Sequence Cloning Primer 3
ggatccttat tcttgggtaa tgt 23 4 27 DNA Artificial Sequence
Description of Artificial Sequence Cloning Primer 4 ccatgggtat
cactactcct gaagaga 27 5 22 DNA Artificial Sequence Description of
Artificial Sequence Cloning Primer 5 ctcgagcgca cctgaaggct ta 22 6
21 DNA Artificial Sequence Description of Artificial Sequence
Cloning Primer 6 ctcgagtgaa ggagggacaa c 21 7 19 DNA Artificial
Sequence Description of Artificial Sequence Cloning Primer 7
catatgggta tcactactc 19 8 22 DNA Artificial Sequence Description of
Artificial Sequence Cloning Primer 8 ggatcctacg cacctgaagg ct 22 9
25 DNA Artificial Sequence Description of Artificial Sequence
Cloning Primer 9 ggatcctatc cagctttatt tgaag 25 10 17 PRT
Adenovirus Type 12 10 Thr Leu Trp Thr Thr Pro Asp Pro Pro Pro Asn
Cys Ser Leu Ile Gln 1 5 10 15 Glu 11 17 PRT Adenovirus Type 5 11
Thr Leu Trp Thr Thr Pro Ala Pro Ser Pro Asn Cys Arg Ile His Ser 1 5
10 15 Asp 12 17 PRT Adenovirus Type 9 12 Thr Leu Trp Thr Thr Pro
Asp Thr Ser Pro Asn Cys Lys Ile Asp Gln 1 5 10 15 Asp 13 17 PRT
Adenovirus Type 4 13 Thr Leu Trp Thr Thr Pro Asp Pro Ser Pro Asn
Cys Gln Ile Leu Ala 1 5 10 15 Glu 14 17 PRT Adenovirus Type 40 Long
14 Thr Leu Trp Thr Thr Ala Asp Pro Ser Pro Asn Ala Thr Phe Tyr Glu
1 5 10 15 Ser 15 16 PRT Adenovirus Type 40 Short 15 Thr Ile Trp Ser
Ile Ser Pro Thr Pro Asn Cys Ser Ile Tyr Glu Thr 1 5 10 15 16 18 PRT
Adenovirus Type 3 16 Thr Leu Trp Thr Gly Val Asn Pro Thr Thr Ala
Asn Cys Ile Ile Glu 1 5 10 15 Tyr Gly
* * * * *