WO2010129265A2 - Compositions comprising monomeric streptavidin and methods for using same - Google Patents

Abstract

Provided are improved streptavidin proteins and methods for using them. The improved streptavidin proteins can contain insertions, deletions and amino acid replacements, and contain at least one engineered disulfide-forming pair of amino acids. Also provided are method of the mutant streptavidin proteins to isolate biotinylated molecules, and complexes that contain mutant streptavidin proteins and biotinylated molecules.

Description

COMPOSITIONS COMPRISING MONOMERIC STREPTAVIDIN AND METHODS

FOR USING SAME

This application claims priority to U.S. patent application no. 61/172,996, filed April 27, 2009, the entire disclosure of which is incorporated herein by reference.

FIELD OF THE INVENTION

[0001] The present invention relates generally to streptavidin and more specifically to compositions comprising altered streptavidin amino acid sequences.

BACKGROUND OF THE INVENTION

[0002] Streptavidin is a protein that has high affinity for biotin. This makes the molecule useful in biotechnology applications, e.g. labeling, surface modeling, and purification of biotinylated compounds. However, the molecule is a tetramer which prevents its use in some situations, e.g. as a fusion partner in endogenous cellular systems. There is an ongoing and unmet need for improved streptavidin-based molecules for use in a wide variety of binding reactions.

SUMMARY OF THE INVENTION

[0003] The present invention provides improved streptavidin molecules, compositions comprising the improved molecules, and methods for using them. In particular, the invention provides streptavidin mutant proteins containing mutations that can be categorized in two groups. Group I mutations comprise insertion mutations between streptavidin amino acids N23 and G26. The insertions contain 8, 9, 10, 11 or 12 amino acids and result in deletion of two amino acids (Q24 and L25). The insertions extend streptavidin strands 1 and 2 and are believed to inhibit domain-swapped dimer formation between chains A and D through steric repulsion. Group II mutations are changes distributed between residues A34 and S 139. The changes are believed to improve biotin binding by stabilizing structure. All mutant proteins comprising Group II mutations contain an engineered disulfϊde-forming pair. In one embodiment, the engineered disulfϊde-forming pair is comprised of E51C and Y83C.

[0004] The invention also provides method of using the mutant streptavidin proteins. In one embodiment, the invention provides a method of isolating a biotinylated molecule. The method comprises providing a composition comprising biotinylated molecules, contacting the biotinylated molecules with a mutant streptavidin protein as further described herein, allowing formation of mutant streptavidin-biotinylated molecule complexes, and separating the complexes from other components of the composition.

[0005] Also provided are compositions comprising a mutant streptavidin protein and a biotinylated molecule, which may be present as a complex.

BRIEF DESCRIPTION OF THE FIGURES

[0006] Figure 1 is an illustration of the process used in reconstruction of the streptavidin hydrophobic binding pocket by insertion of a loop.

[0007] Figure 2 is an illustration of a domain swapped dimer.

[0008] Figure 3 provides a photographic representation of electrophoretic separation of mutant streptavidin proteins provided by the invention.

[0009] Figure 4 is a graphical representation of results obtained from flow cytometry analysis of equimolar amounts of FLAG-tagged M57 and QM proteins mixed with an equal amount of biotinylated microbeads. The amount of bound protein is determined by flow cytometry using an anti-FLAG antibody and a fluorophore-conjugated anti-IgG antibody to detect bound protein. In this histogram, the x-axis represents fluorescence intensity which is correlated with the amount of bound protein; the y-axis represents the number of biotin beads.

[0010] Figure 5 is a graphical representation of data obtained from analysis using a procedure similar to that used to obtain the data summarized in Figure 4 and demonstrates various relative biotin binding affinities for wild type and mutant streptavidin proteins.

[0011] Figure 6 is a graphical representation of data showing relative effects of dithiothreitol (DTT) on wild type streptavidin as compared to a mutant streptavidin comprising engineered disulfides. The experiment was performed as in Figure 4, but wild type tetramer or mutant streptavidin monomer was incubated with DTT at the indicated final concentrations prior to incubating with biotinylated beads. The amount of bound streptavidin was quantified by flow cytometry using anti-FLAG antibody and fluorophore-conjugated anti-IgG antibodies. The amount of bound tetramer does not change, but the amount of bound monomer decreases at high concentrations of DTT because the engineered disulfide bond is reduced and results in loss of activity.

[0012] Figure 7 is a photographic representation of results from electrophoretic separation of a mutant streptavidin protein provided by the invention with and without biotin during purification (left panel) and the effects of DTT . These SDS-PAGE gels show that the streptavidin mutant containing the T76R mutation (Wu and Wong 2005), which and U.S. Patent No. 7,265,205), forms a domain swapped, functional dimer between the subunits A and D. In this regard, the double mutant containing T76R and V 125 C crosslinks to form a dimer that is visible on the gel as a high molecular weight band. As expected of a disulfide - crosslinked dimer, the species corresponding to the ~ 40 kDa band can be reduced with DTT.

DETAILED DESCRIPTION OF THE INVENTION

[0013] The present invention provides novel monomeric streptavidin (mSAV) that does not form a tetramer and is expected to bind biotin with high affinity. Also provided are mutant streptavidin proteins that can form domain swap dimers with improved properties.

