Note: Descriptions are shown in the official language in which they were submitted.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
STREPTAVIDIN-BINDING PEPTIDES AND USES THEREOF
Statement as to Federallx Sponsored Research
This invention was funded by grant number R01GM53936 from the
National Institutes of Health and grant number NCC-2-1069 from NASA. The
government may have certain rights in the invention.
Background of the Invention
In general, the invention features novel compounds and methods for
purifying or detecting proteins of interest.
Determining the enzymatic activity, binding specificity, or three-
dimensional structure of a protein often requires the purification of the
protein
from a complex mixture of other components, such as compounds present in a
cell
lysate or iT2 vitro translation extract. With the number of known proteins
increasing dramatically as a result of whole genome sequencing projects, it
has
become crucial to find alternatives to traditional, time-consuming monoclonal
antibody production for generating affinity reagents for the detection and
purification of proteins. In addition, purifying a novel protein using
traditional
column chromatography methods often requires much trial and error to develop a
purification protocol that results in the recovery of the protein in high
yield and
purity.
Thus, purification methods are needed that may be generally applied to
proteins of interest, that utilize inexpensive reagents, and that result in
highly
purified protein without requiring multiple chromatography steps.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
2
Summary of the Invention
The purpose of the present invention is to provide improved reagents for
the purification, detection, or quantitation of proteins of interest. In
particular, the
high affinity, streptavidin-binding peptides of the present invention may be
used
as affinity tags for the purification of fusion proteins containing proteins
of
interest.
Accordingly, in a first aspect, the invention provides a peptide which binds
streptavidin with a dissociation constant less than 10 ~.M (that is, binds
streptavidin more tightly than a I~ of 10 ~,M) and which is not disulfide
bonded
or cyclized. Preferably, the dissociation constant is equal to or less than 5
,uM, 1
~,M, 100 nM, 50 nM, 25 nM, 10 nM, or even 5 nM. In one preferred
embodiment, the dissociation constant is less than 10 ~,M, 5 ~,M, 1 ~.M, 100
nM,
50 nM, or 25 nM; and greater than 0.01 nM, 0.1 nM, 1 nM, 5 nM, or 10 nM. In
another preferred embodiment, the value of the dissociation constant is
contained
in one of the following ranges: 5 ,uM to 1 ,uM, 1 ,uM to 100 nM, 100 nM to 50
nM, 50 nM to 25 nM, 25 nM to 10 nM, 10 nM to 5 nM, 5 nM to 1 nM, or 5 nM to
0.1 nM, inclusive.
In a related aspect, the invention provides a peptide which binds
streptavidin with a dissociation constant less than 10 ~,M. The amino acid
sequence of the peptide does not contain an HPQ, HPM, HPN, or HQP motif.
Preferably, the dissociation constant is equal to or less than 5 ~,M, 1 ~.M,
100 nM,
50 nM, 25 nM, 10 nM, or 5 nM. In one preferred embodiment, the dissociation
constant is less than 10 ~,M, 5 ,uM, 1 ~,M, 100 nM, 50 nM, or 25 nlV.~ and
greater
than 0.01 nM, 0.1 nM, 1 nM, 5 nM, or 10 nM. In another preferred embodiment,
the value of the dissociation constant is contained in one of the following
ranges:
5 ~,M to 1 ~.M, 1 ,uM to 100 nM, 100 nM to 50 nM, 50 nM to 25 nM, 25 nM to
10 nM, 10 nM to 5 nM, 5 nM to 1 nM, or 5 nM to 0.1 nM, inclusive.
In another related aspect, the invention provides a peptide which binds
streptavidin with a dissociation constant less than 23 nM, 10 nM, or
5 nM. In one preferred embodiment, the peptide is disulfide bonded or
cyclized.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
In another preferred embodiment, the dissociation constant is less than 23 nM,
10
nM, or 5 nM; and greater than 0.01 nM, 0.1 nM, or 1 nM. In another preferred
embodiment, the value of the dissociation constant is contained in one of the
following ranges: 20 nM to 10 nM, 10 nM to 5 nM, 5 nM to 1 nM, or 5 nM to
0.1 nM, inclusive.
In other related aspects, the invention provides nucleic acids encoding the
peptides of the present invention, and vectors that include such nucleic
acids.
In addition, standard gene'fusion techniques may be used to generate
fusion nucleic acids that encode fusion proteins which include a peptide of
the
present invention and a protein of interest. The fusion proteins may be
purified,
detected, or quantified based on the high affinity of the peptides for
streptavidin.
Accordingly, in one such aspect, the invention provides a fusion protein
including a protein of interest covalently linked to one of the following
peptides:
(a) a peptide which binds streptavidin with a dissociation constant less than
10
~,M and which is not disulfide bonded or cyclized, (b) a peptide which binds
streptavidin with a dissociation constant less than 10 ,uM and which does not
contain an HPQ, HPM, HPN, or HQP motif, or (c) a peptide which binds
streptavidin with a dissociation constant less than 23 nM. In preferred
embodiments, the peptide is attached to the amino-terminus or the carboxy-
terminus of the protein of interest, or the peptide is positioned between the
amino
and carboxy-termini of the protein of interest. Preferably, the peptide is
linked to
the protein of interest by a linker which includes a protease-sensitive site.
In related aspects, the invention provides nucleic acids encoding the fusion
proteins of the present invention, and vectors that include these fusion
nucleic
acids.
In addition, the invention provides a method of producing a fusion protein
of the present invention. This method includes transfecting a vector having a
nucleic acid sequence encoding the fusion protein into a suitable host cell
and
culturing the host cell under conditions appropriate for expression of the
fusion
protein.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
The fusion proteins described herein may be used in methods for purifying
proteins of interest from samples. Such a method involves expressing the
protein
of interest as a fusion protein covalently linked to one of the following
peptides:
(a) a peptide which binds streptavidin with a dissociation constant less than
10
~,M and which is not disulfide bonded or cyclized, (b) a peptide which binds
streptavidin with a dissociation constant less than 10 p.M and which does not
contain an HPQ, HPM, HPN, or HQP motif, or (c) a peptide which binds
streptavidin with a dissociation constant less than 23 nM. A sample containing
the fusion protein is contacted with streptavidin under conditions that allow
complex formation between the fusion protein and streptavidin. The complex is
isolated, and the fusion protein is recovered from the complex, thereby
purifying
the protein of interest from the sample. In one preferred embodiment, the
protein
of interest is recovered from the fusion protein by cleaving the streptavidin-
binding peptide from the fusion protein.
