- Split View
-
Views
-
Cite
Cite
Bernard Hallet, David J. Sherratt, Finbarr Hayes, Pentapeptide scanning mutagenesis: random insertion of a variable five amino acid cassette in a target protein, Nucleic Acids Research, Volume 25, Issue 9, 1 May 1997, Pages 1866–1867, https://doi.org/10.1093/nar/25.9.1866
- Share Icon Share
Abstract
Anew insertion method for probing protein functional organization was developed. The method relies on the random insertion of transposon Tn4430 and subsequent in vitro deletion of the bulk of the transposon after which a 15 bp insertion remains within the target gene. This results in pentapeptide insertions randomly distributed in the target protein. Characterization of 23 pentapeptide insertions in TEM-1 β-lactamase demonstrated the utility of the method. The phenotypes associated with the mutated β-lactamase proteins equated both with the sorts of local peptide structures in which the pentapeptide insertions occurred and their position in the three-dimensional structure of the enzyme.
Linker insertion mutagenesis can provide invaluable insights into protein structure-function relationships. However, most linker insertion mutagenesis procedures are technically demanding and are limited to insertions at preexisting restriction enzyme sites (1). In this study, a simple and efficient transposon-based linker insertion mutagenesis strategy is described.
Transposon Tn4430, a Tn3-related transposon from Bacillus thuringiensis, transposes efficiently in Escherichia coli and duplicates 5 bp of host sequences at the insertion point (2, this study). The transposon also contains KpnI restriction enzyme sites 5 bp from the outer ends of its terminal inverted repeats. Following transposition of Tn4430 to a target gene, a 15-bp in-frame insertion remains if the bulk of the transposon is deleted by restriction with KpnI and religation (Fig. 1). This results in a five amino acid insertion in the protein encoded by the target gene. As the sequence of the insertions is dependent on the 10 bp derived from Tn4430 and on the sequence of the target site pentanucleotide duplication, the pentapeptide insertions will vary in composition (Fig. 1). No wild-type amino acids are simultaneously mutated.
The pentapeptide insertion mutagenesis strategy utilizes the formation of a cointegrate as a transposition intermediate by Tn4430. Plasmid pHT385 consists of the conjugative F derivative pOX38 into which Tn4430Ω5, a Tn4430 derivative encoding kanamycin resistance, has been introduced (2). A donor strain which contains both pHT385 and the plasmid harbouring the target gene is mated with a plasmid-free recipient strain. By selecting for appropriate markers on the target plasmid and the recipient strain, transconjugants containing cointegrates between pHT385 and the target plasmid are isolated. These cointegrates are resolved rapidly in vivo resulting in target plasmid derivatives harbouring a copy of Tn4430Ω5 (2). Independent mating experiments are performed to isolate different insertions into the target gene. Isolates in which Tn4430Ω5 has inserted in the gene of interest are identified by an appropriate genetic screen and/or by restriction enzyme mapping. The bulk of Tn4430Ω5 subsequently is deleted by digestion with KpnI and religation.
The utility of the above strategy was tested using the TEM-1 β-lactamase protein encoded by the bla gene on pBR322. Thirty independent insertions of Tn4430Ω5 were generated in bla. The positions of the 15-bp insertions remaining after KpnI deletion of the bulk of the transposon were determined by sequencing. This analysis revealed that Tn4430Ω5 insertion was essentially random although five positions in bla acted as target sites on more than one occasion resulting in 23 unique insertion positions among the 30 clones characterized. The low insertion specificity of Tn4430 is typical of members of the Tn3 family (3).
The effects of the 23 different pentapeptide insertions on β-lactamase activity were assessed by determining the maximum ampicillin concentrations allowing growth conferred by these proteins at 30°C. The wild-type bla gene provided resistance to 5500 µg/ml of ampicillin whereas the host strain was sensitive to 10 µg/ml of this antibiotic. The set of mutated proteins was divisible into three categories: (i) five of the mutated β-lactamases conferred high levels of ampicillin resistance (≥2500 µg/ml); (ii) eight of the pentapeptide insertions abolished detectable β-lactamase activity; and (iii) the remaining 10 insertions resulted in intermediate levels of ampicillin resistance (50–1500 µg/ml). The pentapeptide random insertion mutagenesis procedure therefore can generate mutated proteins with a range of activity levels.
As linker insertions potentially confer thermosensitive properties on the mutated proteins (1), the resistance levels conferred by the mutated β-lactamase proteins were also analyzed at 37 and 42°C. All of the high-level and intermediate-level resistance proteins conferred decreased resistance at 37°C which was further reduced to varying levels at 42°C (data not shown). The wild-type β-lactamase protein itself provides less resistance at higher temperatures but the pentapeptide mutagenesis procedure may be an efficient technique for generating proteins with enhanced thermosensitive characteristics.
TEM-1 β-lactamase has an ellipsoidal structure consisting of two globular domains connected by two hinge regions. The depression lying at the interface of the two domains forms the catalytic cavity (4,5). Mapping of the pentapeptide insertions on this structure illustrated the usefulness of the pentapeptide scanning mutagenesis method as the phenotypes conferred by the mutated β-lactamases correlated both with the types of local location in the three-dimensional organization of the enzyme (Fig. 2). Insertions which abolished resistance to ampicillin were clustered around the catalytic cavity of the enzyme whereas insertions in other regions of the protein conferred a range of resistance levels which presumably reflect a range of structural defects. Pentapeptide insertions in the five mutated β-lactamase proteins which retained high levels of ampicillin resistance mapped to two protruding loops which are distant from the catalytic site (Fig. 2). Insertions which conferred intermediate levels of ampicillin resistance were found either in different regions of secondary structure which are not directly involved in the substrate binding cavity, in one of two hinge regions of the protein, or in a loop whose C-terminus forms the left border of the catalytic site (4,5). Pentapeptide insertions in α-helices and β-strands conferred a higher level of thermosensitivity than insertions in loops.
Pentapeptide scanning mutagenesis is a simple and rapid procedure which requires a single bacterial mating experiment followed by standard in vitro plasmid DNA digestion and religation manipulations. The only technical limitation of the method is the necessity for not having a KpnI site within the target coding sequence. However, in many instances, undesirable KpnI restriction enzyme sites could readily be removed by site-directed mutagenesis before employing the pentapeptide insertion procedure. The method promises to be especially useful in probing structure-function relationships in multifunctional proteins since insertions in some regions of a protein may specifically alter different protein activities whereas positions highly tolerant to pentapeptide insertions may pinpoint inter-domain regions. Indeed, application of the method to the XerD site-specific recombinase shows that a range of protein function alterations can readily be obtained (Y.Cao, B.Hallet, D.J.Sherratt and F.Hayes, unpublished data). The pentapeptide scanning mutagenesis procedure has the advantage of providing a unique KpnI restriction enzyme site at the insertion point within the mutated gene. This not only facilitates mapping of the mutations but also potentially allows further manipulations, e.g., expression of protein subdomains, construction of protein fusions, or the introduction of useful peptides such as protein tags or specific epitopes.
Acknowledgements
B.H. and F.H. contributed equally to this work which was funded by a Wellcome Trust Research Career Development Fellowship (no. 040822/Z/94/Z/1.4U/AWB/CS) to F.H. B.H. was supported by a European Molecular Biology Organization Long-Term Fellowship. We thank François Cornet, Keith Dyke, Jean-Marie Frère, Rene Rezsöhazy and Andrew Spiers for constructive criticism of the manuscript, Didier Lereclus for kindly providing pHT385 and for helpful discussions, and David Levinger for assistance with the MOLSCRIPT program.
Comments