[0014] The invention relies in part on our design of mSAV that reconstitutes the structural properties of wild type tetramer streptavidin, which underlie high affinity biotin binding. The mSAV described here for the first time can serve a unique role in situations where high affinity noncovalent interaction is desired but oligomerization/aggregation is unwanted.

[0015] Wild-type streptavidin is a tetramer (SAV4) and binds biotin with Kd ~ 10-14M. It is expected that the invention will provide mSAV that binds biotin with Kd that is within 1, 2, 3 or 4 orders of magnitude of the biotin binding Kd of wild-type streptavidin. The amino acid sequence of "wild type" streptavidin is known in the art and has the following amino acid sequence:

[0016] OVSKDSKAQVSAAEAGITGTWYNQLGSTFIVTAGADGAL TGTYESA VGNAES RYVLTGRYDSAPATDGSGTALGWTVAWKNNYRNAHSATTWSGQYVGGAEARINTQ WLL TSGTTEANA WKSTL VGHDTFTKVKPS AASID AAKKAGVNNGNPLD AVQQ (SEQ ID NO:1).

[0017] When produced naturally, the 159 residue full-length protein presented in SEQ ID NO:1 is processed to yield a shorter "core" streptavidin sequence that consists of residues 13 - 139 (Pahler et al. 1987). Italicized and bolded amino acids in SEQ ID NO:1 represent the core streptavidin sequence, and the core sequence is separately presented as SEQ ID NO:2 as follows:

[0018] GITGTWYNQLGSTFIVTAGADGALTGTYESAVGNAESRYVLTGRYDSAPATD GSGTALGWTVAWKNNYRNAHSATTWSGQYVGGAEARINTQWLLTSGTTEANAWK STLVGHDTFTKV (SEQ ID NO:2). It is considered for the purposes of the invention that the AEA sequence conventionally shown at the N-terminus of the processed core peptides are not necessary for the peptides to retain their utility, particularly since these three amino acids are frequently disordered in crystal structures of streptavidin, and the core sequence without the AEA sequence has been shown to be at least as active as the core sequence with it (Sano et al. 1995).

[0019] Because wild type (wt) streptavidin binds biotin with high affinity (Kd ~ 10-14 M), the biotin-streptavidin interaction is the basis of many biotechnology applications, including detection, labeling, immobilization, purification, conjugation, and crosslinking (M. Wilchek, E. A. Bayer, Foreword and introduction to the book (strept)avidin-biotin system. Biomol Eng 16, 1 (1999)). Wt streptavidin is a tetramer, which is a dimer of two dimers (A/B and C/D, where A - D are chain numbers). There is experimental evidence that tetramerization is important for high affinity biotin binding. For example, the mutations that disrupt the tetramer association reduce the affinity for biotin by 6 or 7 orders of magnitude. (T. Sano, C. R. Cantor, Intersubunit contacts made by tryptophan 120 with biotin are essential for both strong biotin binding and biotin-induced tighter subunit association of streptavidin. Proc Natl Acad Sci U S A 92, 3180 (1995)).

[0020] High affinity biotin binding requires tetrameric streptavidin. The obligate tetramerization limits its applications where target aggregation is a concern. In order to avoid target aggregation, a mutant streptavidin tetramer has been constructed that contains only one high affinity biotin binding site by mixing active and inactive subunits and isolating the complex containing one active subunit and three active subunits (M. Howarth, D. J. Chinnapen, K. Gerrow, P. C. Dorrestein, M. R. Grandy, N. L. Kelleher, A. El-Husseini, A. Y. Ting, A monovalent streptavidin with a single femtomolar biotin binding site. Nat Methods 3, 267 (2006)). However, preparing a monovalent tetramer in the described manner results in a significant waste of the starting materials, since the complexes with incorrect stoichiometry are discarded. Some of the challenges we faced in arriving at the present invention include but were not limited to the unpredictability of computational protein design, backbone flexibility, side chain entropy, parameterization of van der Waals radii, and discrete rotamers.

[0021] A genuine monomer that binds biotin tightly has an advantage over monovalent tetramer in a variety of applications. It is straightforward to use it in labeling experiments. It is also much easier to prepare in the lab compared to a monovalent tetramer. Monomeric streptavidin can not only detect and label biotinylated proteins but can also be directly fused a heterologous protein for detection with biotinylated chromophores or fluorophores.

[0022] A monomeric streptavidin mutant has been previously reported (U.S. Patent No. 7,265,205 and S. C. Wu, S. L. Wong, Engineering soluble monomeric streptavidin with reversible biotin binding capability. J Biol Chem 280, 23225 (2005)). The designed mutant contains a bulky substitution, T76R, at the dimer interface that disrupts dimer formation. The mutant also contains three additional mutations to stabilize the resulting monomer (V55T, L109T, V 125R). However, we have determined that this reported quadruple mutant (the quadruple mutation is referred to herein as "QM") is not a true monomer but forms a dimer through domain swapping. (A graphical depiction of a domain swapped dimer is provided in Figure 2.) We proved this in two ways. First, we demonstrated that purified QM forms a protein complex that has more than one biotin binding site by immobilizing it on biotinylated microbeads and demonstrating that the bound protein is still capable of recognizing other biotinylated molecules. We use biotinylated antibody (b-Ab) as a nominal reagent for this purpose. The bound b-Ab is then detected using anti-IgG antibody. Second, when we introduced a cysteine mutation at the A/D interface (V125C) together with T76R, L109T, V55T, the mutant forms a disulfide crosslinked dimer, which corresponds to a domain swapped dimer (Figure 7). Additionally, an L124D mutation prevents domain swapping through electrostatic repulsion.