In yet another aspect, the invention provides a method of detecting the
presence of a fusion protein of the present invention in a sample. This method
includes (a) contacting the sample with streptavidin under conditions that
allow
complex formation between the fusion protein and streptavidin, (b) isolating
the
complex, and (c) detecting the presence of streptavidin in the complex or
following recovery from the complex. The presence of streptavidin indicates
the
presence of the fusion protein in the sample. Preferably, step (c) also
involves
measuring the amount of streptavidin in the complex or following recovery from
the complex. The amount of fusion protein in the sample is correlated with,
and
may be calculated from, the measured amount of streptavidin. For example, for
a
fusion protein containing a peptide that binds one molecule of streptavidin
per
molecule of peptide, the amount of fusion protein in the sample is predicted
to be
approximately the same as the amount of streptavidin measured. In one
preferred
embodiment, the amount of streptavidin is determined using Western or ELISA
analysis with an antibody that reacts with streptavidin or that reacts with a
compound that is covalently linked to streptavidin. In another preferred
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
embodiment, streptavidin is covalently linked to an enzyme, radiolabel,
fluorescent label, or other detectable group, and the amount of streptavidin
is
determined using standard techniques based on a characteristic of the
detectable
group such as its enzyme activity, radioactivity, or fluorescence.
In preferred embodiments of various aspects of the invention, the amino
acid sequence of the peptide includes at least 10, 25, 50, 75, or 100
consecutive
amino acids or consists of between 5 and 150, 10 and 100, 20 and 75, or 30 and
50 amino acids, inclusive, of any one of SEQ ID Nos. 1-29 or 35. Preferably,
the
amino acid sequence of the peptides includes an LPQ, QPQ, EPQ, HPA, HPD, or
HPL motif. In other preferred embodiments, the amino acid sequence includes
any one of SEQ ID Nos. 1-29 or 35. In still other preferred embodiments, the
peptide has an amino acid sequence that is at least 20, 30, 40, 50, 60, 70,
80, 90,
95, or 100% identical to any one of SEQ ID Nos. 1-29 or 35.
It is also contemplated that the affinity of the peptides of the present
invention for streptavidin may be increased by incorporating disulfide bonds
into,
or cyclizing, the peptides. By constraining the peptides, the amount of
disorder
inherent in the peptides (i.e., entropy) decreases, and thus binding of these
peptides to streptavidin may require less energy. It is also contemplated that
the
three-dimensional structure of peptides of the invention bound to streptavidin
may
be experimentally determined or modeled based on the known crystal structure
of
streptavidin and used to determine possible modifications to the peptides that
may
further improve their affinity for streptavidin.
As used herein, by "nucleic acid" is meant a sequence of two or more
covalently bonded naturally-occurring or modified deoxyribonucleotides or
ribonucleotides.
By "peptide" is meant a sequence of two or more covalently bonded
naturally-occurring or modified amino acids. The terms "peptide" and "protein"
are used interchangeably herein.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
By "covalently linked" is meant covalently bonded or connected through a
series of covalent bonds. A group that is covalently linked to a protein may
be
attached to the amino-terminus, carboxy-terminus, between the amino- and
carboxy-termini, or to a side chain of an amino acid in the protein.
By "streptavidin" is meant any streptavidin molecule or fragment thereof
or any protein that has an amino acid sequence that is at least 80, 90, 95, or
100%
identical to a streptavidin molecule or fragment thereof (see, for example,
Haeuptle et al. J. Biol. Chem. 258: 305, 1983). A preferred fragment of
streptavidin is "core" streptavidin, which is a proteolytic cleavage product
of
streptavidin (Bayer et al. Biochem. J. 259,369-376, 1989). Preferably, a
streptavidin molecule or fragment thereof is capable of binding biotin or any
other
streptavidin-binding molecule. Streptavidin or a streptavidin fragment may be
modified chemically or through gene fusion technology or protein synthesis so
that it is covalently linked to an enzyme, radiolabel, fluorescent label, or
other
detectable group. These detectable groups may be used to determine the
presence or location of a streptavidin-bound fusion protein in a cell or
sample or
to quantify~the amount of a streptavidin-bound fusion protein, using standard
methods.
By "cyclized" is meant nonlinear. A peptide may be cyclized by the
formation of a covalent bond between the N-terminal amino group of the peptide
or the side-chain of a residue and the C-terminal carboxyl group or the side-
chain
of a residue. For example, a peptide lactam may be formed by the cyclization
between the N-terminal amino group or an amino group of an amino acid side-
chain and the C-terminal carboxyl group or a carboxyl or amide containing side-
chain. Other possible cyclizations include the formation of a thioether by the
reaction of a thiol group in a cysteine side-chain with the N-terminal amino
group,
C-terminal carboxyl group, or the side-chain of another amino acid. A
disulfide
bond may also be formed between two cysteine residues. As used herein, a "non-
cyclized peptide" is a linear peptide that does not have any of the above
cyclizations.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
By "dissociation constant" is meant the dissociation constant for binding
streptavidin as measured using the electrophoretic mobility shift assay
described
herein. By "less than" a particular dissociation constant is meant capable of
binding streptavidin more tightly than the strength of binding represented by
a
particular dissociation constant.
By "purifying" is meant separating a compound, for example, a protein,
from other components that naturally accompany it. Typically, a protein is
substantially pure when it is at least 50%, by weight, free from proteins and
naturally-occurring organic molecules with which it is naturally associated.
Preferably, the protein is at least 75%, more preferably, at least 90%, and
most
preferably, at least 99%, by weight, pure. In other preferred embodiments, the
protein is at least 2, 5, 10, 25, 50, or 100 times as pure as the starting
material.
Purity may be assayed by any appropriate method, such as polyacrylamide gel
electrophoresis, column chromatography, optical density, HPLC analysis,
western
analysis, or ELISA (see, for example, Ausubel et al., Current Protocols in
Molecular Biology, John Wiley & Sons, New York, 2000).
By "recovered from the complex" is meant physically separated from the
complex of streptavidin and the fusion protein. For example the streptavidin-
bound fusion protein may be incubated under conditions that reduce the
affinity of
the fusion protein for stxeptavidin (i.e., at low or high salt concentrations
or at low
or high pH values) or incubated in the presence of molecules that compete with
the fusion protein for binding streptavidin. Preferably, either the fusion
protein or
the streptavidin that has been released from the complex is isolated using
standard
procedures, such as column chromatography, polyacrylamide gel electrophoresis,
HPLC, or western analysis.
The present invention provides a number of advantages related to the
detection and purification of proteins of interest. For example, because the
present methods do not require the generation of an antibody or other affinity
reagent that is specific for each protein of interest, these methods may be
universally applied to any protein. In addition, if desired, the streptavidin-
binding
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
peptide may be connected to the protein of interest through a protease
cleavable
linker, allowing removal of the peptide after purification of the fusion
protein.
Using the methods described herein, purification of a fusion protein based on
its
affinity for streptavidin has allowed the isolation of the fusion protein in
significantly higher purity than that obtained using a hexahistidine affinity
tag or
maltose-binding protein affinity tag. Moreover, streptavidin is an inexpensive
reagent that may be purchased unmodified or covalently labeled with a
detectable
group (such as FITC-streptavidin or alkaline phosphatase-conjugated
streptavidin)
or with a chromatography matrix (such as streptavidin-agarose). The
availability
of these reagents simplifies methods for detecting and purifying the fusion
proteins of the present invention.
Other features and advantages of the invention will be apparent from the
following detailed description and from the claims.
