Patent Application
20060281899
The present invention relates to new mutants of biotin binding proteins with improved properties compared to wild type proteins. Proteins, which bind biotin include chicken avidin and bacterial streptavidin, other poultry avidins, such as avidin proteins isolated from the duck, goose, ostrich and turkey, and chicken avidin-related proteins (AVRs).
Chicken avidin and bacterial streptavidin are widely utilized proteins in many life science applications ranging from purification techniques to modern diagnostics and targeted drug delivery. This methodology, known as (strept)avidin-biotin technology, relies on the extremely tight and specific affinity (Kd˜10−15 M) between (strept)avidin and biotin.
Avidin and streptavidin are exceptionally stable proteins consisting of homo-tetrameric up-and-down β-barrels, which upon biotin binding become even more stable. Transition midpoints of heat denaturation (Tm), analyzed by differential scanning calorimetry (DSC), have shown that avidin is more stable than streptavidin in both the absence and presence of biotin (Gonzalez, M., Argarana, C. E., & Fidelio, G. D., Biomol. Eng. 16, 67-72 (1999). A possible reason for this may be the intramonomeric disulphide bridge found in each avidin monomer. Wild-type (wt) avidin has a high isoelectric point and it is glycosylated, which properties may cause unspecific binding in some applications. Streptavidin is not glycosylated and lacks cysteine residues capable of forming disulfide bridges. It has been shown, that the unwanted properties of avidin can be abolished without markedly affecting the tight biotin-binding affinity and the stability characteristics of avidin (Marttila, A. T. et al., FEBS Lett 467 (2000), 31-6). The avidinand streptavidintetramer is actually a dimer of two dimers. The monomers that form the tetramer interact with each other in a symmetrical manner and form the three types of monomer-monomer interactions described in detail by Livnah (Livnah, O., Bayer, E. A., Wilcheck, M., & Sussman, J. L., Proceedings of The National Academy of Sciences of the United States of America 90 (1993), 5076-5080). The interaction between monomers 1-4 (and the equivalent 2-3) is the strongest, while the interaction between monomers 1-3 (and 2-4) is the weakest. The intensity of the interaction between monomers 1-2 (and 3-4) is also relatively weak, but important because tryptophan 110 in avidin (Trp120 in streptavidin) from subunit 1 participates in biotin binding at the binding-site of subunit 2, forming the function-related monomer-monomer interface.
The 3-D structures of avidin and streptavidin have been solved by x-ray crystallography and their tertiary and quaternary structures show astonishing similarity despite relatively low amino acid sequence identity (Weber, P. C., Ohlendorf, D. H., Wendoloski, J. J. and Salemme, F. R. (1989) Science 243, 85-88; Livnah, O., Bayer, E. A., Wilcheck, M. and Sussman, J. L. (1993) Proc. Natl. Acad. Sci. 90, 5076-5080; Pugliese, L., Coda, A., Malcovati, M. and Bolognesi, M. (1993) J. Mol. Biol. 231, 698-710). Most of the essential biotin-binding residues are conserved, and the affinity for biotin is nearly identical in both avidin and streptavidin. In addition to high biotin-binding capacity, avidin and streptavidin tetramers show remarkable stability at high temperatures as well as under strongly denaturing conditions. The stability increases even more upon biotin binding (Green, M. (1990) Methods Enzymol. 184, 51-67; Gonzalez, M., Argarana, C. E. and Fidelio, G. D. (1999) Biomol. Eng. 16, 67-72).
A streptavidin mutant with enhanced stability characteristics when compared to those of the wild-type streptavidin has been reported previously (Reznik, G. O. et al., Nat Biotechnol 14 (1996), 1007-11). The improved stability was achieved by addition of two intermonomeric disulphide bridges to the streptavidin tetramer, one between monomers 1-3 and the other between monomers 2-4, by changing histidine residue 127 to cysteine. In avidin these interfaces are similar, and histidine residue 127 of streptavidin is analogous to isoleusine residue 117 of avidin.
The avidin gene (AVD) in chicken has homologues, called avidin-related genes (AVRs) (Keinänen, R. A., Laukkanen, M. L. and Kulomaa, M. S. (1988) J. Steroid Biochem. 30, 17-21;. Keinänen, R. A., Wallen, M. J., Kristo, P. A., Laukkanen, M. O., Toimela, T. A., Helenius, M. A. and Kulomaa, M. S. (1994) Eur. J. Biochem. 220, 615-21; Ahlroth, M. K., Kola, E. H., Ewald, D., Masabanda, J., Sazanov, A., Fries, R. and Kulomaa, M. S. (2000) Anim. Genet. 31, 367-75).
Seven different genes, AVR1-AVR7, have been cloned and sequenced. Two of the genes, AVR4 and AVR5, are 100% identical in their coding sequence, exhibiting only a single nucleotide difference in their 5′-flanking region, whereas the others are 94-99% identical to each other. The identity between AVD and the different AVRs ranges from 91 to 95%. Messenger RNAs for AVR1, AVR2 and AVR3 have been detected in chicken under inflammatory conditions (Kunnas, T. A., Wallen, M. J. and Kulomaa, M. S. (1993) Biochim. Biophys. Acta 1216, 441-5), but it is not known whether they are expressed as proteins. The putative avidin-related proteins (AVRs), as deduced from their nucleotide sequences, are 74-81% identical to avidin and 85-100% identical to each other.
Avidin proteins, isolated from the duck, goose, ostrich and turkey, showed similar tetrameric structure, glycosylation and stability both against temperature and proteolytic activity of proteinase K as chicken avidin. The biotin-binding properties of these avidins, measured using IAsys optical biosensor, were similar to those found in avidin from the chicken. Three of these avidins, however, showed different immunological crossreactivity when compared to chicken avidin. The patient sera response to duck, goose and ostrich avidin was also lower than those observed for chicken and turkey avidins. These proteins could offer advantages over the use of chicken avidin and bacterial streptavidin in pretargeting applications (Hytönen et al., Biochem. J. (2003) 372; 219-225).
Because of the similarity of the structure of the biotin binding proteins including chicken avidin and bacterial streptavidin, other poultry avidins, such as avidin proteins isolated from the duck, goose, ostrich and turkey, and chicken avidin-related proteins (AVRs), it is presumed that new mutants of these biotin binding proteins may be created by site-directed mutagenesis of the amino acids of the sequences and that the resulting mutants have similar properties.
The object of the present invention is to provide new mutants of biotin binding proteins with improved properties compared to wild type proteins. The term biotin binding proteins is understood to include chicken avidin and bacterial streptavidin, other poultry avidins, such as avidin proteins isolated from the duck, goose, ostrich and turkey, and chicken avidin-related proteins (AVRs).
This invention provides thermally stabilized biotin binding proteins using site-directed mutation. For avidin and AVRs this was achieved by introducing disulphide bridges between its subunits. These covalent bonds had no major effects on the biotin-binding properties of the respective mutants. Moreover, one of the mutants maintained its tetrameric integrity even in denaturing conditions. These new avidin forms have native→denatured transition midpoints (Tm) close to 100° C. in the absence of biotin; for AVR4/5 Tm is close to 110° C. Furthermore, it is shown that the intramonomeric disulphide bridges found in wild-type avidin have effects on its stability.
A mutant form, in which this bridge was removed, had a lower Tm in the absence of biotin than wild-type avidin, but showed comparable stability in the presence of biotin.
In the present invention the stability of avidin was improved through the addition of intermonomeric disulphide bridges by changing isoleucine residue 117 to cysteine. In addition two supplementary mutants were produced where even more intermonomeric disulphide bridges were introduced. A cysteineless avidin version was also constructed, where the intramonomeric disulphide bridges of wild-type avidin were removed. The stability characteristics of these mutants were studied and compared with those of wild-type avidin and streptavidin.
Site-directed mutagenesis may also be introduced in other biotin binding proteins including bacterial streptavidin, other poultry avidins, and chicken avidin-related proteins (AVRs) in order to alter their properties similarly to avidin. The site-directed mutagenesis may also be a deletion of an amino acid.
The only AVR that exhibited affinity to biotin and 2-iminobiotin comparable to avidin was AVR4/5. Nonetheless, AVR4/5 showed other interesting differences, i.e. an “extra” cysteine residue in every subunit and a different glycosylation pattern, when compared with avidin.