[0023] Wt streptavidin tetramer also has high thermal and chemical stability with the melting temperature Tm = 16 ⁰C The biotin-streptavidin complex has even higher stability with Tm = 112 ⁰C (Gonzalez et al. 1999). High thermal stability is important in most applications to avoid unfolding, degradation, and precipitation. However, QM has poor water solubility and easily aggregates during and post purification (S. C. Wu, et al., J Biol Chem 280, 23225 (2005)) and our observation). Similarly, previously engineered streptavidin dimers also have limited solubility (T. Sano, Set al. Proc Natl Acad Sci U S A 94, 6153 (1997). [0024] The invention provides various streptavidin mutants. The mutations can be categorized as belonging to two groups. The first (Group I) comprise insertion mutations between N23 and G26 containing 8 - 12 amino acids. The mutated residues extend strands 1 and 2, and overlap with the residues from a domain swapped dimer. Without intending to be bound by theory, it is believed that the insertion also prevents domain swapping that leads to a domain-swapped dimer formation between chains A and D (also referred to as functional dimer) through steric repulsion. Therefore, dimerization is not compatible with these mutations. Further, these mutations are believed to contribute to biotin binding by reducing solvent accessibility to the binding pocket. Those skilled in the art will recognize that here the chain designation is as in the PDB structure ISWE. The second group (Group II) of mutations are distributed between residues A34 and S 139 and are believed to indirectly contribute to biotin binding by stabilizing the monomeric structure. The disulfide mutations described more fully below are believed to be part of these stabilizing mutations. The stabilizing mutations also prevent native dimer formation between chains A and B due to introduced bulky side chains at the interface, including E74, N78, R91 and R93. In this regard, we have shown that R93 alone is sufficient to disrupt the native dimer formation.

[0025] With respect to the mutant streptavidin proteins described herein, for convenience of reference, amino acid positions refer to the positions of amino acids in the full-length streptavidin sequence shown in SEQ ID NO: 1. Thus, while the core sequence presented as SEQ ID NO:2 has GIy as its first amino acid, that GIy is referred to as occupying streptavidin position 16, owing to that Ala being present at position 13 in the wild-type sequence. The same rationale applies to other amino acid positions in the sequences presented herein, unless stated otherwise.

[0026] In one embodiment, the invention provides a mutated streptavidin protein having the sequence of SEQ ID NO:2, with the exception that residues 51 and 83 have been changed to cysteines (i.e. E51C and Y83C mutations). Thus, E51C and Y83C, when used to designate changes to Cys in the core sequence, refer to replacing residues that occupy positions 39 and 71 in the core sequence, respectively, with Cys. Those skilled in the art will be familiar with this amino acid position naming convention when referring to the streptavidin core and wild type sequences, as well as the mutant streptavidin proteins disclosed herein. A mutated streptavidin sequence comprising E51C and Y83C is provided as SEQ ID NO:3. The cysteines in SEQ ID NO:3 that replace the wild type amino acids as the E51C and Y83C mutations are italicized, enlarged and bolded.

[0027] GITGTWYNQLGSTFIVTAGADGALTGTYESAVGNACSRYVLTGRYDSAPAT DGSGTALGWTVAWKNN CRNAHSATTWSGQ YVGGAE ARINTQWLLTSGTTEANAW KSTLVGHDTFTKV (SEQ ID NO:3). Representative electrophoretic separations of proteins provided by the invention are depicted in Figure 3 under native and denatured conditions. The SDS-PAGE gel shows that streptavidin wild type is a tetramer and migrates as a tetramer under native conditions but dissociates to monomers if heat denatured. In contrast, all our engineered streptavidin mutants migrate as monomers under native conditions

[0028] In alternative embodiments the protein can have cysteine mutations at positions i) 51 and 83, ii) 52 and 83, iii) or 51 and 84, thus repositioning the disfulide bond location by one amino acid. Amino acid sequences comprising such alternative location of the cysteines will be readily recognizable to those skilled in the art and are part of the instant disclosure.

[0029] In connection with the mutated streptavidin protein sequences presented herein, the invention also provides mSAV that has at least 95% homology to the mutated sequences. Thus, the invention includes proteins having 95%, 96%, 97%, 98% and 99% homology to the protein sequences presented herein. Any of the protein sequences presented herein may alternatively comprise or consist of the presented amino acid sequences.

[0030] In various embodiments, the proteins presented herein may comprise or consist of at least 118 amino acids.

[0031] In various embodiments, a protein of the invention comprises or consists of between 118 and 168 amino acids, inclusive, and including all integers there between.

[0032] In various embodiments, the mutated streptavidin proteins provided by the invention consist of 121, 122, 123, 124, 125, 126, 127, 128, 129, or 130 amino acids.