Brief Description of the Drawings
Figure 1A is a schematic illustration of an ih vitro selection process
according to the invention, showing the structure of the library and the
selection
scheme. Members of the DNA library have, from the 5' to 3' end, a T7 RNA
polymerase promoter (T7), a tobacco mosaic virus translation enhancer (TMV), a
start codon (ATG), 88 random amino acids, a hexahistidine tag (H6), and a 3'
constant region (Const).
Figure 1B is a picture of an SDS-PAGE gel of samples from the library at
different stages of preparation. The first lane shows the result of
translating the
mRNA display template with 35S-methionine. Most of the counts represent free
peptide (free pep), but a significant amount of mRNA-peptide covalent fusions
are
also present (mRNA-pep). There is also another band that is independent of
added
template (NS, non-specific), and some counts remain in the gel well. The band
corresponding to the mRNA-peptide can be shifted to a position slightly higher
than that for the free peptide by the addition of RNase A. The remaining lanes
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
show the result of successive oligo-dT and Ni-NTA purifications, and finally
reverse transcription (RT).
Figure 2A is a bar graph showing the fraction of 35S counts from the
displayed peptides that bound to streptavidin and eluted with biotin, at each
round
of selection. Figure 2B is a graph showing the elution profile for the peptide
library generated from the output of the seventh round'of selection in Figure
2A.
The first fraction represents the flow-through. Biotin was added at the point
indicated. The plot compares the binding of the intact, reverse-transcribed,
displayed peptides (mRNA-pep), the same sample treated with RNase A, and the
RNase-treated sample applied to a streptavidin column pre-saturated with
biotin
(excess biotin was washed away prior to exposing the library to the matrix).
Figure 3 is a list of the sequences of 20 clones from the seventh round of
selection (SEQ )D Nos.: 1-20). The "#" column indicates the number of times
each sequence was observed. The HPQ sequence is in bold type. Non-random
sequences at the termini are underlined. The six C-terminal-most residues are
not
shown.
Figure 4A is a picture of a native gel showing an electrophoretic mobility
shift (EMSA) analysis demonstrating the binding of four different DNA-tagged
peptides to streptavidin. The migration of each clone is shown in the absence
(-)
and presence (+) of 1 ,uM streptavidin. Some of the clones show multiple
bands,
presumably representing different conformations. The arrows show the position
of the gel well, which often contains a fraction of the counts. Figure 4B is a
picture of a native gel showing the titration of the full-length clone SB 19
with
streptavidin. The streptavidin concentration in each lane, from left to right,
is:
3.8, 6.6, 10, 15, 23, 35, and 61 nM. Figure 4C is a curve fit of the data
shown in
Figure 4B (the fraction of peptide bound could not be accurately determined
for
the point with the lowest concentration of streptavidin). Assuming that the
peptide is homogeneous and 100% active, the data from this experiment give a
Kd
of 10 nM for the binding of peptide SB 19 to streptavidin.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
Figure 5 is a list of the sequences of truncation mutants of peptide SB 19
(SEQ ID Nos.: 21-29). The full-length (FL), C-terminal deleted (C1-C4.), N-
terminal deleted (N1-N3), and point mutated (M1) peptide sequences are shown.
The "% binding" refers to the performance of these peptides in the
streptavidin
5 column-binding assay.
Figure 6A is the nucleotide sequence of the plasmid used for expression of
a fusion protein containing a streptavidin-binding peptide (SEQ ID No.: 37).
Figure 6B is the amino acid sequence of the encoded protein (SEQ ID No.: 38)
which contains, from the amino- to carboxy-terminus, maltose-binding protein,
a
10 sheptavidin-binding peptide (SEQ ID No.: 35, Fig. 7A), a hexahistidine tag,
and
another peptide called 2r18-l9dN. Figure 6C is the amino acid sequence of 2r18-
l9dN (SEQ ID No.: 39).
Figure 7A is the amino acid sequence of the streptavidin-binding peptide
(SEQ ID No.: 35) used as an affinity tag for the purification of the fusion
protein
listed in Fig. 6B. This peptide contains the first 38 amino acids of the SB 19-
C4
peptide (Fig. 5). Figure 7B is a picture of an SDS-PAGE gel showing the purity
of the fusion protein after elution from the streptavidin column (lane 2)
compared
to the purity of the E. coli lysate that was applied to the column (lane 1).
Figures 8A-8F are schematic illustrations of the pre-selection method.
Figure 8A is an illustration of an mRNA display template terminating in
puromycin in which the tobacco mosaic virus translation enhancer sequence
(TMV), the initiating methionine codon (AUG), and the sections of the open
reading frame encoding the two protein affinity tags (FLAG and His6) are
labeled.
Figure 8B is an illustration of an mRNA display template that is free of
frameshifts and premature stop codons and thus encodes a full-length protein
containing both affinity tags. Figure 8C illustrates an mRNA display template
that has initiated internally and displays the corresponding truncated protein
lacking the N-terminal FLAG tag. Figure 8D shows an mRNA display template
that has a deletion in its open reading frame and thus displays the
corresponding
frameshifted protein lacking the C-terminal His6 tag. Figure 8E illustrates
the
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
11
reverse transcription of the mRNA display template from Fig. 8B that was
purified based on the presence of both protein affinity tags in the encoded
protein.
Figure 8F shows the cleavage sites for Type IIS restriction enzymes which are
encoded in each cassette. Ligation of pre-selected cassettes which have been
cleaved with these enzymes yields the full-length DNA library.
Figure 9A is the polynucleotide sequence of the vector encoding a fusion
protein containing maltose-binding protein, a streptavidin-binding peptide
(SEQ
ID No.: 35, Fig. 7A), and a hexahistidine tag. Figure 9B is the amino acid
sequence of the encoded fusion protein. The sequence of the streptavidin-
binding
peptide which contains the first 38 amino acids of the SB19-C4 peptide is
underlined.
Figure 10A is a graph of the biacore response units over various lengths of
time for the dissociation of streptavidin from the fusion protein listed in
Fig. 9B
immobilized on a biacore chip. For line "a," the streptavidin concentration is
23
~,M; for line "b," the concentration is 11.5 ,uM, and for line "c," the
concentration
is 5.75 ~.M. This data was used to calculate an upper limit of 2 x 10-3/s for
the
dissociation rate, kd. Figure 10B is a graph showing the association and
subsequent dissociation of streptavidin from the immobilized fusion protein.
For
lines "a" through "f," the streptavidin concentrations are 1.6, 0.8, 0.4, 0.2,
0.1,
and 0.05 ,uM, respectively. This data was used to calculate an association
rate, ka,
of 5 x 104/Mls.
Detailed Description
The present methods stem from the discovery of peptides that have
unusually high affinities for streptavidin (Kd of less than 10 ,uM). 'These
peptides
were selected from a library of randomized, non-constrained peptides using the
mRNA display method. The high affinity of the selected peptides was
particularly surprising, given the fact that non-constrained linear peptide
libraries
generally do not yield high affinity ligands to proteins, except in cases
where the
protein normally functions in peptide recognition (Clackton et al., Trends
Biotech
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
12
12:173-184 (1994); Katz, Annu. Rev. Biophys. Biomol. Struct. 26:27-45, 1997).