Site-directed mutagenesis of AVR4/5 cDNA was performed by mutation of the asparagine 43 of AVR4/5 to glutamic acid (AVR4/5(N43E)) in order to avoid possible glycosylation at that position. Cysteine 124 in AVR4/5 was converted to serine to eliminate the expected extraneous disulfide bridge in AVR4/5. The AVR4/5(C124S) was further mutagenized by changing the tyrosine 117 to cysteine (AVR4/5(Y117C)). The cysteine 124 in AVR4/5 could be converted to any amino acid. (The numbering of the amino acids are according to avidin (see
Suprisingly it was found that AVR4/5 displays the most durable heat stability characteristics of all biotin-binding proteins so far characterized. AVR4/5 exhibited a Tm that evidently exceeds the boiling point of water even in the absence of ligand. Due to its high thermal stability, AVR4/5 or some of its derivatives may prove extremely useful in applications where high heat durability is required.
u Untreated with enzyme
t Treated with enzyme
4/5=AVR4/5
43=AVR4/5(N43E)
43+=AVR4/5(N43E) treated with Endoglycosidase Hf
A=avidin (Belovo)
A+=avidin treated with Endoglycosidase Hf
B) Non-reducing SDS-PAGE. Samples were first acetylated in vitro, boiled for 20 minutes and subjected to 15% SDS-PAGE. The arrow shows the presence of the dimeric form of AVR4/5. Values on the left represent bands for the molecular weight standards (Bio-Rad).
A=avidin
4/5=AVR4/5
124=AVR4/5(C124S)
C) Non-reducing SDS-PAGE. Samples were first acetylated in vitro, boiled for 20 minutes and subjected to 15% SDS-PAGE. The arrow shows the presence of the dimeric form of AVR4/5(Y117C,C124S). Values on the left represent bands for the molecular weight standards (Bio-Rad).
Avidin (SEQ ID NO 1) and streptavidin (SEQ ID NO 2) are valuable and widely used tools in the life sciences. In addition to their high biotin-binding affinity, the robustness and the flexibility of the system relies on their extreme stability under various demanding conditions. In the present application one object is to increase the stability of chicken avidin and other biotin binding proteins even further without losing its strong biotin-binding ability. In order to achieve this goal two, four or six intermonomeric disulphide bridges are introduced to avidin. The same could also be done to avidin related proteins. In addition, the role of the intramonomeric disulphide bridges in the stability of the avidin tetramer was determined, by removing them using site-directed mutagenesis.
According to the DSC results, both the intramonomeric and intermonomeric disulphide bridges have effects on the heat-induced denaturation of avidin. Removal of the intramonomeric disulphide bridges from avidin (Avd-nc) caused a decrease in its Tm in the absence of biotin (ΔTm=−8.9° C.), whereas in the presence of biotin the Tm was virtually the same as that of the wild-type avidin. This underlines the importance of the biotin-induced increase in the stability of avidin. Interestingly, the cysteineless mutant Avd-nc has a slightly higher Tm than that of streptavidin as an apoform and significantly higher as a complex with biotin. Further, avidin has naturally higher Tm than streptavidin.
The most stable of the mutated avidins is Avd-ci, which has an intermonomeric disulphide bridge in monomer interfaces 1-3 and 2-4. In contrast, the disulphide bridges created in interfaces 1-4 and 2-3 do not improve the thermal stability of avidin. Instead these mutations cause a decrease in the Tm of the mutant Avd-cci when compared to the wild-type avidin or even to the cysteineless mutant Avd-nc. This effect was more prominent in the absence of biotin. The high thermal stability of the combined mutant Avd-ccci indicated that the stabilizing effect of the disulphide bridge in interface 1-3 (and 2-4) exceeded the decrease caused by the disulphide bridges in interface 1-4 (and 2-3).
In the presence of denaturing agents it may be preferable to use the mutants Avd-ci, Avd-cci and Avd-ccci instead of wild-type avidin. This may particularly be the case for Avd-ccci since all of its monomers are covalently linked to each other (directly or indirectly). The mutant is capable of remaining as a tetramer even after unfolding, as seen in the SDS-PAGE analysis (
In most applications of biotin binding technology high affinity biotin binding is essential. Consequently, all of the mutants introduced in this application are promising candidates for use with the methodology, as they all showed irreversible biotin binding and high affinity towards 2-iminobiotin. None of the mutations involved residues directly responsible for biotin binding. However, cysteine 106 that substituted isoleucine 106 in mutants Avd-cci and Avd-ccci, is located in the same loop (between β-strands 7 and 8) as the important biotin-binding residue tryptophan 110. The disulphide bridge formed between Cys106 of subunit 1 and Cys86 of subunit 4 may directly or indirectly have an influence on Trp110, and thereby also affect the biotin binding ability of these avidin mutants.
It was also found that the streptavidin tetramer is unstable in certain conditions without its bound ligand. This could be detrimental in many applications due to the loss of signal along with the analytical molecules. Furthermore, it would presumably shorten the life span of materials if they were to contain covalently linked streptavidin molecules. Chilkoti as well as Reznik have described a streptavidin mutant, His127Cys (Chilkoti, A., Schwartz, B. L., Smith, R. D., Long, C. J., and Stayton, P. S. (1995) Bio/Technology 13, 1198-1204; Reznik, G. O. et al., Nat Biotechnol 14 (1996), 1007-11). This mutant has an intermonomeric disulphide bridge between subunits 1 and 3 (2 and 4), which is analogous to the isoleucine 117 to cysteine substitution in our avidin mutant Avd-ci. They reported that the resultant mutant was more stable than wild type streptavidin (Reznik, G. O. et al., Nat Biotechnol 14 (1996), 1007-11. Avidin is, however, more stable than streptavidin, as judged by DSC analysis (Gonzalez, M. et al., Biomol Eng 16 (1999) 67-72). Therefore, the 1-3 (and 2-4) intermonomeric disulphide bridges bearing mutant Avd-ci would also have a higher Tm than the corresponding streptavidin mutant.
The production of streptavidin (or its mutants) is usually performed as inclusion bodies in E. coli. Its downstream processing includes laborious and time-consuming denaturation and renaturation procedures followed by the contrived formation of disulphide bridges (as in the case of the His127Cys mutant) and additional purification steps. In contrast, according to the present invention, the production of avidin and other biotin binding mutants, even those with intermonomeric disulphide bridges, yielded soluble proteins in insect cells that were easily purified in a single step by affinity chromatography on a 2-iminobiotin column.
The avidin mutants and other avidin related protein mutants include a signal sequence and thus are introduced in a eucaryotic insect cell to a special path, where building up of the multiple disulphide bridges is possible as part of the process. The mutants are thus soluble and functional immediately after this process. In this special path the bridging is enhanced by i.a. PDI (Protein Disulphide Isomerase) enzymes. On the contrary the streptavidin mutants are produced in an inactive form in procaryotic bacteria (U.S. Pat. No. 6,022,951). Streptavidin has not been succesfully produced to a special path in an insect cell/baculovirus system, which system has now successfully been used in avidin and avidin mutant production. Avidin also has considerably higher solubility than streptavidin. Previously it was thought that because avidin already has one natural intramonomeric disulfide bridge, introducing new intermonomeric disulphide bridges would cause incorrect pairing leading to low quality or inactive protein in the cells. Streptavidin does not have a natural cysteine and thus this would not be a problem in streptavidin production.
Avidin and streptavidin are totally different proteins even though they have a similar structure. The mutant avidin Avd-ccci differs from the known avidins in that all the subunits are covalently attached to each other. It includes in addition to I117C also changes where intermonomeric bridges are formed by amino acids in different positions of the subunits, i.e. cysteine 86 pairs with cysteine 106 and vice versa (1-4 monomers and 2-3 monomers) and thus differs from the I117C mutant (or H127C streptavidin mutant), wherein the same amino acid in the pairing subunits are bridged in the disulphide bridging. Several disulphide bridges have not been introduced in a streptavidin tetramer than could be achieved by H127C, and such streptavidin mutant could be non-functional and presumably very heterogeneous. When the side chains of isoleucines 117 in each avidin tetramer subunit are substituted with cysteines they form the 1-3 (and 2-4) disulphide bridges in mutants Avd-ci and Avd-ccci. Each subunit has isoleucine 106 and aspartate 86, in mutants Avd-cci and Avd-ccci these are substituted with cysteines in order to form 1-4 (and 2-3) intermonomeric disulphide bridges.