[0033] In one embodiment, the invention provides a mutated streptavidin protein comprising the streptavidin core sequence, or a sequence that has at least 95% homology with the core sequence, with the proviso that the mutated core sequence and the protein having at least 95% homology to it contains both of the E51C and Y83C mutations, or in alternative embodiments, cysteine pairs at positions 51 and 83, 52 and 83, or 51 and 84. Figures 6 shows that the amount of bound tetramer does not change, but the amount of bound monomer decreases at high concentrations of DTT because the engineered disulfide bond is reduced and results in loss of activity, thus demonstrating the importance of the disulfide formation.

[0034] In various embodiments, the mutant streptavidin proteins provided by the invention may further comprise deletions and/or insertions (e.g., the Group I mutations). For example, a mutant streptavidin protein provided by the invention can have a deletion of one or two residues and/or an insertion of 9, 10, 11 or 12 amino acid residues. Any protein described herein can be constructed with or without such an insertion/deletion modification.

[0035] In the case of the deletions, insertions, and combinations thereof, the invention includes proteins having at least 95% homology to the proteins comprising the deletions and/or insertions, and such proteins are expected to be monomeric and yet have avidin binding properties that provide utility in a wide variety of biological applications, including but not limited to protein purification and tagging techniques that, given the benefit of the present disclosure, will be readily recognized by those skilled in the art. Proteins comprising insertions and/or deletions also contain both of the E51C and Y83C mutations.

[0036] In one embodiment, the invention provides mSAV proteins comprising a deletion of two residues and/or an insertion of eleven residues, leading to a net addition of nine residues. Without intending to be bound by any particular theory, it is considered that such an insertion precludes the formation of domain-swapped dimers, and thus facilitates persistence of true monomeric streptavidin. In particular, a unique feature of certain embodiments of the present invention is a loop that covers the biotin binding pocket and prevents solvent from disrupting the biotin-mSAV binding. The loop is created by the 2 residues that are deleted and 11 residues that are added for the net addition of nine. Thus, the formation of a structure that blocks the biotin binding site (a loop-like structure) is one feature of the streptavidin monomers of the present disclosure. An illustration of the construction of a mutant streptavidin having this feature is presented in Figure 1.

[0037] In alternative embodiments, no loop is present (i.e., the protein only comprises Group II mutations), which can result in the formation of domain swapped dimers, if such structures are desired for use in any particular circumstance. In particular, a stable domain swapped dimer can be constructed using the mutations V 125 C. In wt, V 125 is at the dimer-dimer interface and contributes to the tetramer formation. Likewise, V125R mutation destabilizes the tetramer formation. We have shown that V 125 C crosslinks and stabilizes a domain swapped dimer (Figure 7). Because a domain swapped dimer preserves the ligand binding pocket, it binds biotin with a significantly higher affinity than a monomer, but is only half the size of wt tetramer. The two binding sites of a domain swapped dimer are anti-cooperative because they are located physically close and the binding of a biotinylated ligand in one binding site sterically interferes with the binding of another ligand at the other location (G. Kada, H. FaIk, H. J. Gruber, Accurate measurement of avidin and streptavidin in crude biofluids with a new, optimized biotin-fluorescein conjugate. Biochim Biophys Acta 1427, 33 (1999)). Therefore, a domain swapped dimer may be functionally similar to a monomer. Such a construct may be advantageous in certain situations in which the requirement of monovalency is not strictly necessary and high affinity interaction is required. A comparison of a known streptavidin mutant (QM) with a mutant streptavidin provided by the invention (M57) is provided in Figure 4. It is notable that from Figure 4, despite the apparent domain swapping for both m57 and QM, m57 binds biotin better than QM. Further, QM precipitates during purification, but m57 does not, which is likely due to improved solubility and/or stability facilitated by the novel mutations we have engineered into it. We have also demonstrated that M57 yield is higher during purification and it stays active longer after purification relative to QM. In contrast, QM precipitates heavily during and after purification and rapidly loses its biotin binding function. These experiments provide strong indication that the designed mutations are beneficial and help stabilize the structure of the molecule while promoting stronger binding to biotin.

[0038] In one embodiment, a mSAV provided by the invention comprises an insertion between N23 and T26. Thus, a two residue deletion of Q24 and L25 is replaced by the insertion. Q24 and L25 are shown in bold in SEQ ID NO:3. It will be apparent that such a protein has a contiguous mSAV sequence from the N-terminus of its sequence to N23, and a contiguous mSAV sequence from T26 to the C-terminus of the mSAV sequence. It will also be apparent that, notwithstanding an insertion and/or a deletion, the conventional numbering of SAV residues continues to apply to the sequence comprising the insertion/deletion. Accordingly, even when there is an insertion and/or a deletion between, for example, N23 and T26, the N23 and T26 residues retain their customary position assignments for convenience of reference. [0039] In one embodiment, an insertion, such as an insertion between N23 and T26 (with a concomitant deletion of Q24 and L25) comprises 11 amino acids. The insertion may comprise a methionine in the fourth position and a tyrosine in the tenth position (or in the second to last position for a shorter than 11 residue insertion). Thus, in one embodiment, the insertion may comprise the sequence: XXXMXXXXXYX (SEQ ID NO:4), wherein X is any naturally occurring amino acid. The insertion may be configured to facilitate formation of a salt bridge between the third and/or fifth amino acids (numbering from N-terminal to C- terminal residues), such as an R in the third position and an E in the fifth position, with residues 125 and 127, such as E 127 and Rl 25. In one embodiment, a mSAV provided by the invention comprises the following amino acid sequence:

[0040] AEAGITGTWYNXXXMXXXXXYXGSTFIVTAGADGALTGTYESXVGXACSR YXLXGXYDXAPATDGSGTXLXWXVXWKNXCRNAXSXTTWSGQYVGGAEARINTQ WXLXSGTTEANAXKSTXXGXDTFTKV (SEQ ID NO:5).