Many other peptides with high affinity for streptavidin may be isolated using
the
mRNA display method or any other selection method, such as ribosome display
(Roberts, Curr. Opin. Chem. Biol. 3(3):268-73, 1999), or phage display (U.S.
Patent No. 5,821,047).
The binding characteristics of exemplary selected streptavidin-binding
peptides are described in Table 1, and the sequences of these peptides are
listed in
Fig. 3. The first column of Table 1 lists the peptide name (SB 1 - SB20). For
comparison, a non-selected sequence with two HPQ motifs spaced by 19 residues
(called "non-selected") is listed in row one. SB 19-C4 is a truncation mutant
of
peptide SB 19, described below. The peptides are grouped according to the
number of HPQ and similar tripeptide motifs they possess. The second column
shows the number of tripeptide motifs in each peptide, and the number of amino
acid residues separating them. The third column represents the percentage of
peptide binding and specifically eluting from a streptavidin column. This
percentage ranged from 8.3% to as high as 88% for the selected peptides,
compared to only 0.16% for the control, non-selected peptide with two HPQ
motifs.
The fourth column shows the Kd, when known, for the interaction between
streptavidin and the peptides, as measured in the EMSA assay described herein.
The standard deviation in the Kd is shown in the fifth column, based on the
number of independent measurements (n, shown in parentheses). The
dissociation constant ranged from 110 nM for peptide SB5 to 4.8 nM for
peptides
SB2.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
13
Table 1
% binding Standard
Peptide Structure and elutingK~ (nM) deviation
(n)
Non-selected HPQ 19 HPQ 0.16
Two HPQ motifs
SB 1 HPQ 19 HPQ 86 50 5.7 (4)
SB2 HPQ 19 HPQ 48 4.8 0.91 (8)
SB3 HPQ 23 HPQ 20
SB4 HPQ 43 HPQ 49
SB5 HPQ 52 HPQ 72 110 22 (6)
One HPQ and one similar
tripeptide motif
SB6 HPL 4 HPQ 49
SB7 HPD 7 HPQ 28
SB8 HPQ 12 HPL 27
SB9 HPQ 12 HP 64
SB 10 HPQ 21 QPQ 15
SB 11 HPQ 28 HPA 68
SB 12 HPQ 30 EPQ 73
SB 13 HPQ 32 EPQ 64
SB 14 HPQ 43 HPL 11
SB 15 QPQ 50 HPQ 44 92 16 (4)
SB 16 HPQ 74 LPQ 50
One HPQ motif
SB 17 8.3
SB 18 58
SB19 85 10 1.8 (10)
SB19 - C4 88 4.9 0.88 (10)
No HPQ motif
SB20 HPL 34
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
14
To further characterize the binding of the selected peptides to streptavidin,
truncation mutants for peptide SB 19 were constructed to determine which
regions were necessary for high affinity streptavidin-binding (Fig. 5).
Deletion
of up to 56 residues had no observable effect on the binding strength. For
example, peptide SB19-C4 retained only the first 38 residues from the selected
construct (plus the C-terminal sequence MMSGGCKLG, SEQ ID No.: 36) and
had a dissociation constant of 4.9 nM for streptavidin (Table 1). In contrast,
N-
terminal truncation mutations (N1-N3) resulted in a lower percentage of the
encoded peptide specifically eluting from the streptavidin column (0.058 to
69%
for the truncation mutants compared to 85°70 for full length SB 19).
These results
suggested that the determinants for binding streptavidin were spread
throughout
the N-terminal 38 residues of the SB19 peptide.
High affinity streptavidin-binding peptides, such as those shown in Table
1, have a number of uses. For example, these peptides may be used for protein
purification by expressing a protein of interest as a fusion protein joined to
one or
more of the streptavidin-binding peptides of the invention. In one such
purification method, a sample containing the fusion protein is incubated with
immobilized streptavidin. Proteins with no or weak affinity for streptavidin
are
washed away, and the fusion protein is then selectively eluted from the
streptavidin matrix by addition of biotin, a biotin analog, another
streptavidin-
binding peptide, or any compound that competes with the fusion protein for
binding to the matrix. Alternatively, the fusion protein may be eluted from
the
matrix by increasing or decreasing the pH of the buffer applied to the matrix.
As described in detail below, this general protocol was used in a one-step
purification of a fusion protein containing a streptavidin-binding peptide
from an
E. coli extract, resulting in a high yield of very pure protein. This fusion
protein
contained the first 38 amino acids of the SB 19-C4 peptide, which due to its
small
size was not expected to affect the three-dimensional structure or activity of
the
covalently-linked protein of interest. Purification of fusion proteins
containing
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
other streptavidin-binding peptides of the present invention may be performed
similarly.
In addition, various modifications of the above purification protocol
would be apparent to one skilled in the art (see, for example, Ausubel et al.,
5 supra), and such modifications are included in the invention. In particular,
use of
the streptavidin-binding peptides as affinity tags is desirable for high
throughput
protein production and purification. For example, purification of fusion
proteins
in a multi-well format may be conducted using magnetic streptavidin beads that
are washed and eluted robotically. The methods of the present invention may
10 also be adapted to purify fusion proteins from ifa vitro translation
mixtures or
from other extracts, such as those from prokaryotic, yeast, insect, or
mammalian
cells, using standard techniques. If necessary, avidin may be added to the
extract
to bind any free biotin in the extract before contacting a sample from the
extract
with streptavidin. Allowing any free biotin to bind avidin may prevent biotin
15 from competing with the streptavidin-binding peptides for binding to
streptavidin.
If desired, the presence of a fusion protein of the invention in a sample
may be detected by incubating the fusion protein with streptavidin (i.e.,
unlabeled
streptavidin or streptavidin that is labeled with a detectable group) under
, conditions that allow streptavidin to bind the fusion protein. Preferably,
the
unbound streptavidin is separated from the streptavidin-bound fusion protein.
Then, the streptavidin that is bound to the fusion protein is detected.
Alternatively, the streptavidin bound to the fusion protein is physically
separated
from the fusion protein and then detected, using standard methods. For
example,
to detect streptavidin that is bound to the fusion protein or that has been
separated
from the fusion protein, Western or ELISA analysis may be performed using an
antibody that reacts with streptavidin or that reacts with a compound that is
covalently linked to streptavidin. If streptavidin is covalently linked to an
enzyme, radiolabel, fluorescent label, or other detectable group, the amount
of
streptavidin may be determined using standard techniques based on a
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
16
characteristic of the detectable group such as its enzyme activity,
radioactivity, or
fluorescence (see, for example, Ausubel et al., supra). Alternatively,
streptavidin
may be contacted with a streptavidin-binding compound that is covalently
linked
to an enzyme, radiolabel, fluorescent label, or other detectable group, and
the
detectable group may be assayed as described herein.
We have also developed an improved method to generate synthetic DNA
libraries encoding full-length proteins, which may be used in a variety of
selection methods to isolate proteins with desired binding affinities or
activities.