The gene encoding avidin (AVD) has seven homologues in chicken, named avidin-related genes (AVRs, AVR1-AVR7 (SEQ ID NOs 3-9)). Since it was not known if the avidin-related or AVR genes, are expressed as proteins in chicken, their cDNAs were recently cloned and they were produced using a baculovirus-insect cell expression system (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J 363, 609-617) Two of the genes, AVR4 and AVR5, are 100% identical in their coding sequence, exhibiting only a single nucleotide difference in their 5′-flanking region, whereas the others are 94-99% identical to each other. The identity between AVD and the different AVRs ranges from 91 to 95% (Keinanen, R. A., Wallen, M. J., Kristo, P. A., Laukkanen, M. O., Toimela, T. A., Helenius, M. A. and Kulomaa, M. S. (1994) Eur. J. Biochem. 220, 615-21; Ahlroth, M. K., Kola, E. H., Ewald, D., Masabanda, J., Sazanov, A., Fries, R. and Kulomaa, M. S. (2000) Anim. Genet. 31, 367-75). The putative avidin-related proteins (AVRs), as deduced from their nucleotide sequences, are 74-81% identical to avidin and 85-100% identical to each other.
Amino acid sequence analysis and molecular modelling are used to predict and explain the properties of the AVRs. It was found that the AVR proteins are highly similar to avidin both structurally and functionally. Despite the numerous amino acid substitutions in the subunit interface regions, the AVRs form extremely stable tetramers similar to avidin.
Throughout the present application the sequences of avidin and AVRs are aligned so that the amino acids 55 (threonine) and 56 (isoleucine) of the avidin sequence which do not exist in the sequence of the AVRs, are counted in the AVR sequence (see
Differences are found in some physico-chemical properties of the AVRs as compared to avidin, including lowered isoelectric point, increased glycosylation and, most notably, reversible biotin binding for two AVRs (AVR1 and AVR2). Molecular modelling shows how the replacement Lys-111-Ile in AVR2 alters the shape of the biotin-binding pocket and thus results in reversible binding. Biochemical analyses show that disulfide bonds can form and link monomers in AVR4/5, a property not found in avidin.
Despite the relatively high primary sequence conservation (
All of the pI-mutants, as well as the non-glycosylated avidins, bind biotin with high affinity and form stable tetramers. Therefore, differences in pI or the lack of glycosylation at positions corresponding to residue 17 in avidin is not expected to affect biotin binding and tetramer formation of the AVRs. Consequently, the observed differences must be based on yet unknown features that, while affecting biotin binding, do not impede tetramer stability.
Another structurally interesting feature of the AVRs (except for AVR2) is the third cysteine residue. In addition to AVR4/5, this may at least partially explain the properties of AVRs 1, 3, 6 and 7. Their inability to bind 2-iminobiotin in IASyS analyses may reflect the presence of such structural differences as compared to avidin.
In conclusion, the recombinant AVRs are functional and show interesting physico-chemical properties. These new biotin binders may provide advantages over avidin and streptavidin in several applications. For example, AVR2 could be used in affinity purification protocols that require mild elution conditions.
Further because of the similarity of the sequences of avidin and AVRs, the site-directed mutations performed to the avidin sequence in order to create thermostabile mutant avidin may be performed in a similar way for the AVRs thus creating intermonomeric disulphide bridges in the tetramer. The tetramer may have one disulphide bridge between monomers 1-3 and one disulphide bridge between monomers 2-4 as in Avd-ci, two disulphide bridges between monomers 1-4 and two disulphide bridges between monomers 2-3 as in Avd-cci, or one disulphide bridge between monomers 1-3 and one disulphide bridge between monomers 2-4 and two disulphide bridges between monomers 1-4 and two disulphide bridges between monomers 2-3 as in Avd-ccci.
The AVRs AVR1 and AVR2 have asparagine 117 instead of the isoleucine 117 in avidin, which is changed to cysteine in these mutants, and in AVR3-7 the amino acid residue 117 is tyrosine, which is changed to cysteine. The isoleucine residue 106 and aspartate residue 86 are the same in the avidin and the AVRs (see
AVR4/5 is the closest relative to the avidin gene in the family, and its chromosomal location is closest to the avidin gene (Ahlroth, M. K., Grapputo, A., Laitinen, O. H. & Kulomaa, M. S. (2001) Biochem. Biophys. Res. Commun. 285, 734-741). It is the only AVR that exhibited affinity to biotin and 2-iminobiotin comparable to avidin. The most obvious structural differences would arise from the diverse patterns of glycosylation, and the extraneous cysteine residue found in most AVRs that is not present in avidin. In addition, residues in contact at the subunit interfaces of the model structure showed significant differences when compared to the avidin structure. However, according to the analysis of stability by SDS-PAGE (Bayer, E. A., Ehrlich-Rogozinski, S. & Wilchek, M. (1996) Electrophoresis 17, 1319-1324), these differences at the interfaces did not reduce the stability of the AVRs.
In the present study, we have carried out a more detailed examination of the structural and stability properties of the AVR4/5 protein using comprehensive biochemical and mutational analyses. A surprising observation of the present invention was the superior heat stability of AVR4/5 and its mutants when compared to avidin or streptavidin (Gonzalez, M., Argarana, C. E. & Fidelio, G. D. (1999) Biomol. Eng. 16, 67-72), and to the thermally improved avidin forms, too (Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483). In the latter case, the improvements in the stability were achieved by introducing disulfide bridges between avidin subunits. In contrast, the disulfide bridge due to the extraneous Cys-124 of AVR4/5 is not connected to the enhanced stability, since AVR4/5(C124S) was observed to be even more stable than AVR4/5. According to ITC, DSC and optical biosensor data, the biotin and 2-iminobiotin binding properties of AVR4/5 resemble those of avidin. The measured affinity towards biotin (Kd≈3.6×10−14 M) is weaker than that for avidin, but this interaction is still one of the tightest measured for ligand-protein-interactions. It actually resembles values measured between biotin and bacterial streptavidin (Kd≈4.0×10−14 M) (Green, N. M. (1990) Methods Enzymol. 184, 51-67). The thermodynamic explanation for the weaker affinity of AVR4/5 towards biotin (as compared to avidin) seems to lie in the entropy change on binding. Due to its high thermal stability, AVR4/5 or some of its derivatives may prove extremely useful in applications where high heat durability is required.
The modeling results suggest that possible reasons for the improved thermal stability include both a more stable tertiary structure of the AVR4/5 β-barrel as well as more favorable subunit interactions of the tetramer. The structure of avidin may be more flexible and in a more open conformation to accept biotin entry into the binding pocket than the potentially more rigid AVR4/5 barrel. This might also explain the higher affinity of avidin towards biotin and 2-iminobiotin. The potential salt bridge between Asp-39 and Arg-114 in AVR4/5 could contribute to the observed slower association rate of AVR4/5 for 2-iminobiotin, and it may also stabilize the monomer structure. Furthermore, the interaction of Asp-39 with the positively charged Lys-111 from the neighboring subunit may enhance the stability of the tetramer. Another contribution to the slower association rate may be the presence of the Pro-41-Gly-42 dipeptide in AVR4/5 that could form a conformation that partially blocks entry of biotin into the binding-pocket.
One important factor for the enhanced stability is probably connected to the deletion of two residues in loop 4 of AVR4/5, corresponding to residues Thr-55 and Ile-56 in avidin. This deletion may eliminate the unfavorable interaction seen in avidin. In general, comparison of the crystal structure of related proteins have shown that shorter loops are less mobile and presumably more stable (Usher, K. C., de la Cruz, A. F., Dahlquist, F. W., Swanson, R. V., Simon, M. I. & Remington, S. J. (1998) Protein Sci. 7, 403-412; Davail, S., Feller, G., Narinx, E. & Gerday, C. (1994) J. Biol. Chem. 269, 17448-17453; Thompson, M. J. & Eisenberg, D. (1999) J. Mol. Biol. 290, 595-604). In addition, the Ser-41 to proline conversion in loop 3 may add rigidity to the secondary and tertiary structure of AVR4/5. The replacement in AVR4/5 of glycine residues, present in avidin, may have a further stabilizing effect, since glycine is potentially more mobile and has more freedom to adopt torsion angles not permitted for other amino acids (Liu, H. L., Doleyres, Y., Coutinho, P. M., Ford, C. & Reilly, P. J. (2000) Protein Eng. 13, 655-659; Shibuya, H., Kaneko, S. & Hayashi, K. (2000) Biochem. J. 349, 651-656).