[0041] SEQ ID NO: 5 comprises an eleven amino acid insertion that replaces Q24 and L25. In various embodiments, the insertion comprises the sequence XXRMXXXXXYX (SEQ ID NO:20), XXXMEXXXXYX (SEQ ID NO:21), and XXXMXXWXXYX (SEQ ID NO:22). In specific embodiments, the insertion can comprise TGTMEGWGD YT (SEQ ID NO:6) or TGTMQGWGDYT (SEQ ID NO:7), or TGTMEGWGD YT (SEQ ID NO:8). In alternative specific embodiments, the sequence can comprise TGRSEGSTYT (SEQ ID NO:23) and TGRSEGGTTYT (SEQ ID NO:24).

[0042] It is considered that the 11 amino acid insertion in SEQ ID NO: 5 inhibits and/or prevents domain swap dimer formation. However, we have also shown that different insertion sequences will also function and will produce high affinity monomers. For example, insertions of TGRSEGSTYT (SEQ ID NO:23) and TGRSEGGTTYT (SEQ ID NO:24) can be used to replace SEQ ID NO:20 in SEQ ID NO:5 and result in high affinity monomers m54 and m56, respectively.

[0043] In one embodiment, a protein of the invention does not contain the insertion, and thus may form domain swap dimers. SEQ ID NO: 9 provides an example of such a sequence: AEAGITGTWYNQLGSTFIVTAGADGALTGTYESXVGXACSRYXLXGXYDXAPATD GSGTXLXWXVXWKNXCRNAXSXTTWSGQYVGGAE ARINTQWXLXSGTTEANAX KSTXVGHDTFTKV (SEQ ID NO:9). [0044] In one embodiment, a protein provided by the invention comprises a mutation selected from the group consisting of V55T, T76R, L109T, V125R, and combinations thereof. When all four of these mutations are present, the protein is designated a "QM" protein. In certain embodiments, proteins having at least 95% homology to a protein sequence presented herein contain the QM mutation. In alternative embodiments, only 3, 2, 1 or none of the mutations of which the QM mutation is comprised are present in a protein of the invention.

[0045] In certain embodiments, the mSAV disclosed herein comprises a mutation selected from A46S, V47L, the QM mutation, and combinations thereof

[0046] In various embodiments, a mutant streptavidin protein provided by the invention comprises a mutation selected from the group consisting of A46S, V47L, L 125 V, and combinations thereof.

The proteins of the invention may further comprise any one, or any combination of the following modifications of SAV: a disulfide between residues 57 and 76 (i.e. T57C and T76C mutations); 8 amino acid insertion with Tyr at position 7 (each of the insertions described here can be between N23 and T26 and can include a concomitant deletion of Q24 and L25); 9 amino acid insertion with Tyr at position 8; 10 amino acid insertion with Tyr at position 9; 11 amino acid insertion with Tyr at position 10; 12 amino acid insertion with Tyr at position 11; mutants containing a salt bridge between strand 1 and strand 8 by introducing complementary mutations at H 127 and +3 position (i.e. 3^rd amino acid) of an insertion, such as an R; mutants containing an additional salt bridge between strand 1 and 8 by introducing complementary mutations at V 125 and +5 position (i.e. 5^th inserted amino acid), such as an E; mutants containing methionine in the fourth position of an insertion; mutants containing an 11 amino acid insertion with W at position 7; mutant containing a salt bridge between residues 89 and 111 (i.e., A89R and TI l IE); mutant containing a salt bridge between residues 59 and 74 (i.e., R59 and G74E); mutant containing a mutation at G74 that replaces the residue with a larger side chain in order to prevent dimerization, e.g. G74E; mutant containing H87P; mutant containing S93R; mutant containing T91E; mutant containing L109Y; mutant containing V55Y; mutant containing N82E; mutant containing A46S; mutant containing A72T; mutant containing A78N; mutant containing V125C; mutant containing V125R, mutant containing N49Q; mutant containing T55Y; mutant containing S62R; mutant containing W120A; mutant containing L124K; mutant containing L124D; mutant containing L124C, mutant containing H127E, and mutants containing any combination of the foregoing.

[0047] In one embodiment, a protein provided by the invention does not include a substitution of Thr76 with a charged amino acid.

[0048] The present invention in various embodiments provides the mSAV proteins comprising the following amino acid sequences; letter and number combinations (i.e., "m54") signify alternative designations for the proteins comprising the recited amino acid sequences.

m54 (high affinity monomer)

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Arg Ser GIu GIy GIy Thr Thr Tyr Thr GIy Ser Thr Phe lie VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg lie Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr Lys Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO: 10)

m56 (high affinity monomer)

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Arg Ser GIu GIy Ser Thr Tyr Thr GIy Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr Lys Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO: 11)

m57 (can form a domain swapped dimer)

GIy He Thr GIy Thr Trp Tyr Asn GIn LeuGly Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr Lys Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID

NO: 12)

m60 (does not form domain swapped dimer)

GIy He Thr GIy Thr Trp Tyr Asn GIn LeuGly Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr Asp Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO:13)

M44f52c

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Thr MetGlu GIy Trp GIy Asp Tyr Thr GIy Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Trp Lys Ser Thr Lys Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO: 14)

M44f52c-A120

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Thr MetGlu GIy Trp GIy Asp Tyr Thr GIy Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr Lys Arg GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO: 15)

M62

GIy He Thr GIy Thr Trp Tyr Asn GIn LeuGly Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ser VaI GIy GIn Ala Cys Ser Arg Tyr Tyr LeuCys GIy Arg Tyr Asp Arg Ala Pro Ala Thr Asp GIy Ser GIy Thr Thr LeuGlu Trp Cys VaI Asn Trp Lys Asn GIu Cys Arg Asn Ala Pro Ser Arg Thr Arg Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg lie Asn Thr GIn Trp Tyr LeuGlu Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr LeuCys GIy GIu Asp Thr Phe Thr Lys VaI (SEQ ID NO:16)

M44FQmC

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Thr MetGlu GIy Trp GIy Asp Tyr Thr GIy Ser Thr Phe lie VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ala VaI GIy Asn Ala GIu Ser Arg Tyr Thr LeuThr GIy Arg Tyr Asp Ser Ala Pro Ala Thr Asp GIy Ser GIy Thr Ala LeuGly Trp Arg VaI Ala Trp Lys Asn Asn Tyr Arg Asn Ala HisSer Ala Thr Thr Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg lie Asn Thr GIn Trp Thr LeuThr Ser GIy Thr Thr GIu Ala Asn Ala Trp Lys Ser Thr LeuArg GIy HisAsp Thr Phe Thr Lys VaI (SEQ ID NO: 17)

M44FQmC-A120

GIy He Thr GIy Thr Trp Tyr Asn Thr GIy Thr MetGlu GIy Trp GIy Asp Tyr Thr GIy Ser Thr Phe He VaI Thr Ala GIy Ala Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ala VaI GIy Asn Ala GIu Ser Arg Tyr Thr LeuThr GIy Arg Tyr Asp Ser Ala Pro Ala Thr Asp GIy Ser GIy Thr Ala LeuGly Trp Arg VaI Ala Trp Lys Asn Asn Tyr Arg Asn Ala HisSer Ala Thr Thr Trp Ser GIy GIn Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp Thr LeuThr Ser GIy Thr Thr GIu Ala Asn Ala Ala Lys Ser Thr LeuArg GIy HisAsp Thr Phe Thr Lys VaI (SEQ ID NO: 18 ).

R76/C125

GIy He Thr GIy Thr Trp Tyr Asn GIn LeuGly Ser Thr Phe He VaI Thr Ala GIy Ala

Asp GIy Ala LeuThr GIy Thr Tyr GIu Ser Ala VaI GIy Asn Ala GIu Ser Arg Tyr VaI

LeuThr GIy Arg Tyr Asp Ser Ala Pro Ala Thr Asp GIy Ser GIy Thr Ala LeuGly Trp

Arg VaI Ala Trp Lys Asn Asn Tyr Arg Asn Ala HisSer Ala Thr Thr Trp Ser GIy GIn

Tyr VaI GIy GIy Ala GIu Ala Arg He Asn Thr GIn Trp LeuLeuThr Ser GIy Thr Thr

GIu Ala Asn Ala Trp Lys Ser Thr LeuCys GIy HisAsp Thr Phe Thr Lys VaI (SEQ ID NO: 19)

[0049] Proteins provided by the invention can be made using any suitable method. However, functional purification and proper storage of purified streptavidin are important to avoid loss of activity. We showed that the addition of 10 - 25 % glycerol during the purification increases the yield by several folds. Currently, we get approximately 2 mg of purified protein from 250ml of bacterial culture. Without it, the yield is approximately 100 ug of purified protein from the identical culture. More importantly, the purified protein loses its activity slowly over a period of 1 wk when stored at 4 ⁰C, and almost completely if frozen. In contrast, the purified protein can be stored in 25% glycerol at -20 ⁰C without freezing or frozen at -80 ⁰C with no apparent loss of activity. Three rounds of freeze thaw cycles using liquid nitrogen had no effect on the biotin binding function.

[0050] The following techniques were used to produce and analyze the mutant streptavidin proteins described herein.

[0051] Cell induction: Streptavidin mutant constructs were cloned into pRSET A vector were transformed into BL21 (DE3) pLysS E. coli and plated on Luria Broth (LB)-Ampicillin (Amp). A 10 mL LB-Amp liquid culture (100 ug/mL ampicillin) was grown overnight starting from a single colony with stirring at 300 rpm at 37oC. The following day, the overnight culture was diluted 100-fold into 25OmL LB-Amp and grown to OD600= 0.9-1.0. The cells were induced with 0.4 mM of isopropyl-β-D-thiogalactopyranoside (IPTG) and were harvested after 3 hours at 37oC. The cells pellet was stored in -80oC prior to the purification.

[0052] Protein purification from the inclusion bodies: To lyse the cells, the cell pellet was resuspended in 4mL of bacterial protein extraction reagent (B-PER) with 2mM of phenylmethylsulfonylfluoride (PMSF) and transfered into a 8OmL centrifuge tube. The mixture was then vortexed for 1 minute and centrifuged at 12,000 rpm for 20 minutes at 4oC. The supernatant was discarded and the inclusion bodies were resuspended in 4mL of B-PER. 17OuL of lysozyme (10 mg/mL) was added to the mixture and incubated at room temperature for 5 min. 17mL of inclusion bodies washing buffer (5OmM Tris-HCl at pH 8.0, 10OmM NaCl, 0.5% triton x-100) was added into the mixture and vortexed for 1 minute to obtain a homogenous mixture. The mixture was then centrifuged at 12,000 rpm for 20 minutes at 4oC. The pellet was washed in 17mL of inclusion body wash buffer at least twice to achieve high purity. Finally, the inclusion body was resuspended in 1.5mL of denaturing buffer (6M guanidine hydrochloride, 5OmM Tris-HCl at pH 7.0, 15OmM NaCl), centrifuged at 12000 rpm for 20 minutes at 4oC. The denatured and solubilized protein was incubated with 25OuL of nickel resin for 1 hour at 4oC. (The resin was kept in suspension by periodic pipetting). After one hour, the unbound fraction was removed and the resin was washed twice with denaturing buffer. The protein was finally eluted off the resin with 50OuL of denaturing buffer containing 1OmM DTT.

[0053] Protein refolding by dialysis: The elution fractions containing the protein was transferred into a dialysis column with membrane cut-off of 1OkDa and dialyzed against a 30OmL buffer containing decreasing amounts of guanidine hydrochloride/ 5OmM Tris-HCl, pH7/ 15OmM NaCl/ 10% glycerol. Each dialysis step lasted for 6 hours before the GuHCl concentration was reduced stepwise by a fixed amount. After the GuHCl concentration has reached 0.25M, the buffer volume was increased to IL and the glycerol content was increased 25% to remove the remaining guanidine hydrochloride. The refolded protein was transferred to a microcentrifuge tube and clarified at 14000 rpm for 5 minutes at 4oC. The soluble fraction was stored at -20oC prior to gel electrophoresis analysis and biochemical/biophysical experiments.

[0054] The invention also provides compositions comprising polynucleotide sequences encoding each and every protein described herein. The polynucleotide sequence may be present in an expression vector or other types of vectors, such as shuttle vectors, and may be used to express the proteins in any suitable cell type, including but not necessarily limited to bacterial, yeast, insect and mammalian cells. The polynucleotides may be provided in a composition comprising a cell culture. The polynucleotides may also be provided with other reagents useful for expressing and/or purifying the proteins from the cell culture or for using the proteins encoded by the polynucleotides in any of a variety of assays and/or other processes. The polynucleotides may be linked to one or more cloning sites, such as a polycloning site, which would facilitate synthesis of fusion proteins comprising a protein of the invention fused to any desired amino acid sequence. The invention accordingly provides a protein purification system that enables production and selective purification of a protein that is encoded in the same reading frame as a protein of the invention. For example, the fusion proteins can be isolated and/or purified to any desired degree of purity using a biotin resin/column. On a large scale, a systematic proteomics study can be designed to examine many and potentially all proteins expressed in an organism.

[0055] mSAV can be used to label/immobilize biotinylated compounds without causing aggregation of the targeted molecule. Aggregation of the targeted molecules is a concern if wild type streptavidin is used as a labeling reagent. For example, adding wild type tetrameric streptavidin to cells expressing biotinylated cell surface proteins induces the proteins to aggregate because the lipid bilayer is fluid and allows the proteins to laterally diffuse. A solution to this problem involves monovalent streptavidin, in which three binding sites have been mutated to prevent biotin binding. Preparing the reagent, however, is difficult because it is a multi-step process involving recombinant expression, denaturation, renaturation and purification, thus significantly adding to the total labor. Thus, the mSAV described herein provides an improved composition for labeling and/or immobilizing biotinylated compounds.

[0056] In one embodiment, the invention provides a method for isolating a biotinylated molecule, examples of which include but are not limited to biotinylated nucleic acids, carbohydrates, lipids, peptides and polypeptides. The biotynlated molecules are also considered to include molecules conjugated to biotin analogues and biotin derivatives such as iminobiotin and desthiobiotin. The method comprises providing a composition comprising biotinylated molecules, contacting the biotinylated molecules with mutant streptavidin protein provided by the invention, allowing formation of mutant streptavidin-biotinylated molecule complexes, and separating the complexes from other components of the composition. By mutant streptavidin-biotinylated molecule "complexe" it is meant that the mutant streptavidin is bound to the biotin moiety of the biotinylated molecule. Either the mutant streptavidin protein or the biotinylated molecules can be attached to a substrate, such as a resin, beads, a column, etc. If the biotinylated molecules are attached to a substrate, the method can function to isolate the mutant streptavidin molecules from a composition, which has particular value, for example, when screening fusion proteins comprising a mutant streptavidin protein of the invention wherein the non-mutant streptavidin portion of the fusion protein comprises the amino acid sequence of a protein of interest.