The generation of libraries of proteins containing a desired number of amino
acids is often limited by the number of internal initiation events that result
in
truncated proteins and the number of frameshifts that result in either
premature
stop codons or the removal of desired stop codons. For example, during solid
phase DNA synthesis, insertions and deletions which cause frameshifts may
occur due to imperfect coupling and capping efficiencies. In addition, the
random regions in DNA templates may encode stop codons, resulting in
premature truncation of the encoded protein. To address these problems, we
have developed a method in which small DNA cassettes are synthesized, and an
ih vitro selection using the mRNA display technology is performed to enrich
the
library of DNA cassettes for sequences encoding two protein affinity tags.
These
DNA cassettes lack frameshifts and premature stop codons. The selected DNA
cassettes are then cleaved with restriction enzymes and ligated to generate
the
full-length DNA library (Figs. 8A-8F) (Cho et al., J. Mol. Biol. 297:309-319,
2000).
In one preferred embodiment of this method, mRNA display templates that
contain a translation enhancer sequence operably-linked to an open reading
frame
and that terminate in puromycin are generated as described previously (Cho et
al.,
supra). The open reading frame encodes two different protein affinity tags,
such
as a FLAG tag and a hexahistidine tag, Preferably, one of the tags is located
at
the amino-terminus of the encoded peptide, and the other tag is located at the
carboxy-terminus. The mRNA display templates are ire vitro translated to
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
17
generate mRNA displayed peptides (Cho et al., supra). mRNA displayed
peptides encoded by templates that do not contain frameshifts or premature
stop
codons should contain both affinity tags. In contrast, templates that contain
frameshifts or premature stop codons encode peptides without the C-terminal
affinity tag (Fig. 8D). Additionally, mRNA display templates that initiate
internally produce peptides without the N-terminal affinity tag (Fig. 8C). The
library of mRNA displayed peptides is enriched for peptides containing both
affinity tags by purification of the mRNA displayed peptides based on the
presence of these tags (see, for example, Ausubel et al., supra). For example,
the
mRNA displayed peptides may be applied to a matrix designed to bind peptides
containing one of the affinity tags, and the mRNA display peptides without the
affinity tag are washed away. The mRNA display peptides containing the
affinity
tag are then eluted and applied to a second matrix designed to bind the other
affinity tag. The mRNA display peptides recovered from this purification step
are enriched for members containing both affinity tags and thus for full-
length
peptides. These mRNA displayed peptides are reversed transcribed to generate
double-stranded DNA. The amplified DNA is then cleaved by restriction
enzymes. Preferably, this restriction digestion removes the sequences encoding
the affinity tags from the DNA cassettes. The cleaved DNA cassettes are then
ligated to generate the full-length DNA templates.
The experiments described above were carried out as follows.
Generation of a Str~tavidin-Binding Peptide Library
The mRNA display method for selecting peptides or proteins of interest
takes advantage of the translation-terminating antibiotic puromycin, which
functions by entering the A site of ribosomes and forming a covalent bond with
the nascent peptide. By covalently attaching puromycin to the 3' end of an
mRNA, a covalent link between a polypeptide and its encoding message can be
achieved in situ during i~c vitro translation (Roberts et al., Curr. Opin.
Struct.
Biol. 9:521-529, 1999; Liu et al., Methods Enzymol. 318:268-293, 2000). These
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
18
mRNA-peptide fusions can then by purified and subjected to in vitro selection,
yielding the isolation of novel peptide ligands.
A DNA library encoding polypeptides of 108 amino acids was synthesized
as described (Cho et al., supra). The library consisted of short cassettes
concatamerized together. Each cassette encoded a random peptide with a pattern
of polar versus non-polar amino acid side chains compatible with forming an
amphipathic a-helix or (3-strand (Cho et al., supra). The random region was 88
amino acids long, followed by a C-terminal invariant region containing a
hexahistidine tag (Fig. 1A).
The library had a complexity of 2.4 x 1014 at the DNA level. It was
transcribed using T7 RNA polymerase (Fig. 1A), after which a "linker"
oligonucleotide was added to the 3' end using T4 DNA ligase as described (Liu
et
al., supra; Cho et al., supra). The linker consisted of a 21 nucleotide long
dA
stretch, followed by a polyethylene glycol linker, followed by the sequence dA-
dC-dC-puromycin (Liu et al., supra).
This puromycin-terminated mRNA was translated ifa vitro, using the
Ambion (Austin, TX) ioz vitro translation kit under standard conditions for
capped
mRNA. The 10 mL reaction mixture was supplemented with 2 mCi 35S-
methionine and a total methionine concentration of 10 ~,M. The reaction
mixture
also included 300 nM of the library of puromycin-linked mRNA molecules.
After 1 hour at 30°C, MgCl2 and KCl were added to 20 and 710 mM,
respectively, and the reaction mixture was further incubated at room
temperature
for five minutes to increase the yield of displayed peptides. This in vitro
translation produced 1.2 x 1014 polypeptides linked via the puromycin moiety
to
their encoding mRNAs.
These mRNA displayed peptides were then purified on oligo-dT cellulose
(which binds to the oligo-dA sequence in the linker) to remove polypeptides
not
fused to mRNA. For this purification procedure, the reaction mixture was
diluted
10-fold into oligo-dT-binding buffer (1M NaCl, 50 mM HEPES, 10 mM EDTA,
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
19
0.25% Triton X-100, and 5 mM 2-mercaptoethanol at pH 7.9) and 80 mg oligo-
dT cellulose (type 7, Amersham-Pharmacia, Piscataway, NJ) and incubated with
agitation at 4°C for 30 minutes. The mixture was applied to a column
(Poly-Prep
chromatography column, Biorad, Hercules, CA), drained, washed with 10 mL
oligo-dT-binding buffer, washed with 10 mL oligo-dT-wash buffer (300 mM
NaCI, 20 mM HEPES, 1 mM EDTA, 0.25% Triton X-100, and 5 mM 2-
mercaptoethanol at pH 7.9), and washed with 1 mL of 0.5x oligo-dT-wash
buffer. The mRNA-displayed peptides were eluted with 4.5 mL water plus 5 mM
2-mercaptoethanol into tubes containing Triton X-100 and bovine serum albumin
(BSA, New England Biolabs, Beverly, MA) at final concentrations.of 0.15% and
~. g/mL, respectively.
The mRNA-displayed peptides that eluted from the oligo-dT cellulose
column were further purified on Ni-NTA agarose, which binds to the
hexahistidine tags on the polypeptides, to remove any mRNA not fused to
15 polypeptides. The eluted fractions from the oligo-dT cellulose purification
were
exposed to 0.5 mL Ni-NTA-agarose (Qiagen, Valencia, CA) in Ni-binding buffer
[6 M guanidinium chloride, 0.5 M NaCI, 100 mM sodium phosphate, 10 mM
Tris(hydroxymethyl)aminomethane, 0.1% Triton X-100, 5 mM 2-
mercaptoethanol, 4 ~,g/mL tRNA (Boehringer-Mannheim, Indianapolis, IN), and
5 ~,g/mL BSA at pH 8.0)] and incubated for 30 minutes at room temperature.
