In looking at exist http www apexbt com

In looking at existing FGE substrates from other organisms, there are no reported effects on conversion efficiency of the general XCXPXRX pattern with respect to placement within or at the N′ or C′ terminus of polypeptides, and it has been suggested that the aldehyde formation is independent of the tag location within the protein primary sequence [12]. In contrast, the efficiency of enzymatic oxidation of the cysteine in this motif to formylglycine has been noted as being affected by the interior and flanking amino acids of this short motif [13]. Previous works utilizing site directed mutagenesis of native FGE substrates from various organisms have determined the importance of individual sequence components to show, depending on the species of origin, there were significant differences in specificity [11]. In contrast, we have carried out in this work a directed evolution process for identifying substrate sequences capable of conversion by the formylglycine generating enzyme of M. tuberculosis. In doing so, we confirmed a peptide sequences screening approach for the discovery of substrates of enzymatic reactions such as oxidation of a thiol to an aldehyde in the case of FGE and also, as we have inadvertently found, sites of proteolytic cleavage. In order to identify substrate motifs for FGE from M. tuberculosis, we carried out evolutionary screening of a library of filamentous bacteriophage possessing the XCXPXRX sequence space presented on the p3 protein coat. While a number of combinatorial screening approaches have proven to be effective including the use of peptide arrays [14] and directed evolution via error prone PCR [15], phage-based based evolutionary screening was chosen given that it has shown in the past to be an effective and versatile means of screening short peptide sequences [16], [17]. In our yohimbine hydrochloride of phage based screening, we utilize the ability of effective substrates of FGE to be modified to possess an aldehyde tag to provide population enhancement by applying selection of the FGE modified phage through chemical coupling of their aldehyde bearing p3 protein coat directly onto amine groups present on magnetic beads. Subsequent removal of unbound or non-specifically bound species provided selective enrichment of only the covalently bound phage. The utilization of a linker containing a trypsin cut site between the formylglycine used in covalent attachment and the p3 protein coat allowed for liberation of the phage attached covalently to the magnetic support. In retrospect, the use of a second sequence dependent enzymatic step (i.e., trypsin cleavage) provided an auxiliary selection pressure from which sequences possessing trypsin cut sites were also enriched by preferential elution for subsequent infection into host Escherichia coli. This process proved effective in allowing amplification of phage from the library particularly for those displayed peptides possessing desirable sequences for recognition by FGE. After four rounds of selection, only three of the original 160,000 sequences remained and within these a motif was found which was confirmed to be converted by FGE, albeit it was not a better FGE substrate than present in wild-type sulfatases. We also serendipitously found that this process yielded an FGE substrate which was particularly amenable to cleavage by trypsin. We can anticipate that this adaptable method may be versatile enough for evolutionary screening of other enzymatic substrates, particularly those in which the product may potentially be selectively immobilized or selectively eluted, such as in the case of certain post-translation modifications or proteolytic cleavage reactions.
Materials and methods
Results and discussion The FGE of M. tuberculosis has shown to exhibit some promiscuity in recognizing substrates possessing substitutions within the XCXPXRX motif, and hence, has become widely adopted as a tool for engineering proteins with aldehyde tags [8]. In recent works, we have shown, by engineering the p3 coat protein of phage to display the XCXPXRX motif, that phage could be modified in vitro by the FGE of M. tuberculosis to provide a formylglycine (Fgly) residue to facilitate covalent immobilization of the phage onto amine bearing supports via imine formation [10], [18]. We carried out an evolutionary screening approach using the aforementioned concept of covalent capture of phage as the selection pressure. An additional source of selection pressure arose unanticipated from the enzymatic elution of the immobilized phage since proteolytic cleavage is sequence dependent. Utilizing a pCanta phagemid system, we produced a library of the entire XCXPXRX sequence space at the N terminus of the p3 coat protein separated from the wild-type p3 by a spacer containing a trypsin cut site. As depicted in Fig. 1 for a typical round of phage screening, the engineered phage were modified by FGE to afford Fgly residues on viable substrates, followed by exposure to amine bearing magnetic beads to allow covalent immobilization via Fgly displayed on modified phage, washing to remove non-covalently linked phage, enzymatic cleavage to elute the phage, and finally amplification in E. coli. The substrate sequence present on the phage were analyzed after each round (Fig. S1). With the input concentration held constant, we found ∼7×102 output CFUs after the first round, ∼5×103 output CFUs after the second round, and ∼104 output CFUs after the third and fourth rounds. After four rounds of screening, individual phage found in the 4th round output were found to possess one of the three following substrate sequences: 4(1)-QCSPKRP; 4(2)-HCTPRRP; 4(3)-TCYPTRK. From Fig. 2a, we can see the important role of FGE in allowing these three individual phage to be captured onto the magnetic beads for selection, since the same reaction performed without the enzyme FGE yields no covalent capture of these phage. As a first comparison of these three candidate substrate sequences, individual phage containing 4(1), 4(2), 4(3), or a negative control phage (NC) possessing no XCXPXRX motif were amplified and combined in equal proportions to compete in an FGE reaction followed by screening as described above for capture onto the amine bearing beads. The eluted phage were sequenced to reveal the proportion of the eluted phage having a particular substrate sequence. In this initial combined screening, 4(2) (having the sequence HCTPRRP) was found to represent 50% of the output colonies (see Fig. 2b for details of the other sequences).