[0057] The invention also provides complexes comprising a mutant streptavidin protein of the invention and a biotinylated molecule.

Cited references: Denny WA. Hypoxia-activated prodrugs in cancer therapy: progress to the clinic. Future

Oncol 6(3) :419-28. Gonzalez M, Argarana CE, Fidelio GD. 1999. Extremely high thermal stability of streptavidin and avidin upon biotin binding. Biomol Eng 16(l-4):67-72. Pahler A, Hendrickson WA, Kolks MA, Argarana CE, Cantor CR. 1987. Characterization and crystallization of core streptavidin. J Biol Chem 262(29): 13933-7. Sano T, Pandori MW, Chen X, Smith CL, Cantor CR. 1995. Recombinant core streptavidins.

A minimum-sized core streptavidin has enhanced structural stability and higher accessibility to biotinylated macromolecules. J Biol Chem 270(47):28204-9. Tredan O, Garbens AB, Lalani AS, Tannock IF. 2009. The hypoxia-activated ProDrug AQ4N penetrates deeply in tumor tissues and complements the limited distribution of mitoxantrone. Cancer Res 69(3):940-7. Wu SC, Wong SL. 2005. Engineering soluble monomeric streptavidin with reversible biotin binding capability. J Biol Chem 280(24):23225-31.

Claims

We claim:

1. A mutated streptavidin protein comprising the amino acid sequence of SEQ ID NO:5 or SEQ ID NO:8.

2. The mutated streptavin protein of claim 1 comprising the sequence of SEQ ID NO:5, wherein SEQ ID NO:5 comprises a mutation selected from the group consisting of T57C, T76C, G74E, H87P, L109Y; V55Y, N82E, A46S, A72T, A78N, V125C, V125R, N49Q, T55Y, S62R, W120A, L124K, L124D, L124C, H127E, and combinations thereof.

3. The mutated streptavidin protein of claim 1 comprising the sequence of SEQ ID NO: 8 comprising a mutation selected from the group consisting of T57C, T76C, R59, G74E, H87P, L109Y; V55Y, N82E, A46S, A72T, A78N, V125C, V125R, N49Q, T55Y, S62R, W120A, L124K, L124D, L124C, H127E, and combinations thereof.

4. The mutated streptavidin protein of claim 1 comprising the sequence of SEQ ID NO:5, wherein the XXXMXXXXXYX (SEQ ID NO:4) portion of SEQ ID NO:5 sequence comprises a sequence selected from the group of sequences consisting of XXRMXXXXXYX (SEQ ID NO: 19), XXXMEXXXXYX (SEQ ID NO:20), and XXXMXXWXXYX (SEQ ID

NO:21).

5. The mutated streptavidin of claim 1 comprising the sequence of SEQ ID NO:5, wherein the XXXMXXXXXYX (SEQ ID NO:4) portion of SEQ ID NO:5 sequence comprises a sequence selected from the group of sequences consisting of TGTMEGWGD YT (SEQ ID NO:6), TGTMQGWGD YT (SEQ ID NO:7), and TGTMEGWGD YT (SEQ ID NO:8).

6. The mutated streptavidin of claim 1 comprising the sequence of SEQ ID NO:5, wherein the XXXMXXXXXYX (SEQ ID NO:4) portion of SEQ ID NO:5 is replaced by TGRSEGSTYT (SEQ ID NO:23) or TGRSEGGTTYT (SEQ ID NO:24).

7. The mutated streptavidin of claim 6, wherein the mutated streptavidin comprises SEQ ID NO: 23 and SEQ ID NO: 10, or SEQ ID NO:24 and SEQ ID NO:11.

8. The mutated streptavidin of claim 1 comprising the sequence of SEQ ID NO:5, wherein the mutated streptavidin comprises a sequence selected from the group of sequences consisting of SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO:16, SEQ ID NO:17, SEQ ID N0:18 and SEQ ID N0:19.

9. The mutated streptavidin protein of claim 1 , wherein the mutated streptavidin protein is part of a fusion protein.

10. The mutated streptavidin of claim 9, wherein the mutated streptavidin protein is present in a complex with a biotinylated molecule.

11. A method of isolating a biotinylated molecule comprising providing a composition comprising biotinylated molecules, contacting the biotinylated molecules with a mutant streptavidin protein of claim 1, allowing formation of mutant streptavidin-biotinylated molecule complexes, and separating the complexes from other components of the composition.

12. The method of claim 11 , wherein the mutant streptavidin protein or the biotinylated target molecules are attached to a substrate.

13. The method of claim 11 , wherein the mutant streptavidin protein is present in a fusion protein.

14. A composition comprising a mutant streptavidin protein of claim 1 and a biotinylated molecule.

15. The composition of claim 14, wherein the mutant streptavidin protein and the biotinylated molecule are present in a complex.