The matrix was then drained, washed with 12 column volumes Ni-binding buffer,
and eluted with the same buffer plus 100 mM imidazole. Eluted fractions were
combined and de-salted using two successive NAP columns (Amersham-
Pharmacia, Piscataway, NJ) equilibrated in 1 mM
Tris(hydroxymethyl)aminomethane, 0.01% Triton X-100, 50 ~,M EDTA, 0.5 mM
2-mercaptoethanol, 0.5 ,ug/mL tRNA (Boehringer-Mannheim, Indianapolis, IN),
and 50 ~,g/mL BSA at pH 7.6).
The mRNA portion was then reverse transcribed using Superscript II
(Gibco BRL, Rockville, MD) according to the manufacturers instructions, except
that the mRNA concentration was about 5 nM and the enzyme concentration was
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
1 U/~,L. To ensure a high yield in the reaction, a mixture of two primers wexe
used: 1 ~,M of "splint" from the splinted ligation (Cho et al., supra), and 1
,uM of
the 3' PCR primer. After 30 minutes at 42°C, the temperature of the
reaction
mixture was raised to 50° for 2 minutes, and then cooled over 5 minutes
to room
5 temperature to allow gradual peptide folding. Finally, the contents were de-
salted
using NAP columns and subjected to scintillation counting. By comparing the
3sS counts of the purified, reverse transcribed mRNA-peptide fusions to the
35S-
methionine stock and taking into consideration the total methionine
concentration
in the translation reaction (10 ,uM), the number of displayed peptides in this
10 sample was determined to be 6.7 x 1012. This number also represents the
complexity of the library, since it contained virtually no redundancy (the
complexity of puromycin-linker template used in the translation exceeds the
number of recovered displayed peptides by a factor of about 35).
Samples from the synthesis and purification of the mRNA displayed
15 peptides were run on an SDS-PAGE gel, as shown in Fig. 1B.
Selection of Streptavidin-Binding Peptides
For selection of peptides with high affinity for streptavidin, the above
mRNA displayed peptide library was incubated with immobilized streptavidin
20 (Ultralink immobilized streptavidin plus, about 4 mg/mL; Pierce, Rockford,
IL)
in streptavidin-binding buffer under reducing conditions (40 mM
Tris(hydroxymethyl)aminomethane, 300 mM KCI, 2 mM EDTA, 0.1% Triton X-
100, 5 mM 2-mercaptoethanol, 100 ,ug/mL BSA, and 1 ,ug/mL tRNA at pH 7.4).
The amount of gel used was 0.5 mL in a total volume of 5.5 mL. After
incubating for 20 minutes at room temperature, the contents were loaded onto a
disposable chromatography column, drained, washed with 14 column volumes of
streptavidin-binding buffer, and eluted with five successive aliquots, at 10
minutes intervals, of streptavidin-binding buffer plus 2 mM D-biotin (Sigma,
St.
Louis, MO) (Fig. 1A). The fraction of the library that survived this
purification
was 0.08%. Elution fractions were combined, de-salted on NAP columns, and
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
21
then PCR-amplified to regenerate the double-stranded DNA library using the
described conditions and primers in a 8 mL reaction mixture (Cho et al.,
supra).
This concluded the first round of selection, and the remaining six rounds
followed the same protocol except that the translation was scaled down 10-
fold,
and the number of column volumes for washing the streptavidin column was
increased (32 volumes for round 2; 40 volumes for rounds 3, 4 and 6; and 25
volumes for rounds 5 and 7). The streptavidin-binding selection for rounds 5
and
7 was performed directly on the streptavidin-column eluate from the preceding
selection rounds, without intervening amplification (the biotin was removed by
three successive passages through NAP columns). PCR products amplified after
the seventh selection round were cloned using the TOPO TA cloning kit
(Invitrogen, Carlsbad, CA), following the manufacture's protocol. The fraction
of the library that bound and eluted from the streptavidin column increased in
each round, reaching 61 % at round seven (Fig. 2A).
Characterization of the Selected Library
The eluate from the seventh round of selection was amplified by PCR.
The resulting PCR DNA was used to synthesize a library of displayed peptides
to
confirm that the displayed peptides, rather than the RNA or DNA portion of the
library constructs, were responsible for the interaction with streptavidin.
Treatment of the library with RNAse A did not reduce the extent of
bindinglelution from the matrix (Fig. 2B). Also, biotin-saturated streptavidin
showed no binding to the peptide library (Fig. 2B). These results demonstrated
that the interaction of the selected peptides with the streptavidin matrix was
specific for the unligated protein, rather than for any other component of the
matrix.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
22
Seduence Analysis of Selected Peptides
Thirty-three randomly chosen clones from~the PCR DNA from round
seven were chosen for sequencing. Twenty different sequences were observed
(Fig. 3). Surprisingly, all 20 sequences were frame-shifted from the intended
frame (frame 1) to frame 3 by deletion of two nucleotides or addition of one
nucleotide. The designed pattern of polar and non-polar residues was therefore
discarded, leaving an unpatterned, essentially random sequence. Prior to the
selection, about half of the library members were in frame 1 throughout their
entire open reading frames (Cho et al., supf~a). Frame 3 appears to have been
enriched over frame 1 due to the increased frequency of the sequence HPQ.
Frame 1 has a low incidence (1:45,000 library members) of the sequence HPQ
due to the designed polar/non-polar pattern. By contrast, frame 3 had a much
higher expected incidence of the HPQ sequence ( 1:64), similar in frequency to
that of a library of the same length and with equal mixtures of all four
nucleotides
at each position (1:193). Also, frame 3 was rich in histidine, thus allowing
retention on the Ni-NTA column. The Ni-NTA purification protocol was
intended to eliminate library mRNA molecules not displaying peptide, but was
not performed under sufficiently stringent conditions so as to eliminate
peptides
with small numbers of histidines. Frame 2 had a high incidence of stop codons.
Nineteen of the 20 clones had at least one HPQ motif, and five clones
contained two such motifs (Table 1). The clones were organized according to
the
number of times the HPQ and related tripeptide motifs occur (Table 1). The
number of amino acids between the two motifs, when present, ranged from four
to 74.
Binding Affinities of Peptides
To rapidly assay each of the 20 selected peptides to determine their affinity
for streptavidin, a new method for preparing, tagging and purifying the
peptides
was employed. For generation of the DNA-tagged peptides, plasmids containing
single inserts were used as templates for PCR-amplification using the same 5'
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
23
PCR primer as described for the library construction (Cho et al., supra), and
a
new 3' primer (5'-ATAGCCGGTGCCAAGCTTGCAGCCGCCAGACCAGT-3';
SEQ ID No. 30), which altered the 3' RNA sequence to
ACUGGUCUGGCGGCUGCAAGCUUGGCACCGGCUAU (SEQ ID No. 31).