Finally, the replacement of Ile-117 in avidin by tyrosine in AVR4/5 should increase the stability of the 1-3 (and 2-4) dimer interface in comparison with avidin. Furthermore, the stabilizing effect might also extend to subunits 2 and 4 (1 and 3), which are in close proximity to the 1-3 (2-4) dimer interface (
Although the predicted disulfide bridges seemed to have no effect on the thermal stability of AVR4/5, they had a clear influence on the overall structure of the protein. In contrast to the previous hypothesis (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617), the FPLC data suggested that these disulfide bridges connect AVR4/5 tetramers together, thus forming larger oligo-tetrameric structures instead of linking the monomers within the same tetramer (
Based on the sequence, AVR4/5 contains three potential N-glycolysation sites. In a previous study it seemed that AVR4/5 was heavily glycosylated, but it was impossible to judge if all possible sites were utilized (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617). Mutational analysis showed that glycosylation occurs at least at Asn-43 (
By constructing and analyzing the AVR4/5(Y117C,C124S), it was shown that it is possible to apply the mutation(s) described in (Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483) to AVR4/5(C124S). The obtained protein AVR4/5(l 17C,C124S) is a stable tetramer including covalent bonds between subunits. This kind of protein could offer even better properties in applications, where high resistance against environmental factors is needed. One potential use may be, for example, in surface-based applications, where biotin-binding protein is attached to a matrix and the subunit dissociation is harmful event.
The behaviour of AVR4/5-protein was changed significantly due to mutation Cys124->Ser. AVR4/5 forms oligomers of tetramers via intertetrameric disulphide bonds. AVR4/5(C124S) behave like a native avidin with the high stability of AVR4/5. The obtained protein AVR4/5(C124S) was found to be even more stable than original one (AVR4/5). This protein (AVR4/5(C124S)) may found to be useful in many applications in avidin-biotin technology.
It is shown in this application that the intramonomeric disulphide bridges of wild-type avidin seem to be an important factor in making avidin so thermostable. On the other hand, it is possible to enhance the high thermostability of avidin and other biotin binding proteins further by the introduction of intermonomeric disulphide bridges. The most thermostable avidin mutants and other biotin binding protein mutants described in this application provide more stable tools for biotin binding technology, suitable also for new kinds of applications.
The invention will be further described with reference to the following non-limiting examples.
Design of the Avidin Mutants
Mutations were designed by using the sequence and structure information obtained from analyses with GCG (Genetics Computer Group, Madison, Wis.), EMBOSS (European Molecular Biology Open Software Suite), WHAT IF (Vriend, G., J. Mol. Graph. 8 (1990), 52-6, 29) and InsightII (Molecular Simulations Inc., San Diego, Calif.) programs. Genetic engineering of the coding sequence of avidin was performed by megaprimer (Sarkar, G. & Sommer, S. S., Biotechniques 8 (1990), 404-407) and QuikChange (Stratagene) methods, by using oligonucleotide primers containing the desired mutations.
The avidin mutant Avd-nc (C4A, C83Y) was constructed to obtain information about the importance of the intrinsic disulphide bridges to the overall stability of the avidin tetramer. According to the sequence alignment with streptavidin, the cysteine residues were substituted with the same residues that streptavidin bears in the analogous positions in its primary structure (Table 1). The avidin mutant Avd-cci (D86C, I106C) adopted also the cysteines using the evolutionary approach (Laitinen, O. H. et al., FEBS Lett 461 (1999), 52-8). The sequence information came from the avidin-like domain of the sea urchin fibropellins (Hunt, L. T. & Barker, W. C., Faseb 3 (1989), 1760-1764; Delgadillo-Reynoso, M. G. et al., J Mol Evol 29 (1989), 314-27; Bisgrove, B. W. et al., Journal of Molecular Evolution 41 (1995), 34-45; Bisgrove, B. W. & Raff, R. A., Dev Biol 157 (1993), 526-38. In the case of Avd-cci, two intermonomeric disulphide bridges, designed to form between monomers 1-4 and 2-3 were introduced, thereby constituting a total of four new disulphide bridges per tetramer (
Avd-ci (I117C) has an extra cysteine residue in each subunit of the tetramer. It was designed according to the mutational strategy that Chilkoti (Chilkoti, A. et al., Bio/Technology 13 (1995), 1198-1204) and Reznik (Reznik, G. O. et al., Nat Biotechnol 14 (1996), 1007-11) used to stabilize the streptavidin tetramer. In Avd-ci, cysteine 117 from subunit 1 faces cysteine 117 from the neighboring subunit 3. Identically, the corresponding subunit interface 2-4 of the wild-type protein contains two equivalent isoleucine residues, which were replaced by cysteines. Therefore, two intermonomeric disulphide bridges in the avidin tetramer were expected to form between subunits 1-3 and 2-4, intensifying the firm association between dimers 1-4 and 2-3 by the addition of two covalent bonds. Finally, in order to introduce six intermonomeric disulphide bridges into the avidin tetramer, Avd-ccci, a combination of mutants Avd-ci and Avd-cci, was constructed.
Production, Purification and Characterization of Mutant Avidins
All the mutants were produced by a baculovirus expression system (Bac-To-Bac, Gibco BRL, Life Technologies, Gaithersburg, Md., USA) in the infected insect cells and purified by affinity chromatography on 2-iminobiotin agarose as previously described in detail by Airenne (Airenne, K. J. et al., Protein Expression and Purification 9(1997), 100-108) and Laitinen (Laitinen, O. H. et al., Biochem J. 363 (2002), 609-17).
Wild-type avidin was purified from chicken egg-white. Using a Vibra cellTM sonicator, the egg-white was sonicated for 3 minutes on ice at power setting 8 and 50% duty cycle with a one-minute break between bursts. After sonication the sample was diluted with two volumes of PBS and centrifuged (20 minutes, 20.000 g, 4° C.). The soluble fraction was further purified by affinity chromatography on 2-iminobiotin agarose as previously reported (Laitinen, O. H. et al., J Biol Chem 276 (2001), 8219-24). Protein samples were concentrated and subjected to change of buffers with Centricon YM-3 filters (Millipore, cat. no. 4202). The SDS-PAGE analysis was performed with a sample buffer without β-mercaptoethanol.
All the mutants showed excellent purification efficiency, indicating that the mutations had no major effects on the 2-iminobiotin-binding properties. Moreover, the mutants showed irreversible biotin-binding properties indistinguishable from that of wild-type avidin as measured with an IAsys optical biosensor (data not shown). Affinities towards 2-iminobiotin were determined for wild-type avidin and the four mutants (Table 1). The results indicated that the mutations had not significantly altered the 2-iminobiotin binding-characteristics of the mutants. The formation of the intermonomeric disulphide bridges was studied with SDS-PAGE analysis (
Biotin-Binding Assays for Avidin and the Mutants
Reversibility of biotin binding was measured for avidin and the mutants with an IAsys optical biosensor (Laitinen, O. H. et al., FEBS Lett 461 (1999), 52-8). Protein samples were allowed to bind to a biotin-aminosilane cuvette in PBS containing 1 M NaCl. After the equilibrium was reached, biotin-containing buffer was added and the dissociation of the proteins was monitored. The affinities of the proteins towards 2-iminobiotin were determined with an IAsys optical biosensor (Marttila, A. T. et al. FEBS Lett 441 (1998), 313-7). Biotin-binding activity of the mutants Avd-ci and Avd-ccci after heat treatment was studied with a microtiter plate assay. Protein samples, 5 μg/ml concentration in PBS, were heated for various time periods at 99.9° C. and then chilled on ice. The samples were transferred to a Nunc Maxisorp-plate and incubated at 37° C. for 2 h. The wells were washed three times with PBS-Tween (0.05% v/v). After that the wells were blocked with PBS containing 1% BSA at 37° C. for 30 minutes and washed again with PBS-Tween. Next, biotinylated alkaline phosphatase (Sigma) in PBS, 1% BSA was added and incubated at 37° C. for one hour. Once again, the wells were washed with PBS-Tween and the PNPP substrate in DEA buffer (1 mg/ml) was applied to the wells. The absorbances at 405 nm were measured after incubation for 45 minutes.
Differential Scanning Calorimetry of the Avidin Mutants
In order to study the thermal stability of the purified avidin mutants, they were subjected to analysis by differential scanning calorimetry. A Nano II differential scanning calorimeter (Calorimetric Science Corporation, Provo, Utah) was used. The protein sample concentrations were 0.032 mM (given as the monomer concentration) and the biotin-containing samples had a biotin:avidin molar ratio of 3:1. The reference cell was filled with the same buffer that the sample proteins were dissolved in (100 mM Na-phosphate buffer pH 7.4). The thermograms were recorded as a function of temperature, between 25 and 130° C. at a temperature scan rate of 55.6° C./h. The baselines were subtracted and the transition midpoints (Tm) of the samples were calculated using proprietary software provided by the instrument manufacturer.
The results are shown in a baseline-subtracted form (
Sequence Comparison and Molecular Modelling of AVR Proteins
The EMBL database accession numbers for the AVR genes used to deduce the corresponding amino acid sequences are as follows: AVR1, Z21611; AVR2, Z21554; AVR3, Z21612; AVR4, Z22883, AVR6, AJ237658; AVR7, AJ237659. AVR4 represents also the identical AVR5, and we therefore use here the term AVR4/5. The AVR cDNAs were translated with the GCG software package program Map (Genetic Computer Group, Madison, Wis., USA). The theoretical pI for the AVRs and avidin was determined by using the GCG-package program Peptidesort. The comparison of the sequences of avidin and AVR1-AVR7 was performed using Malign (Johnson, M. S. and Overington, J. P. (1993) J. Mol. Biol. 233, 716-38; Johnson, M. S., May, A. C., Rodionov, M. A. and Overington, J. P. (1996) Methods Enzymol. 266, 575-98) in the BODIL Modelling Environment (Lehtonen, Rantanen, Still and Johnson, unpublished). Model structures of AVR1-AVR7 were made using the rapid homology modelling program HOMODGE in BODIL on the basis of the x-ray structure of chicken avidin in complex with biotin (PDB code: lavd; (Pugliese, L., Coda, A., Malcovati, M. and Bolognesi, M. (1993) J. Mol. Biol. 231, 698-710). HOMODGE aims to reduce errors in models of closely related proteins by using as many atomic coordinates from the known structure as possible to build the conserved main chain and side chain structures. Where the side-chains differ, HOMODGE uses a rotamer library to select the side chain conformation in order to optimise contacts.
The putative AVRs were compared to avidin to identify differences that might affect their biotin binding, structural or other biochemical properties. The biotin-binding residues are well conserved in all AVR proteins, showing only three amino acid changes (out of the 16 possible) (
The theoretical pIs of AVR3 and AVR4/5 are basic, similar to the pI of avidin (avidin 10.4, AVR3 10.2 and AVR4/5 10.0). On the other hand, AVR1, AVR6 and AVR7 are neutral (pI 7.3) and AVR2 is calculated to be acidic with a pI of 4.7. All AVR proteins, except for AVR2, have three cysteine residues (
The interface regions between different subunits show a variable number of amino acid substitutions between the AVR proteins and avidin. The interface between subunits 1 and 2 (as well as between 3 and 4) is perfectly conserved in all of the AVRs. In contrast, two out of three interface residues between subunits 1 and 3 (2 and 4) show changes in all of the AVR proteins, except for AVR4/5, which shows only one substitution (Ile-117-Tyr). Interestingly, the tightly interacting 1-4 (2-3) subunit interface shows several amino acid substitutions in all of the AVRs. In avidin, a total of 22 residues confer intersubunit contacts within this interface. The AVR proteins have nine substitutions in this region, seven of which are found in all AVRs and two in all AVRs but AVR4/5 (Table 3).
Amino acids, which in avidin participate in intersubunit interactions but are substituted in AVRs.
aMain-chain contact
bOxygen atom of main chain
cSide-chain contact
dVan der Waals contacts
eNumber of hydrogen bonds
Most of the sequence differences in the subunit interface of avidin and AVR1-AVR7 would not interfere with tetramer formation. The most critical sequence difference is located at position 96 where methionine in avidin and AVR4/5 is replaced by lysine in AVR1-AVR3 and AVR6-AVR7. In avidin and AVR4/5, Met-96 interacts with Met-96 from a second monomer of the tetramer (
As compared to avidin, the biotin-binding pocket is highly conserved in AVR1-AVR7, where only minor changes in the sequences and therefore in the shape of the binding pocket are observed. These changes are limited to the area surrounding the most flexible part of the biotin molecule and, thus, biotin can easily compensate for these changes. The only exception occurs at position 111. In avidin, AVR1 and AVR3-AVR7 lysine is found at this position, whereas in AVR2 isoleucine is present. In the avidin crystal structure, the hydrophobic part of the lysine side-chain interacts with the indole-ring of Trp-110 keeping its position fixed (
All AVRs, with the exception of AVR2, contain a third single cysteine (in addition to the two present in avidin) located on the surface of the proteins. Since AVR2 does not have the extra cysteine, there is no reason to believe that AVR2 could form tetramers in a way that is different to avidin. However, in AVR1, AVR3, AVR6 and AVR7, the extra cysteine is located in L4 at position 60 (avidin numbering), where it is impossible to form an extra disulfide-bond between subunits without severely disturbing tetramer formation.
Production and Purification of Recombinant AVR Proteins
The AVR1-4/5 genes (Keinänen, R. A., Laukkanen, M. L. and Kulomaa, M. S. (1988) J. Steroid Biochem. 30, 17-21; Keinänen, R. A., Wallen, M. J., Kristo, P. A., Laukkanen, M. O., Toimela, T. A., Helenius, M. A. and Kulomaa, M. S. (1994) Eur. J. Biochem. 220, 615-21) were cloned into the EcoRI-HindIII sites of pGEM4 and were in vitro transcribed and spliced using Ribomax and RNA Splicing System kits (Promega) according to the manufacturer's instructions. The cDNAs were then produced by RT-PCR and further amplified by PCR using the oligonucleotide primers: AK33 (5′-CTGCTAGATCTATGGTGCACGCAACCTCCCC-3′) and AK44 (5′-GTTGCAAGCTTTGCGGGGCCATCCT-3′) containing BglII and HindII restriction sites, respectively. After cutting with BglII and HindIII, the AVR1-4/5 cDNAs were cloned into BamHI and HindIII sites of pFASTBAC. For AVR6 and AVR7, cDNAs were produced by subcloning the corresponding genes (Ahlroth, M. K., Kola, E. H., Ewald, D., Masabanda, J., Sazanov, A., Fries, R. and Kulomaa, M. S. (2000) Anim. Genet. 31, 367-75) into pDsRed1 (Clontech) where the red fluorescent protein-encoding region had been removed, and by subsequent transfection of the constructs into NIH/3T3 cells. Total RNA was extracted from the cells using the SV Total RNA Isolation System (Promega). The AVR6 and AVR7 cDNAs were produced by RT-PCR (RobusT RT-PCR Kit, Finnzymes, Espoo, Finland) using the oligonucleotide primers AK33 and AK44 to produce BglII and HindIII restriction sites to the 5′- and 3′-ends of the cDNAs, respectively. The synthesised cDNAs were cloned into BamHI/HindIII digested pFASTBAC1, the cloning vector for the Bac-To-Bac Baculovirus Expression System (Gibco BRL, Life Technologies, Gaithersburg, Md., USA). The nucleotide sequences of the cDNAs were confirmed by sequencing. The virus vectors for producing the AVR proteins were constructed and amplified according to the Bac-To-Bac system instructions. Recombinant AVR proteins were produced in Sf9 insect cells as previously reported (Airenne, K. J., Oker-Blom, C., Marjomäki, V. S., Bayer, E. A., Wilchek, M. and Kulomaa, M. S. (1997) Protein Expr. Purif. 9, 100-10822). Proteins were purified from the cells using affinity chromatography on a 2-iminobiotin (AVR1, AVR3, AVR4/5-7) and/or biotin agarose column (AVR1 and AVR2) as previously described (Laitinen, O. H., Airenne, K. J., Marttila, A. T., Kulik, T., Porkka, E., Bayer, E. A., Wilchek, M. and Kulomaa, M. S. (1999) FEBS Lett. 461, 52-8). AVR2 was eluted from biotin agarose with 1 M acetic acid and the elution fractions were immediately neutralised with NaOH. The elution of AVR1 was achieved using 1 M HCl followed by neutralisation with Tris (1 g/ml).
All AVRs showed different patterns in SDS-PAGE as compared to avidin (
Biotin-Binding Analyses of AVRs
The biotin-binding characteristics of the AVR proteins were studied as previously described using the IASyS optical biosensor (Marttila, A. T., Airenne, K. J., Laitinen, O. H., Kulik, T., Bayer, E. A ., Wilchek, M. and Kulomaa, M. S. (1998) FEBS Lett. 441, 313-317; Laitinen, O. H., Airenne, K. J., Marttila, A. T., Kulik, T., Porkka, E., Bayer, E. A., Wilchek, M. and Kulomaa, M. S. (1999) FEBS Lett. 461, 52-8). Affinities were measured for 2-iminobiotin or biotin at 20.0° C. using a stirring power of 100%. Briefly, the protein under study in biotin-free buffer was allowed to bind to the biotin-coated surface of the IASyS cuvette. As a negative control, the protein under study was saturated with excess biotin prior to (and was also maintained under biotin excess upon) addition to the cuvette. Under these conditions, residual binding to the biotin surface of the cuvette could be regarded as unspecific, and was used as a measure of baseline variation. Binding equilibrium in the experimental cuvette was considered to be attained after no binding was detected as compared to negative control.
The reversibility of biotin binding was studied as follows: Avidin and the AVRs were allowed to bind to a biotin-coated cuvette. After measuring the initial binding (A) as described above, biotin-saturated (0.17 mg/ml) buffer was injected into the cuvette and the amount of proteins remaining bound was measured until no dissociation was seen, again as compared to the negative control cuvette (B). The measurement time was at least 20 minutes. The percent reversibility was calculated as follows: reversibility (%)=100*(A−B)/A. A−B is the amount of liberated proteins after biotin adding.
In the reversibility assay, AVRs 3-7 showed totally irreversible biotin binding, similarly to avidin. In contrast, AVR1 exhibited 18% and AVR2 93% reversibility following addition of free biotin (Table 4). The actual dissociation constants (Kd) were calculated for the AVRs according to the kass and kdiss rates measured in a separate assay in different protein concentrations (varying between 10 μM and 100 pM) (Table 4 and
Biotin-binding properties of AVRs compared to avidin (AVD), determined using the IASyS biosensor. The values for avidin are from Laitinen et al. (Laitinen, O. H. et al., Biochem. J 363, 609-17 (2002)) and Marttila et al. (Marttila, A. T. et al., FEBS Lett 467, 31-6 (2000)). Reversibility of binding was determined by using a biotin cuvette.
aMeasured in a 2-iminobiotin cuvette.
bMeasured in a biotin cuvette.
cThe dissociation constants were calculated from the equilibrium response data.
dThe dissociation constants were calculated directly from the binding curves.
ND Not determined.
Structural Analyses of AVRs
The heat stability of the AVRs was studied by using an SDS-PAGE-based method (Bayer, E. A., Ehrlich-Rogozinski, S. and Wilchek, M. (1996) Electrophoresis 17, 1319-1324). In this method, each protein sample was divided into two fractions one of which was kept biotin-free and the other was supplemented with an excess of biotin. The samples were then mixed with denaturing SDS-PAGE sample buffer (containing β-mercaptoethanol) and incubated at a given temperature for 20 minutes. After heat treatment, the samples were subjected to SDS-PAGE and visualised by staining with Coomassie brilliant blue. The relative proportions of tetrameric and monomeric forms of the AVR proteins (with avidin as the control) were detected. A similar assay, with β-mercaptoethanol omitted from the sample buffer, was performed to examine the presence of intersubunit disulfide-bridges. Sensitivity to proteinase K was studied both in the absence and in the presence of biotin as described (Laitinen, O. H., Airenne, K. J., Marttila, A. T., Kulik, T., Porkka, E., Bayer, E. A., Wilchek, M. and Kulomaa, M. S. (1999) FEBS Lett. 461, 52-8). The isoelectric point of each AVR was determined by isoelectric focusing as previously reported (Marttila, A. T., Airenne, K. J., Laitinen, O. H., Kulik, T., Bayer, E. A., Wilchek, M. and Kulomaa, M. S. (1998) FEBS Lett. 441, 313-317). The glycosylation patterns of the AVR proteins were studied by treating the proteins with Endo Hf and PNGase F glycosidase according to the manufacturer's instructions (New England Biolabs, Mass., USA).
All AVRs exhibited heat stability comparable to avidin in a reducing SDS-PAGE assay (not shown). In the absence of biotin, AVR tetramers began to dissociate into monomers at around 60° C. In the presence of biotin, a portion of AVRs remained tetrameric even upon boiling. Non-reducing SDS-PAGE showed that AVR1 and AVRs 3-7 have a tendency to form dimers. Under reducing conditions, only tetramers and monomers could be discerned. This result suggests that the dimers are crosslinked via disulfide bridges (stable enough to hold the dimers together under non-reducing conditions) and are slowly degraded into monomers by heat (
The AVRs also showed remarkable resistance against proteolysis. In the presence of biotin, all AVRs remained intact after 16 hours of proteinase K treatment. Proteolysis occurred, albeit slowly, in the absence of biotin. Taken together, AVRs showed stability similar to or even greater than avidin (not shown).
The multiple bands observed for the AVRs in SDS-PAGE were found to be differently glycosylated forms. Treatment with Endo Hf glycosidase and PNGase F eliminated the higher-molecular weight bands. The number of putative glycosylation sites seems to correlate well with the observed glycosylation patterns. Preliminary deglycosylation results using non-denaturing conditions suggest that N-glycan is attached to the glycosylation site in L5 in AVRs 1,2,4/5. Isoelectric focusing showed that the pIs of the AVRs were approximately the same as expected from theoretical calculations; ≈7 for AVRs 1, 6, and 7; ≈5 for AVR2; and ≈10 for AVRs 3, and 4/5.
The relative proportions (as percentage) of different oligomeric forms of AVRs and avidin (AVD) in SDS-PAGE under non-reducing conditions, boiled 20 minutes before loading onto the gel. Note the presence of dimeric forms in the case of AVRs 1, 3-7.
Mutagenesis of AVR4/5 and the Purification of Expressed Recombinant Proteins
Site-directed mutagenesis of cDNA encoding AVR4/5 protein (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617) was performed by using Quik-Change (Stratagene, La Jolla, Calif., USA) mutagenesis method.
Based upon the sequence alignment of AVR4/5 and avidin (
Recombinant baculoviruses producing AVR4/5 and its mutated forms were generated according to instructions of Bac-To-Bac™-system (Gibco BRL, Life Technologies, Gaithersburg, Md., USA). Wild type and mutant AVR4/5 proteins were produced in baculovirus infected Sf9 insect cells in biotin-free medium as reported earlier (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617). The produced proteins were purified with affinity chromatography using 2-iminobiotin agarose as previously described (Laitinen, O. H., Marttila, A. T., Airenne, K. J., Kulik, T., Livnah, O., Bayer, E. A., Wilchek, M. & Kulomaa, M. S. (2001) J. Biol. Chem. 276, 8219-8224).
Briefly, the insect cell infections were carried out with cells that were transferred to a biotin-free medium. Infection was allowed to proceed for 72 hours, and after that the cells were collected by centrifugation (1500 g, 5 min). The cell pellet was resuspended into a lysis buffer (50 mM Tris-HCl, 2 mM EDTA, 150 mM NaCl, 1% Triton X-100, pH 8) and incubated on ice for 30 minutes. After that the cell extract was sonicated and centrifuged (15000 g 20 min) and the pH of the supernatant was adjusted to 11 with 5 M NaOH and NaCl was added to a final concentration of 1 M. Then the cell extract was applied to a 2-iminobiotin agarose column washed previously with binding buffer (50 mM Na-carbonate, 1 M NaCl). Bound proteins were eluted from the column with 50 mM Na-acetate, pH 4.
The concentrations of the avidin solutions were calculated from the absorbance at 280 nm using an extinction coefficient of 24280 M−1 cm−1 for avidin, 26960 M−1 cm−1 for AVR4/5 and AVR4/5(N43E), 26840 M−1 cm−1 for AVR4(C124S) and 25680 M−1 cm−1 for AVR4/5(Y117C, C124S) (Green, N. M. (1975) Adv. Protein Chem. 29, 85-133).
Glycosylation Analysis of AVR 4/5 and its Mutants
The glycosylation patterns of the proteins were studied by treating the proteins with Endo Hf glycosidase according to the manufacturer's instructions (New England Biolabs, MA, USA) as described in (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617). Samples were then analyzed with 15% SDS-PAGE and stained with Coomassie Brilliant Blue. Non-treated protein samples were used as a control.
Treatment with Endo Hf glycosidase caused the disappearance of higher molecular weight bands in the protein samples (
Gel Filtration Analysis of AVR 4/5 and its Mutants
The oligomeric state of the AVR4/5 and its mutant forms were assayed with FPLC gel filtration using Shimadzu HPLC liquid chromatography equipped with a Superdex 200 HR 10/30 column (Amersham Pharmacia Biotech, Uppsala, Sweden) and fluorescence detector (excitation 280 nm, detection 350 nm). The column was calibrated using the gel filtration mixture (thyroglobulin, IgG, ovalbumin, myoglobin, vitamin B-12; Bio-Rad Laboratories, Hercules, Calif., U.S.A) and bovine serum albumin (Roche Diagnostics, Mannheim, Germany) as molecular weight standards. Sodium phosphate buffer (50 mM, pH 7.2) or sodium carbonate buffer (50 mM, pH 11) with 650 mM NaCl was used as the liquid phase. Protein samples varying between 5 and 25 μg in a volume of 20 μl were used in the analysis. The reduction of intertetrameric disulphide bridges were performed by adding 2-mercaptoethanol to final concentration of 10% (v/v) after which the protein sample was incubated 1 h at 37° C.
AVR4/5 and its mutated form N43E appeared to be heterogeneous in gel filtration, since higher molecular weight forms were seen (
aΔTm is the change in Tm upon addition of a three-fold molar excess of biotin.
bΔH value obtained from sample without biotin. The value could not be determined accurately from samples saturated with biotin because the Tm values were too close to the temperature limit of the DSC equipment.
cThree protein forms with lowest molecular weights are indicated.
dNot measured.
eNot determined due to low signal.
fUnpublished results (V. P. Hytönen 2003)
gNordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483.
hGonzalez, M., Argarana, C. E. & Fidelio, G. D. (1999) Biomol. Eng. 16, 67-72.
Non-Reducing SDS-PAGE Analysis of AVR 4/5 and its Mutants
Protein samples in the absence or presence of biotin (0.02 mg/ml) were diluted to a final concentration about 0.2 mg/ml in 50 mM Na-carbonate buffer pH 8.7, and proteins were acetylated by adding NHS-acetic acid conjugate to a final concentration of 0.2 μg/ml. The mixture was incubated at room temperature (23° C.) for 10 min. Equal amount of SDS-PAGE sample buffer without 2-mercaptoethanol was added to the sample and it was boiled for 20 min. After analysis by 15% SDS-PAGE, the gel was stained with Coomassie brilliant blue.
In the SDS-PAGE analysis AVR4/5(Y117C, C124S) (
Optical Biosensor Studies of AVR 4/5 and its Mutants
The biotin-binding characteristics of the different avidins were studied using an IAsys optical biosensor at 20° C. The reversibility of binding and binding affinities to the 2-iminobiotin surface in 50 mM borate buffer (pH 9.5, 1 M NaCl) were measured as previously reported (Laitinen, O. H., Hytönen, V. P., Ahlroth, M. K., Pentikäinen, O. T., Gallagher, C., Nordlund, H. R., Ovod, V., Marttila, A. T., Porkka, E., Heino, S., Johnson, M. S., Airenne, K. J. & Kulomaa, M. S. (2002) Biochem. J. 363, 609-617).
Briefly, The 2-iminobiotin ligand was immobilized onto the carboxymethyldextran surface of a cuvette using N-hydroxysuccinimide activation. Binding of various concentrations of avidin or avidin mutants onto 2-iminobiotinylated surface were measured in a 50 mM borate buffer (pH 9.5) containing 1 M NaCl at 20° C. The iminobiotin cuvettes were regenerated with 20 mM HCl. The kinetic rate constants for association (kass) and dissociation (kdiss) were calculated using the Fast Fit program package (Affinity Sensors). The obtained rate constants were used to calculate Kd(rel) according to Kd=kdiss/kass. Equilibrium data obtained from measurements with different protein concentrations were used to calculate the dissociation constant Kd(eq).
All proteins exhibited in the IAsys interaction analysis strong binding to the biotin surface, similar to that of chicken avidin (Table 7). All AVR4/5 variants showed similar affinities towards the 2-iminobiotin surface (Table 7). These affinities were, however, an order of magnitude lower than that exhibited by avidin for 2-iminobiotin. In all cases, the slower association rate was observed as the reason for the lowered 2-iminobiotin affinities.
aKd detected from the equilibrium response data.
bKd detected directly from the binding curves.
cValues are from (Hytönen, V. P., Laitinen, O. H., Grapputo, A., Kettunen, A., Savolainen, J., Kalkkinen, N., Marttila, A. T., Nordlund, H. R., Nyholm, T. K., Paganelli, G. & Kulomaa, M. S. (2003) Biochem. J. 372, 219-225).
dNordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483.
Differential Scanning Calorimetry
The thermodynamics of the heat-induced unfolding of the proteins were studied using DSC. The transition midpoint of heat-induced denaturation (Tm) of the avidin and different AVR4/5 forms was studied using a Nano II differential scanning calorimeter (Calorimetric Sciences Corporation, Provo, Utah, USA). The reference cell was filled with buffer whereas the sample cell was filled with proteins dissolved in 50 mM sodium phosphate buffer, pH 7.0 containing 100 mM NaCl. Avidin and AVR4/5(C124S) were also analyzed in 50 mM Na-acetate pH 4.0 containing 100 mM NaCl. Furthermore, AVR4/5 was analyzed in the presence of 2-mercaptoethanol (0.4% v/v). The samples (1.6-9.6 μM) were analyzed both in the absence and presence of a three fold molar excess of biotin/subunit. All samples were degassed before being loaded into the DSC. The thermograms were recorded as a function of temperature; between 25 and 130° C. at a rate of 55.6° C./h (Gonzalez, M., Argarana, C. E. & Fidelio, G. D. (1999) Biomol. Eng. 16, 67-72; Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483). Data was analyzed using Microcal Origin 6.0 and Cpcalc 2.1 (Applied Thermodynamics, Hunt Valley, Md.). The change in heat capacity upon unfolding (ΔCp) was measured for avidin and AVR4/5(C124S) using DSC data from measurements in pH 4.0 and 7.0. ΔCp was then obtained from the slope of ΔH vs. Tm (Liu, Y. & Sturtevant, J. M. (1996) Biochemistry 35, 3059-3062).
Biotin was obtained Sigma Chemicals and was used as delivered. A 0.395 mM solution of biotin was prepared in the measurement buffer. The concentration of biotin was determined gravimetrically with a microbalance.
In comparison to wild type avidin, all AVR4/5 proteins unfolded at a markedly higher temperature (Table 6). The enthalpy change ΔH upon unfolding was larger for AVR4/5 and AVR4/5(C124S) as compared to avidin. Due to addition of biotin, the Tm was shifted to a higher temperature for all proteins. None of the AVR4/5 forms were, however, stabilized to the same degree as wild type avidin (ΔTm, Table 6). Treatment with 2-mercaptoethanol caused no significant decrease in the Tm of AVR4/5.
The Tm values of avidin and AVR4/5(C124S) at pH 4 were 76.6±0.2° C. and 103.2±0.5° C., respectively. The change in heat capacity upon unfolding (ΔCp) was rather similar for AVR4/5(C124S) and avidin (Table 6). The ΔCp for avidin was similar to the value reported by Donovan and Ross (Donovan, J. W. & Ross, K. D. (1973) Biochemistry 12, 512-517).
Isothermal Titration Calorimetry
Binding enthalpies (ΔH) where measured with an isothermal titration calorimeter (ITC-4200, Calorimetric Sciences Corporation, Provo, Utah, USA). The ITC was calibrated chemically by titrating aliquots of 1.00 mM HCl into 265 mM Tris dissolved in water. The measurements where performed at 25 and 37° C. with 1.0 ml protein solution (20-25 μM) kept under constant stifling. Biotin (0.395 mM) was titrated into the sample cell in 10 μL aliquots. The heat of dilution, which was small compared to the heat of the binding reaction, was corrected for in the analysis of the results. Analysis of the data was performed using Titration Bindworks 3.0.69 (Applied Thermodynamics, Hunt Valley, Md., USA). ΔCpL was calculated from the slope of ΔH vs. temperature (T).
The interaction between AVR4/5(C124S) and biotin was compared to that between avidin and biotin using ITC. Because AVR4/5(C124S) was found to bind biotin as AVR4/5 and to oligomerize like avidin, it was selected for ITC measurements. As these proteins have a very high affinity for biotin, no binding constant could be calculated from the ITC experiments alone. However, since all ligand is bound in the titration, the binding enthalpy could be calculated from the heat released upon the injection of biotin. The ITC experiments, performed at 25 and 37° C., showed that both the ΔH and ΔCp were similar for biotin binding to AVR4/5(C124S) and avidin (Table 6). The binding enthalpy for biotin binding to avidin was similar to a previously reported value, and the ΔCpL was close to the reported value too (Swamy, M. J. (1995) Biochem. Mol. Biol. Int. 36, 219-225).
The estimated binding constant Kb calculated from a combination of ITC and DSC data showed that the binding of biotin to AVR4/5(C124S) (Kb≈2.8×1013 M−1) was almost two orders of magnitude lower than for avidin (Kb≈9.3×1015 M−1). The difference in how biotin binds to AVR4/5(C124S) and avidin may be interpreted from the thermodynamic data in Table 6. Since ΔH is rather similar for biotin binding to both AVR4/5(C124S) and avidin, the difference in ΔG is expected to be mainly due to an entropic contribution.
Microplate Assay
Biotin-binding activity of the proteins after heat treatment was studied with a microtiter plate assay (Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483). Protein samples, 5 μg/ml in PBS (10 mM Na2HPO4, 1.8 mM KH2PO4, 140 mM NaCl, 2.7 mM KCl, pH 7.2), were heated for various time periods at 99.9° C. (T3 thermocycler, Whatman Biometra, Göttingen, Germany; lid temperature 102° C., slope 0.2° C./sec) and then chilled on ice. The samples were transferred to a 96-well Maxisorp-plate (Nalge Nunc International) and incubated at 37° C. for 2 h. The wells were washed three times with PBS-Tween (0.05% Tween-20 v/v). The wells were then blocked with PBS-BSA (1% w/v) at 37° C. for 30 minutes and washed again with PBS-Tween. Biotinylated alkaline phosphatase (Sigma) in PBS-BSA was added and incubated at 37° C. for one hour. Once again, the wells were washed with PBS-Tween and the para-nitrophenyl phosphate in DEA buffer (Sigma, 1 mg/ml) was applied to the wells. The absorbencies at 405 nm were measured after incubation for 45 minutes.
The microplate assay showed that AVR4/5 and its mutants resist boiling and remain in an active form significantly longer than chicken avidin or even a stabilized avidin AVD-ci with additional intermonomeric disulphide bonds (Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483). Of the studied proteins AVR4/5(C124S) showed the greatest stability, since there was about 30% of the initial binding activity present after boiling for 32 minutes (
Structural Modeling and Sequence Analysis of AVR4/5
The three-dimensional structure of avidin in complex with biotin (PDB access code: lavd;) (Pugliese, L., Coda, A., Malcovati, M. & Bolognesi, M. (1993) J. Mol. Biol. 231, 698-710) was obtained from the Protein Data Bank (PDB). The sequence alignment of avidin and AVR4/5 was made using MALIGN (Johnson, M. S., Overington, J. P. & Blundell, T. L. (1993) J. Mol. Biol. 231, 735-752) in the BODIL Modeling Environment1 using a structure-based sequence comparison matrix (Johnson, M. S., May, A. C., Rodionov, M. A. & Overington, J. P. (1996) Methods Enzymol. 266, 575-598) with gap-penalty 40.
The program HOMODGE in BODIL was used to construct 3D model structures by keeping the side-chain conformations of all identical residues fixed and by maintaining the corresponding torsion angles of similar residues in the alignment. The intramolecular interactions of the amino acids at sequence positions that are different from those in the template structure were analyzed by using the amino acid side-chain rotamer library (Lovell, S. C., Word, J. M., Richardson, J. S. & Richardson, D. C. (2000) Proteins 40, 389-408) incorporated within BODIL.
Molecular modeling and sequence analysis identified amino acid substitutions between avidin and AVR4/5 that might explain the observed differences in the stability and functional properties of AVR4/5. For example, Gly-21 and Gly-107 are found in loops 2 and 7 in avidin, but they are not present in AVR4/5. On the other hand, loop 3 of AVR4/5 contains the Pro-41-Gly-42 sequence that is not present in avidin (
In
Asp-39 in AVR4/5 corresponds to alanine in avidin being located at the interface between subunits 1 and 2. The side chain of alanine in the avidin structure is unable to form any intramolecular interactions, but only lines the ligand-binding pocket (
Microparticle Analysis of AVR4/5
Before the analysis, 3.22 μm microparticles with COOH-group/0.852 nm2 (Bangs laboratories, Fishers, Ind., USA) were covalently coated with biotin-binding proteins. Microparticles (1*108 pcs) were washed twice with H20 by sequential centrifuging (5000 g; 3 min) and aspiration, and resuspended in 200 μl PB buffer (50 mM NaH2PO4-HCl, pH 3.0). A solution of biotin binding protein (50 μl, 1 mg/ml or 3 mg/ml, 50 mM Na-acetate pH 4.0) was added while vortexing. The suspension was incubated for 1 h (500 rpm; 23° C.). Particles were washed with PB and with MES buffer (50 mM, pH 5.5) and suspended in MES buffer (200 μl). A solution of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (Pierce, Rockford, Ill., USA) in MES buffer (200 μl, 1 mg/ml) was added while vortexing and the suspension was incubated for one hour (500 rpm; 23° C.). Particles were then washed once with MES-buffer and four times with assay buffer (50 mM Tris-HCl, 10 mM NaN3, 1 mM EDTA, 0.1% Tween-20, pH 8.0). After the last wash particles were suspended in 100 μl volume and concentration of particles was determined with Multisizer 3 Coulter Counter (Beckman Coulter, Fullerton, Calif., USA). Streptavidin coated mircospheres (3.12 μm diameter) were purchased from Bangs laboratories.
Microparticles (5*106 pcs) were suspended in 900 μl of assay buffer and aliquoted in four PCR-tubes (i-v; 180 μl/aliquot). A solution of BF523-biotin (ArcDia Ltd., Turku, Finland) in N,N-dimethylformamide (Riedel-De-Haen, Seeize, Germany), was diluted with assay buffer to concentration of 200 nM and 20 μl of this solution was added in aliquots (i) and (ii) which were then incubated for 1 h under continuous shaking (22° C., 1100 rpm Thermomixer Comfort, Eppendorf, Hamburg, Germany). The aliquots (ii) and (iv) were then incubated for 30 minutes at 95° C. (PTC-100 equipped with Hot bonnet™ heated lid; M&J Research; Waltham, Mass., USA) while the aliquots (i) and (iii) were kept that time in room temperature. The BF523-biotin solution (20 μl, 200 nM) was added to aliquots (iii) and (iv) after thermal treatment. All aliquots were incubated for 2 h (22° C., 1100 rpm) and transferred in 3 replicates (15 μl) to wells of 384-plate (Greiner Bio-One, Frickenhausen, Germany) and the plate was sealed with adhesive PCR film (Abgene, Surrey, UK). The samples were measured with ArcDia TPX PlateReader (ArcDia Ltd.) (Soini, E., Meltola, N. J., Soini, A. E., Soukka, J., Soini, J. T. & Hanninen, P. E. (2000) Biochem. Soc. Trans. 28, 70-74). In order to improve the precision and accuracy of the results, the individual data points (microparticles) characterized with the focal duration time less that 5 msec were omitted.
The negative control for the assay was prepared by saturating the coated particle with unlabeled biotin (100 pmol/1*106 particles) prior to incubation with BF523-biotin.
The fluorescence emission efficiencies of the BF523-biotin conjugate (200 nM) was determined before and after the incubation (30 min) at 95° C. The measurements were carried out with the ArcDia TPX PlateReader using six sample replicates (15 μl). The obtained 14.1% decrease in probe fluorescence was used throughout the analyses to correct the results.
Characteristics of AVR4/5 and AVR4/5(C124S) were studied by coating plastic microparticles with proteins. The capacity of AVR4/5 coated particles was remarkably higher as compared to commercial streptavidin particles and particles coated with wild-type avidin and high-stability (Tm=98.6° C.) AVD-ci with two intersubunit disulfide bonds (Nordlund, H. R., Laitinen, O. H., Uotila, S. T., Nyholm, T., Hytönen, V. P., Slotte, J. P. & Kulomaa, M. S. (2003) J. Biol. Chem. 278, 2479-2483) (Table 4, i, iii). The specificity of BF523-biotin binding to protein-coated particles was confirmed by saturating the particles with unlabeled biotin prior to analysis.
When particles coated with biotin-binding proteins were heat-treated, AVR4/5 forms showed excellent heat resistance as compared to other proteins (Table 4, iv). AVR4/5 and AVR4/5(C124S) showed 93% and 75% of the original signal after treatment, respectively. The streptavidin particles showed 21% of the original signal after heating. AVD-ci showed higher durability (35%) as compared to wild-type avidin (9%). When the biotin-conjugate was added before heat-treatment (ii), all particles showed better durability than treated alone.
The invention has been illustrated by examples and embodiments, but it may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the invention and all such modifications are intended to be included within the scope of the enclosed claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 20021518 | Aug 2002 | FI | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/FI03/00619 | 8/22/2003 | WO | 11/7/2005 |