This sequence was designed to anneal to the photo-crosslinking linker, which
has
the sequence 5'-psoralen-TAGCCGGTG-A17-CC-puromycin-3', in which the
underlined bases are 3'-methoxy nucleotides and the remaining bases are
deoxynucleotides (the oligonucleotide was synthesized using reagents from Glen
Research, Sterling, VA). This new primer changed the constant C-terminal
peptide sequence from WSGGCHI3HI3HHSSA (SEQ ID No. 32) to
WSGGCKLGTGY (SEQ ID No. 33), of which the last three amino acids may not
be translated because they are annealed to the linker. Each DNA template was
transcribed and gel purified, as described (Cho et al., supra), and then
incubated
with the psoralen linker under the following conditions: 2 ~,M mRNA, 4 ~.M
linker, 50 mM Tris(hydroxymethyl)aminomethane, 200 mM KCI, and 10 mM
spermidine at pH 7.4 and 70°C for 2 minutes, and then cooled to
4°C over 5
minutes. Samples were then placed in the cold room in a 96 well plate (50
~,L/well), one inch above which was suspended a UV lamp (366 nm, Ultraviolet
Products, Inc., San Gabriel, CA, model number UVL-21) for 15 minutes. Then,
the reactions mixtures were de-salted using a G-50 Sephadex spin column
(Boehringer Mannheim, Indianapolis, IN). The translation/display reactions and
oligo-dT-purification were carried out as above. Finally, RNase A (200 ng/mL,
10 minutes, room temperature) was added to degrade the mRNA, leaving
peptides fused to a short DNA oligonucleotide. Complete degradation was
confirmed by SDS-PAGE analysis.
The resulting purified DNA-tagged peptides (DTP) were analyzed in a
streptavidin column-binding assay, in which 500 pM 35S-labeled DTP were
mixed with 50 ~,L of the streptavidin matrix in streptavidin-binding buffer,
in a
total volume of 300 JCL, and incubated for 10 minutes at room temperature with
agitation. Then, the contents were loaded onto a chromatography column. The
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
24
column was drained and washed with 80 column volumes of streptavidin-binding
buffer, and then eluted with three consecutive aliquots (3 column volumes
each)
of streptavidin-binding buffer plus 2 mM biotin over a 15 minute period. All
fractions (flow-through, washes, elutions, and irreversibly bound counts) were
analyzed by scintillation counting to determine the fraction of DTP that bound
streptavidin and eluted with biotin (Table 1). The non-selected clone in which
two HPQ motifs (separated by 19 amino acids) were introduced encoded the
sequence
MDEAHPQAGPVDQADARLVQQGALQHHPQGDRMMSGGCKLGTGY
(SEQ ID No. 34), in which the underlined portions are identical the
HPQ,regions
of clone SB2.
The results of this analysis are shown in Table 1. For comparison, two
HPQ motifs, separated by 19 residues, were introduced into a control,
unselected
member of the library. The low percentage of this control peptide that
specifically eluted from the streptavidin column (0.16%) indicated that the
presence of two HPQ motifs was not sufficient for high affinity binding. In
contrast, a greater percentage of the selected peptides (8.3 to 88%) was
retained
on the column during the washing step and then specifically eluted with
biotin.
The dissociation constants of the selected peptides for streptavidin were
measuring using an electrophoretic mobility shift assay (EMSA). In this assay,
DTP's were incubated with varying amounts of pure streptavidin (Pierce
Immunopure Streptavidin, Rockford, IL) in streptavidin-binding buffer plus 5%
glycerol to increase the density of the solution so that it could collect at
the
bottom of the gel well. After incubating at room temperature for 20 minutes,
the
reactions mixtures were moved into the cold room, where they remained for 10
minutes before being carefully loaded onto a 10% polyacrylamide:bisacrylamide
(37.5:1, National Diagnostics, Atlanta, GA) gel (thickness 0.7 mm, height 16
cm,
width 18 cm) containing 2X TBE, 0.1% Triton X-100 and 5% glycerol. The gel,
which had been pre-run for 30 minutes at 13 watts, and the running buffer were
pre-cooled to 4°C. Then, the gel was run in the cold room at 13 watts,
which
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
increased the temperature of the gel to about 20°C. The gel was run for
45 to
120 minutes, depending on the mobility of the particular DTP. Then, the gel
was
fixed in 10% acetic acid and 10% methanol for 15 minutes, transferred to
electrophoresis paper (Ahlstrom, Mt. Holly Springs, PA), dried, and analyzed
5 using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA).
The short DNA oligonucleotide tag on the DTP's allowed them to migrate
in a native gel, and the addition of unlabeled ligand (i.e., streptavidin)
caused a
mobility shift for several of the clones. The concentration of DTPs was less
than
1 nM in each titration, and thus the dissociation constant (Kd) can be
10 approximated by the concentration of streptavidin that results in half of
the DTP
being mobility-shifted. To determine the Kd, several different measurements
were taken in the range of 25-75% of DTP bound (values outside of this range
were unreliable due to background and close proximity of the bound and
unbound bands in the gel). The Kd was determined using the equation Kd =
15 [streptavidin]*R, where R is the ratio of unbound to bound DTP (ratio of
unshifted to shifted band). Independent measurements on gels prepared at
different times were used for each clone (the number of different
measurements,
n, is shown in Table 1). Streptavidin concentrations were measured by UV28z,
using the molar extinction coefficient of 57,000 per monomer.
20 Examples of these mobility shifts in the presence of streptavidin are shown
in Fig. 4A. Some clones showed either no shift or poorly defined bands,
suggesting that the lifetime of these complexes was too short for detection
using
this method. We chose five of the most well behaved clones and quantitatively
examined their mobility shifts in response to a range of streptavidin
25 concentrations. An example of a streptavidin titration experiment for
peptide
SB 19 is shown in Fig. 4B, and the data is graphed in Fig. 4C. The
dissociation
constants for the clones ranged from 110 nM to less than 5 nM (Table 1). These
surprisingly high affinities were comparable to those for monoclonal antibody-
antigen interactions, demonstrating that even random, non-constrained peptide
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
26
libraries can be a source of avid ligands to proteins that do not normally
function
in peptide binding.
Dissection of Clone SB 19
Clone SB19 possessed only one HPQ motif, bound to 85% in the column
binding assay, and had a Kd for streptavidin of 10 nM (Table 1). A series of C-
terminal truncation constructs (Cl-C4) were constructed and assayed in the
streptavidin column-binding assay (Fig. 5). C-terminal truncation analysis of
clone SB 19 was performed using standard methods by amplifying the clone with
the original 5' primer and a series of 3' primers that truncated the sequence
at
various positions and also replaced two codons (encoding Asp and Trp) in the C-
terminal constant region with methionine codons to increase the 35S-
incorporation. Analogous primers were used for the N-terminal truncation
analysis, except that no change was made in the N-terminal constant sequence.
Deletion of up to 56 residues had no observable effect on the binding
strength. Peptide SB 19-C4 retained only the first 38 residues from the
selected
construct (plus the C-terminal MMSGGCKLG sequence, SEQ ID No. 36).
Mutating the HPQ motif to HGA reduced the activity by three orders of
magnitude (compare construct C4 to M1). Results from the N-terminal
truncation constructs (N1-N3) suggested that binding determinants were spread
throughout the N-terminal 38 residues of peptide SB 19. Of the peptides
tested,
SB 19-C4 was therefore the minimal peptide retaining full activity in this
assay.
EMSA analysis of peptide SB 19-C4 confirmed high affinity streptavidin-
binding,
but a fraction (13%) of the peptide was inactive even at streptavidin
concentrations >1 ,uM. The majority (87%), however, had an apparent Kd of 4.9
nM after correction for the amount of inactive peptide.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
27
Purification of Fusion Protein Containing Strentavidin-Binding Peptide
A fusion protein containing the first 38 amino acids of the SB19-C4
streptavidin-binding peptide (Fig. 7A) was expressed in E. coli and then
purified
from the cell lysate. For the expression of the fusion protein, BL21 (DE3)
cells
were transformed with a plasmid containing a
Maltose Binding Protein--Streptavidin-binding Peptide--Hiss-Protein of
Interest
insert which encodes a fusion protein containing, from the amino- to carboxy-
terminus, maltose-binding protein, the first 38 residues of the SB 19-C4
sequence,
a hexahistidine tag, and another peptide called 2r18-l9dN (Figs. 6A-6C). This
insert was constructed using standard molecular biology techniques (see, for
example, Ausubel et al., supra). Each of these domains of the fusion protein
is
separated by a few amino acids to allow proper folding of the domains.
A kanamycin-resistant colony was selected and grown overnight in 10 ml
LB media with 50 mg/liter kanamycin at 37°C. This starter culture was
diluted
100-fold into 1000 ml LB with 50 mg/liter kanamycin, and the culture was grown
at 37°C to OD6oo of 1.8 at 37°C. Expression of the fusion
protein was induced by
addition of 1mM IPTG, and the culture was grown for another two hours. The
cells were pelleted by centrifugation at 5000 X g for 20 minutes. The pelleted
cells were resuspended in 5% of the original volume of 1 mM EDTA and MBP
buffer (10 mM HEPES.HCI, 10 mM HEPES.Na+, 200 mM KCl, 0.25% w/w
Triton X-100, and 10 mM BME at pH 7.4) and frozen slowly at -20°C
overnight.
The sample was thawed in the morning and sonicated on ice. The cell lysate was
obtained by collection of the supernatant after centrifugation at 14,000 X g
for 20
minutes at 4°C.
To purify the fusion protein, the cell lysate was applied to a column
containing immobilized streptavidin, with a capacity of about 1 mg/ml, that
had
been washed with eight column volumes of MBP buffer. Then, the column was
washed with 12 column volumes of MBP buffer. The fusion protein was eluted
with MBP buffer containing 2 mM biotin. Samples of the cell lysate and eluted
protein were analyzed by SDS-PAGE on an 8% gel (Fig. 7B). The lane
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
28
containing the purified protein had a band of the expected size. No other
bands
were observed, except for a faint band of slightly higher mobility (Fig. 7B).
This
band was probably a degradation product of the fusion protein that was missing
a
few amino acids from either the amino- or carboxy-terminus but retained the
streptavidin-binding peptide and thus retained the ability to bind the
streptavidin
column. Thirty percent of the fusion protein loaded onto a column containing
immobilized streptavidin was recovered after washing the column with 12
column volumes of buffer. Thus, the high affinity of the fusion protein for
streptavidin allowed extensive washing of the column to remove contaminating
proteins, while retaining a significant amount of the desired fusion protein.
Similar attempts to purify the same protein from the cell lysate, using
either amylose resin, which binds to the maltose-binding protein portion of
the
fusion protein, or Ni-NTA resin, which binds to the hexahistidine tag,
resulted in
the same expected size band. However, several contaminating bands were also
present.
Biacore Analysis of the Affinity of a Fusion Protein for Streptavidin
Another fusion.protein containing the first 38 amino acids of the SB 19-C4
streptavidin-binding peptide was expressed and purified from E. coli. This
fusion protein contained, from the amino- to carboxy- terminus, maltose-
binding
protein, the first 38 amino acids of the SB 19-C4 sequence, and a
hexahistidine
tag (Fig. 9B, SEQ ID No. 41). The plasmid (Fig. 9A, SEQ ID No. 40) encoding
this fusion protein was constructed using standard molecular biology
techniques
and used to express the fusion protein in E. coli. as described above. This
fusion
protein was purified from the E. coli extract using amylose resin to bind the
maltose-binding protein portion of the fusion protein and then Ni-NTA resin to
bind the hexahistidine tag.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
29
To measure the affinity of the fusion protein for streptavidin, the fusion
protein was immobilized on a biacore chip through the crosslinking of free
amino
groups in the fusion protein to the biacore chip. Buffer containing
streptavidin
was washed over the chip, allowing streptavidin to bind the immobilized fusion
protein (Fig. 10B). This resulted in an increase in the biacore response units
which are proportional to the amount of streptavidin adhering to the biacore
chip.
Then buffer without streptavidin was washed over the chip, and the biacore
response units decreased as streptavidin dissociated from the immobilized
fusion
protein (Figs. 10A and 10B). To measure the association rate for the binding
of
streptavidin to the fusion protein, streptavidin concentrations of 1.6, 0.8,
0.4, 0.2,
0.1, or 0.05 ~,M, (lines "a" to "p' in Fig. 10B, respectively) were washed
over the
biacore chip. The buffer also contained 40 mM
Tris(hydroxymethyl)aminomethane, 300 mM KCl, 2 mM EDTA, 0.1% w/v
Triton X-100, and 5 mM 2-mercaptoethanol at pH 7.4. This data was used to
calculate an association rate, ka, of 5 x 10~/M/s, as described previously
(BIACORE X Instrument Handbook, version AA, Biacore AB, Uppsala Sweden,
1997). To measure the dissociation rate, a pulse of 23, 11.5, or 5.75 ~,M
streptavidin in the buffer described above was administered, and then buffer
without streptavidin was washed over the chip (Fig. 10A). This data was used
to
calculate an upper limit of 2 x 10-3/s for the dissociation rate, kd (BIACORE
X
Instrument Handbook, supra). Based on these calculated association and
dissociation rates, the dissociation constant, Kd, for the binding of
streptavidin by
this fusion protein was less than 40 nM. This result confirms the high
affinity
binding of the SB 19-C4 peptide for streptavidin that was observed in the
streptavidin column-binding assay and the EMSA assay (Table 1). Additionally,
this result demonstrates that this peptide maintains its high affinity for
streptavidin when expressed as part of a fusion protein.
CA 02426035 2003-04-16
WO 02/38580 PCT/US00/41717
Other Embodiments
From the foregoing description, it will be apparent that variations
and modifications may be made to the invention described herein to adopt it to
various usages and conditions. Such embodiments are also within the scope of
5 the following claims.
All publications mentioned in this specification are herein
incorporated by reference to the same extent as if each independent
publication or
patent application was specifically and individually indicated to be
incorporated
by reference.
What is claimed is:
