Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Mar 17.
Published in final edited form as: ACS Chem Biol. 2015 Oct 29;11(1):61–68. doi: 10.1021/acschembio.5b00745

Construction of Lasso Peptide Fusion Proteins

Chuhan Zong 1, Mikhail O Maksimov 2, A James Link 2,3,*
PMCID: PMC4795474  NIHMSID: NIHMS766857  PMID: 26492187

Abstract

Lasso peptides are a family of ribosomally-synthesized and posttranslationally-modified peptides (RiPPs) typified by an isopeptide-bonded macrocycle between the peptide N-terminus and an aspartate or glutamate sidechain. The C-terminal portion of the peptide threads through the N-terminal macrocycle to give the characteristic lasso fold. Because of the inherent stability, both proteolytic and often thermal, of lasso peptides, we became interested in whether proteins could be fused to the free C-terminus of lasso peptides. Here we demonstrate fusion of two model proteins, the artificial leucine zipper A1 and the superfolder variant of GFP, to the C-terminus of the lasso peptide astexin-1. Successful lasso cyclization of the N-terminus of these fusion proteins requires a flexible linker in between the C-terminus of the lasso peptide and the N-terminus of the protein of interest. The ability to fuse lasso peptides to a protein of interest is an important step toward phage and bacterial display systems for the high-throughput screening of lasso peptide libraries for new functions.

Keywords: lasso peptide, protein fusion, phage display

Graphical Abstract

graphic file with name nihms766857u1.jpg

INTRODUCTION

Lasso peptides1, 2 are a growing class of ribosomally-derived natural products or RiPPs3 in which the N-terminus of the peptide is covalently cyclized with an acidic sidechain. This forms an isopeptide-bonded “ring” structure through which the C-terminal portion of the peptide is threaded (Figure 1). Bulky residues in the threaded portion of the peptide prevent unthreading, and these residues are referred to as the steric lock or plug residues. Lasso peptides exhibit improved protease resistance4 relative to their linear peptide counterparts. Because of their inherent protease resistance, lasso peptides are attractive starting points for therapeutic peptide development, and our group has previously demonstrated the construction of saturation mutagenesis libraries of the lasso peptide microcin J25 (MccJ25).5 In addition, the integrin-binding tripeptide epitope RGD has been inserted into MccJ25.6, 7 These studies provide the first glimpses at using lasso peptides as a reprogrammable scaffold. In order to perform high-throughput screens on lasso peptide libraries, it would be desirable to fuse the lasso peptide to an appropriate carrier protein for either phage display or bacterial display. Thus we became interested in whether such lasso peptide fusions could be constructed.

Figure 1. Lasso peptide structure and gene cluster architecture.

Figure 1

Open in a new tab

A) Comparison of the sequence and structure of two lasso peptides, microcin J25 (PDB 1Q71) and astexin-1 (this work) with the ring (salmon), loop (blue), and tail (green) sections indicated. The steric lock residues are shown in pink. Whereas microcin J25 has a long loop and a short tail, astexin-1 has a short loop and a long tail. B) Gene clusters for astexin-1 production (top) and astexin-1-fusion production (bottom). The fusion protein gene is added to the 3′ end of the astexin-1 precursor gene atxA1. Gene sizes not drawn to scale.

Since lasso peptides do not have a free N-terminus, fusions can only be constructed by adding a peptide or protein to the C-terminus of the peptide. Previous work on MccJ25 showed that addition of 1 or 2 amino acids to the C-terminus of this peptide reduced the production of the peptides dramatically, to levels just above the limit of detection.8 Among the structurally characterized lasso peptides,2 MccJ25 is fairly unique because it has a relative long “loop” region with a short “tail” (Figure 1A). Other lasso peptides, such as astexin-1,9, 10 have shorter loop regions and longer tails (Figure 1A). We hypothesized that the longer C-terminal tail region of astexin-1 would be more amenable to fusion with proteins. Here we have shown that astexin-1 fusion proteins can be successfully processed into a lassoed form as long as there is a sufficiently long flexible linker between the C-terminus of the lasso peptide and the fusion protein.

RESULTS AND DISCUSSION

Revisiting the structure of astexin-1

Astexin-1 is a polar, 23 aa lasso peptide, and our group has previously solved the structure of the full-length peptide in DMSO.9 Marahiel and co-workers have also solved a structure of astexin-1 in which the four C-terminal amino acids are truncated, astexin-1 (1–19), in water.10 These two structures differ in the positioning of the steric lock; the structure solved in organic solvent showed Arg-19 serving as the steric lock residue (PDB 2LTI) whereas the structure solved in water (PDB 2M37) implicated Tyr-14 and Phe-15 as the steric lock residues. We suspected that our structure solved in DMSO was partially unthreaded, so here we have solved the full-length astexin-1 structure in H2O/D2O. COSY, TOCSY, NOESY, and 13C-HSQC NMR experiments were carried out on full-length astexin-1 (Figure S1, Figure S2). This new structure (Figure 1A) confirms the steric lock residues as Tyr-14 and Phe-15. When calculated over residues 1–15 (ie, the ring, the loop, and the two steric lock residues), the RMSD between the backbones of the two structures solved in water is 1.09 Å (Figure S3). Thus in its physiologically-relevant aqueous form, astexin-1 consists of a 9 aa ring, a 5 aa loop, and a 9 aa tail. This contrasts with MccJ25 where the ring is 8 aa, the loop is 11 aa, and the tail is only 2 aa (Figure 1A).

Direct fusion of protein to the C-terminus of astexin-1

We next turned our attention to whether astexin-1 could accept fusions at its C-terminus. We fused the artificial leucine zipper protein A111, 12 directly to the C-terminus of astexin-1. A1 is small (62 aa, 6.9 kDa), has a simple helical structure, is well-folded, and expresses well in E. coli,12 thus we thought it would be a good starting point as a model protein for fusion to lasso peptides. We generated a plasmid, pCZ1, with the astexin-1 precursor gene (atxA1) fused to the A1 leucine zipper gene. The fusion gene was placed upstream of the maturation enzymes atxB1 and atxC1 (Figure 1B).9 Lasso peptide precursor proteins, such as AtxA1, are comprised of an N-terminal leader peptide13 and the C-terminal core peptide. The leader peptide likely enables docking of the precursor14 within the AtxB1 enzyme while the core peptide is transformed into the final lassoed product. AtxB1-like enzymes are cysteine proteases that cleave the N-terminal leader peptide from AtxA1 and thus generate a new N-terminus that becomes cyclized. AtxC1 and similar enzymes are related to asparagine synthetases and use ATP to cyclize the lasso peptide.15, 16 In addition, two control plasmids were constructed: pCZ5 lacks the atxC1 gene and pCZ2 lacks both maturation enzymes.

After purifying the protein products from each of these plasmids, we analyzed them using mass spectrometry. The protein produced from both pCZ1 and pCZ5 had mass of 10349 Da, the mass expected for an astexin-1-A1 fusion with its leader peptide removed, but without dehydration due to cyclization (Figure 2, Figure S4). As expected, the protein product of pCZ2 had a mass matching the unprocessed fusion protein (Figure 2, Figure S4). The lack of any lasso peptide formation in these constructs indicated that perhaps a flexible linker was necessary for the proper function of the cyclization enzyme AtxC1. However, it is noteworthy that AtxB1 is able to completely cleave the leader peptide from the precursor both in the presence and absence of AtxC1, indicating that AtxB1 is not sterically hindered from accessing the cleavage site between the leader and core peptide. This is in contrast to an observation in an in vitro study of MccJ25 biosynthesis16 in which McjB, homologous to AtxB1, was unable to cleave the leader peptide from the precursor McjA in the absence of McjC, the homolog of AtxC1.

Figure 2. Maturation of astexin-1 precursor fusion proteins.

Figure 2

Open in a new tab

A) Schematic of lasso peptide fusion protein maturation. The precursor to astexin-1 consists of a leader peptide (purple) and the core peptide (orange) that is converted into the lasso. Linkers (green), when present, are placed between the C-terminus of the core peptide and the fusion protein (red). The cysteine protease AtxB1 cleaves the leader peptide while the asparagine synthetase homolog AtxC1 generates the isopeptide bond that defines the lasso fold. B) Constructs tested in this study and average masses (expected) of the various processed forms. A1: A1 artificial leucine zipper, sfGFP: superfolder GFP, (GSSG)x: Gly-Ser-Ser-Gly repeat, Thb: thrombin cleavage site.

Introduction of flexible linkers

We hypothesized that the inability of AtxC1 to cyclize C-terminally fused lasso peptides could be due to steric interference of the well-folded A1 structure. Whereas AtxB1, the lasso peptide precursor protease, is 209 aa long, the AtxC1 enzyme is much larger, at 572 aa. Thus we generated constructs in which a flexible linker composed of Gly-Ser-Ser-Gly (GSSG) repeats was inserted between AtxA1 and the A1 leucine zipper. Two different linker lengths were investigated, (GSSG)2 and (GSSG)5, generating the plasmids pCZ25 and pCZ16, respectively. Gratifyingly, both of these constructs exhibited masses consistent with formation of the lasso peptide at the N-terminus (Figure 3). The conversion of the peptide to the lasso form, however, was not complete. The mass spectrometry data provided a qualitative indication that roughly half of the protein was lassoed. Controls lacking AtxC1 or both of the maturation enzymes behaved as expected (Figure S5, S6).

Figure 3. Mass spectrometry of astexin-1-A1 fusion proteins with flexible linkers.

Figure 3

Open in a new tab

A) Deconvoluted electrospray mass spectrum of the astexin-1-(GSSG)2-A1 construct showing a roughly 50:50 split between the lassoed and linear core peptide forms of the protein. B) As in A, but for the astexin-1-A1-(GGSG)5-A1 construct. a.u.: arbitrary units.

In order to obtain more quantitative estimates of the extent of lasso formation, we designed another set of constructs in which a thrombin protease cleavage site (LVPRGS, with cleavage between R and G) was placed after the (GSSG)2 or (GSSG)5 linker. This allowed us to remove the lasso peptide from the fusion protein and determine its abundance using the UV detector of an HPLC. The astexin-1-(GSSG)x-thrombin-A1 fusion proteins were analyzed first using SDS-PAGE. Constructs with both maturation enzymes clearly generated two protein products that were resolvable on the gel, presumably the lassoed product and its linear counterpart (Figure 4A) in a roughly 50:50 ratio. Mass spectrometry analysis of the intact protein also showed complete removal of the leader peptide along with roughly 50% conversion to the lasso form (Figure 4B). After thrombin cleavage of the fusion proteins, HPLC analysis with a UV detector at 215 nm revealed three peaks corresponding to the lasso peptide, its linear counterpart, and the cleaved A1 protein (Figure 4C, Figure S7). Integration of the peaks for the lasso and linear peptides allowed for quantitative analysis of the extent of lasso peptide formation. For the astexin-1-(GSSG)2-thrombin-A1 construct, this analysis showed that 49.7 ± 0.3 % of the peptide was lassoed. Results were similar with the astexin-1-(GSSG)5-thrombin-A1 construct with 53.1 ± 0.4 % of the peptide lassoed.

Figure 4. Quantification of the extent of lasso formation in astexin-1 fusion proteins.

Figure 4

Open in a new tab

Left panels correspond to astexin-1-(GSSG)2-thrombin-A1 protein, right panels correspond to astexin-1-(GSSG)5-thrombin-A1 protein. A) SDS-PAGE gel of purified astexin-1 fusion proteins, lane 1: intact astexin-1 gene cluster, lane 2: gene cluster lacking atxC1, lane 3: gene cluster lacking atxB1 and atxC1. When both maturation enzymes are present, a mixture of lassoed and linear core peptide forms of the protein is produced (lane 1). If only the AtxB1 enzyme is present, the leader peptide is removed completely, but no lasso formation occurs (lane 2). If neither AtxB1 nor AtxC1 is present, the leader peptide is not cleaved. B) Electrospray mass spectra of the mixture of lassoed and linear protein corresponding to the protein in lane 1 of panel A. C) HPLC analysis of thrombin-digested astexin-1 fusion proteins. The area under the curve was computed to determine the ratio of lassoed to linear protein.

To provide further confirmation that the thrombin-cleaved peptides were indeed lassoed, we purified the astexin-1-(GSSG)2 and astexin-1-(GSSG)5 lasso and linear peptides and analyzed them using MALDI-MS. The masses observed agree with the predicted masses for the lasso and linear forms of the peptide (Figure S8). The lasso peptides were also subjected to carboxypeptidase digestion. A mixture of carboxypeptidases B and Y removes amino acids from the C-terminus of the lasso peptide, but are usually unable to cleave beyond the steric lock residues of the lasso peptide, presumably due to steric hindrance.17, 18 In the case of astexin-1, Marahiel and coworkers reported that the four C-terminal amino acids of full-length astexin-1 were removed upon carboxypeptidase Y treatment, converting astexin-1 (1–23) into astexin-1 (1–19).10 When we subjected the astexin-1-(GSSG)2 and astexin-1-(GSSG)5 lasso peptides to carboxypeptidase B and Y treatment, we observed astexin-1 (1–20)as a major product in MALDI-MS along with some products resulting from incomplete cleavage (Figure S9). The fact that the carboxypeptidase cocktail cannot cleave beyond this residue is strong confirmation that the astexin-1-(GSSG)2 and astexin-1-(GSSG)5 peptides are indeed lassoed.

Fusing astexin-1 to GFP

Having established that a small, helical leucine zipper protein could be successfully appended to the C-terminus of astexin-1, we next wanted to test the generality of the fusion approach with a larger protein with different tertiary structure. We selected the superfolder variant of GFP19 (sfGFP) for such experiments. sfGFP is much larger than A1, with a mass of 26.8 kDa, and has a β-barrel structure, making it bulkier than the A1 protein. Since the (GSSG)5 linker gave the highest conversion to lasso peptide for the A1 protein, an astexin-1-(GSSG)5-thrombin-sfGFP construct was designed and expressed in E. coli (Figure 5A). The inherent fluorescence of sfGFP is a convenient property to test whether fusion of the lasso peptide has any effect on the fusion protein function. We measured the fluorescence intensity of three different concentrations of purified astexin-1-(GSSG)5-thrombin-sfGFP and compared these measurements to the fluorescence of purified sfGFP alone (Figure S10). We saw no statistically significant difference in fluorescence between the two proteins, indicating that the lasso peptide fusion does not interfere with the folding or chromophore formation of sfGFP.

Figure 5. Fusion of astexin-1 to superfolder GFP (sfGFP).

Figure 5

Open in a new tab

A) Gene cluster for astexin-1 sfGFP fusion production and model of the astexin-1-(GSSG)5-thrombin-sfGFP fusion protein. The (GSSG)5-thrombin-sfGFP portion of the protein was modeled in I-TASSER and then connected to the NMR structure of astexin-1. This illustration is intended to provide a sense of size of the linker and the lasso peptide relative to GFP. B) Mass spectrum of the mixture of lassoed and linear protein produced by the gene cluster in A. C) HPLC analysis of thrombin-digested astexin-1-(GSSG)5-thrombin-sfGFP protein.

As was the case with the A1 fusion proteins, ESI-MS analysis of intact astexin-1-(GSSG)5-thrombin-sfGFP fusion protein showed both linear and lassoed forms of astexin-1 (Figure 5B). To quantify the extent of lasso formation, we turned again to thrombin digestion of the fusion protein followed by HPLC quantification (Figure 5C). The percentage of lasso peptide formed was 44.4 ± 0.6 %, slightly lower than what was observed for the A1 protein. The lasso peptide was also collected and analyzed by MALDI-MS and carboxypeptidase digestion, confirming the identity of this peptide.

Here we have fused two model proteins, the artificial leucine zipper A1 and the superfolder variant of GFP, to the C-terminus of the lasso peptide astexin-1. The successful transformation of the linear lasso peptide precursor into the final threaded form is contingent upon the presence of a flexible linker between the C-terminus of astexin-1 and the N-terminus of the protein. An 8 aa (GSSG)2 linker is sufficient for conversion of ~50% of the astexin-1 precursor into lasso peptide when fused to A1. A longer (GSSG)5 linker also allows for ~50% conversion of astexin-1 when fused to either A1 or sfGFP. Our data indicates that the flexible linker is only required by the AtxC1 enzyme, which forms the isopeptide bond in astexin-1. The other astexin-1 maturation enzyme, AtxB1 is able to cleave the leader peptide from the astexin-1 precursor in both the presence and absence of a flexible linker. In all of our experiments leader peptide cleavage, catalyzed by AtxB1, went to completion. This held true even when the gene for AtxC1 was not present in the gene cluster. This result challenges the view that B and C enzymes in lasso peptide biosynthesis can function only in the presence of each other.8

In order to quantify the extent of lasso peptide formation, we developed constructs which include a thrombin protease cleavage site after the (GSSG)x linker. In so doing, we have generated lasso peptides with tails as long as 33 amino acids (for the astexin-1-(GSSG)5-thrombin fusions). This construct has a far longer tail than any known, naturally-existing lasso peptide. The lasso peptide tail is often susceptible to cleavage both in the native producer17 and in heterologous expression systems such as E. coli.9, 20, 21 Thus our work here has provided a route to such “tail-extended” lasso peptides. The ability to make such constructs opens up the possibility of grafting epitopes of interest into the tail of lasso peptides, which is complementary to the studies cited above6, 7 in which epitopes are grafted directly onto the lasso peptide scaffold.

We originally became interested in the problem of lasso peptide fusion proteins because of a desire to develop high-throughput screens for lasso peptide binding to a target of interest. Yeast22 and bacterial23, 24 display systems have been developed for the screening of libraries of other constrained peptide scaffolds such as those based on cystine knots. Adapting these systems to lasso peptides is challenging because the N-terminus of the lasso peptide disappears during its maturation. This precludes the use of popular phage display methods, such as fusion to the pIII protein of M13 bacteriophage,25 as well as many bacterial display technologies.26 However, with the ability to fuse lasso peptides to the N-terminus of a display protein, alternative techniques27 including T4,28 P4,29 and lambda30 phage display can be envisioned for lasso peptide display. Another intriguing possibility for lasso peptide display is display on Caulobacter crescentus via its S-layer protein RsaA.31 This protein is secreted via a C-terminal secretion sequence, and is thus compatible with fusion of a lasso peptide to the N-terminus of RsaA or one of its fragments. The fusion of lasso peptides to these phage or bacterial protein scaffolds is expected to expand the toolbox for high-throughput screening of constrained peptide libraries.

METHODS

Strains and reagents

PicoMaxx DNA polymerase (Agilent) was used for PCR amplification with primers purchased from IDT. T4 ligase and restriction enzymes EcoRI, BglII, and KpnI for cloning were purchased from New England Biolabs. The XL-1 Blue E. coli strain was used for all recombinant DNA steps, while BL-21 ΔslyD was used for protein expression.

Plasmid construction

Table S1 includes a list of all plasmids used in this paper. All new plasmids were derived from pQE-80 which includes a T5 promoter, a copy of lacIq, and an ampicillin resistance marker. The primers used to generate each construct are listed in Table S1, and the sequences of all primers appear in Table S2. The pCA14 plasmid includes the atxB1C1 operon encoding the astexin-1 maturation enzymes AtxB1 and AtxC1. This operon was amplified using PCR from the pMM32 plasmid9 and digested with BglII and HindIII. This product was ligated into similarly digested pQE80. To generate fusions between astexin-1 and the A1 leucine zipper,11 overlap PCR was employed. The atxA1 gene encoding the astexin-1 precursor was amplified from pMM32 with a 5′ EcoRI site. The gene encoding the A1 leucine zipper includes a C-terminal 6x-His tag and was amplified from pQE9-A111 with a 3′ BglII site. These two fragments were joined by overlap PCR, digested with EcoRI and BglII, and ligated into similarly digested pCA14, resulting in the plasmid pCZ1. The pCZ5 plasmid is a control for pCZ1, and it consists of the gene encoding the astexin-1-A1 fusion and just the AtxB1 maturation enzyme. A PCR product spanning the astexin-1-A1 fusion gene and AtxB1 was generated with a 3′ end KpnI site. This product was digested with EcoRI and KpnI and ligated into pQE-80 to generate pCZ5. The pCZ2 plasmid is an additional control for pCZ1 and lacks both the AtxB1 and AtxC1 maturation enzymes. It was constructed in the same way as pCZ5.

The pCZ16/pCZ20/pCZ21 plasmid series is analogous to pCZ1/pCZ5/pCZ2 described above except that a flexible glycine-serine linker, (GSSG)5 was inserted between astexin-1 and the A1 leucine zipper. The gene encoding astexin-1-(GSSG)5-A1 was assembled by overlap PCR. The (GSSG)5 fragment was amplified from pJP4715 while the astexin-1 and A1 gene fragments were amplified from pCZ1. Once assembled, the PCR product was digested as above for pCZ1 and ligated into pCA14 to generate pCZ16. The pCZ20 and pCZ21 plasmids were generated in an analogous fashion to pCZ5 and pCZ2. The pCZ43 plasmid is similar to pCZ16 except it also includes a thrombin cleavage site between the glycine-serine linker and A1. pCZ43 was assembled in the same way as pCZ16, except that primers encoding the thrombin cleavage site were used. Control plasmids pCZ44 and pCZ45 were made in an analogous fashion to pCZ20 and pCZ21. A set of plasmids with shorter glycine-serine linkers, (GSSG)2, were also constructed, both without (pCZ25, pCZ26, pCZ27) and with (pCZ29, pCZ40, pCZ39) thrombin cleavage sites. These plasmids were derived from the constructs with the (GSSG)5 linker.

Finally, plasmids with genes encoding for astexin-1-(GSSG)5-superfolder GFP (pCZ22) and astexin-1-(GSSG)5-thrombin-superfolder GFP (pCZ46) were constructed in the same way as pCZ16 and pCZ43, respectively. The superfolder GFP fragment was amplified from pDA38, which is a pQE60 derivative with the C-terminally 6x His-tagged superfolder GFP gene19 inserted in the multiple cloning site. All plasmid sequences were verified by Genewiz.

Purification of astexin-1

Expression and purification of astexin-1 were carried out in similar fashion as the purification of astexin-321 with some minor changes. Briefly, BL21 E. coli cells harboring the plasmid pMM32 (contains the atxA1B1C1 operon) were grown in M9 minimal medium and induced at an OD600 of 0.2 to 0.3 to produce astexin-1. The cells were lysed in methanol, and the insoluble fraction was removed by centrifugation. The methanol-soluble fraction, which contains astexin-1, was dried in a rotary evaporator, reconstituted in minimal water, and cleaned up using solid-phase extraction (Strata C8, 6 mL). Crude astexin-1 was eluted from the column using methanol which was subsequently removed using a rotary evaporator. The peptide was purified using HPLC (Zorbax 300SB-C18 Semi-Prep HPLC Column, 9.4 by 250 mm, Agilent Technologies) by applying the following solvent gradient at a flow-rate of 4.0 mL/min: 10% acetonitrile (ACN) in water for 1 min, ramp up to 50% ACN over 19 min, ramp up to 90% ACN over 5 min, 90% ACN for 5 min, ramp down to 10% ACN in 2 min. HPLC fractions containing astexin-1 were pooled and lyophilized (Labconco Freezone 4.5/−105 °C) resulting in a white powder.

NMR Spectroscopy of Astexin-1

The HPLC-purified astexin-1 (1.3 mg) was dissolved in 95:5 H2O/D2O (180 μL). COSY, TOCSY, NOESY (100 ms mixing time), and 13C-HSQC spectra were acquired on a Bruker Avance III 800 MHz spectrometer. All residues were assigned, and an initial structure was calculated using CYANA.32 The structure was energy-minimized in explicit water using GROMACS33 and the amber94 force field using a procedure adapted from Spronk et al.34 Refinement of structures in explicit water has been shown to improve overall quality and provide a more accurate representation of the conformational flexibility within the structure.35–37 After introducing the isopeptide bond, each structure was energy-minimized in vacuo for 100 steps using the limited-memory BFGS algorithm. The resulting structure was placed under a periodic boundary condition such that no atom was closer than 10 Å from the simulation box and solvated in water using the tip3p model. A short energy minimization was applied to soak the water molecules into the structure, followed by 0.5 ps of MD at 300 K. The isothermal-isobaric ensemble was chosen. A total of 25 ps of MD were performed for each structure, while cooling the system from 300 K to 50 K in four steps. During the simulation, the force constant associated with NMR derived distance restraints was decreased from 20,000 to 8,000 kJ mol−1 nm−2. The top 20 structures are presented in the PDB file.

Fusion protein expression

Plasmids were introduced into BL-21 ΔslyD for protein expression. Cells harboring plasmids encoding either astexin-1-A1 fusions or astexin-1-sfGFP fusions were grown in LB medium (10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl) with 100 mg/L ampicillin (Affymetrix) for plasmid maintenance. In a typical experiment, an overnight culture was diluted to an OD600 of 0.02, and the culture was grown at 37 °C until reaching an OD600 of 0.5. At this time, the culture was shifted to room temperature (~21 °C), induced with 1 mM isopropyl-β-D-1-thiogalactopyranoside (Gold Biotechnology), and allowed to express protein for 16 h. Cells then were harvested by centrifugation at 8,000 × g for 10 min at 4 °C, and cell pellets were frozen at −80 °C until needed for purification.

Fusion protein purification

Following protein expression, astexin-1-A1 fusion proteins and astexin-1-sfGFP fusion proteins were purified under native conditions. Cell pellets were resuspended in lysis buffer (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole, pH 8). The cells were incubated for 30 min on ice with the addition of 0.25 mg/ml lysozyme (Sigma Aldrich), and lysed by sonication. Following lysis, cells were centrifuged at 10,000 × g for 10 min at 4 °C, and the supernatants containing fusion proteins were purified on Ni-NTA resin according to the manufacturer’s recommendation (Qiagen). Purified proteins were analyzed by SDS-PAGE.

ESI-MS analysis of intact fusion protein

Electrospray ionization mass spectrometric (ESI-MS) analysis of fusion proteins was performed on a Bruker Daltonics MicroTOF-Q II mass spectrometer. Purified fusion proteins were first purified by HPLC using the same gradient and conditions described in the “Purification of astexin-1” section. Fusion proteins were collected based on their retention times on HPLC and subsequently lyophilized. Lyophilized samples were reconstituted in MS buffer (50% acetonitrile, 0.1% formic acid in H2O) and injected into the mass spectrometer. The observed average molecular weight of each fusion protein sample was obtained by deconvolution of the acquired intensity vs. m/z data using the Maximum Entropy algorithm (Spectrum Square Associates). Theoretical average molecular weights for the fusion proteins were calculated using mMass.

Thrombin digestion of fusion proteins and HPLC analysis

Prior to thrombin digestion, proteins were purified via HPLC as described in the previous paragraph. The peak corresponding to the intact protein was collected and the protein solution was lyophilized. This dried protein was reconstituted in thrombin cleavage buffer 1 (1x PBS, pH 7.3) for the A1 fusion proteins or thrombin cleavage buffer 2 (50 mM Tris, pH 8.0) for the sfGFP fusion proteins. Thrombin digestion assays were performed with 1.5–2 U thrombin (GE Healthcare) to cleave 450–750 μg of astexin-1-A1 fusion protein in 150–200 μL thrombin cleavage buffer 1 for 16 h at room temperature (~21 °C). For each astexin-1-sfGFP fusion protein sample, thrombin digestion assays were performed with 3 U thrombin (GE Healthcare) to cleave 900–1300 μg protein in 300 μL thrombin cleavage buffer 2 (50 mM Tris, pH 8) for 16 h at room temperature (~21 °C). Following thrombin digestion, digested fusion proteins were analyzed on HPLC using the same solvent gradient described above in the ESI-MS section. The typical injection volume of astexin-1-A1 fusion protein digests was 75–100 μL while the injection volume of astexin-1-sfGFP fusions was 290 μL. To quantify the percentage of lasso peptide, the area under the peaks of the lasso peptide and the linear were determined using the HPLC software and the percentage of lasso peptide was calculated as (area under lasso peak)/(area under lasso peak + area under linear peak). The percentage of lasso formation was calculated as the average of 3 or 4 biological replicates, and the error reported is the standard deviation of those measurements. The cleaved peptides (both lasso and linear) were also collected based on their retention time, dried under reduced pressure (DyNAVap, LabNet), and analyzed by MALDI mass spectrometry as well as carboxypeptidase digestion.

MALDI-MS analysis of thrombin-cleaved peptides

Acquisition of mass spectra in the m/z 500–6000 range was performed using an ABSciex 4800 Plus MALDI TOF/TOF Analyzer (ABSciex). The instrument was set to positive ion mode. Dried samples were resuspended in water, diluted with a 2.5 mg/ml solution of α-Cyano-4-hydroxycinnamic acid matrix and spotted onto an Applied Biosystem(ABI) 384 Opti-TOF 123 mm × 81 mm SS plate.

Carboxypeptidase digestion assays

Dried thrombin-cleaved lasso peptides and linear peptides were digested with 0.5 U carboxypeptidase B (Sigma-Aldrich) and 0.5 U carboxypeptidase Y (Affymetrix) in carboxypeptidase digestion buffer (50 mM sodium acetate, pH 6.0) for 3 h at room temperature. Digested samples were further analyzed by MALDI-MS as described above.

sfGFP Fluorescence intensity measurement

For a typical fluorescence intensity measurement, 100μl of purified sfGFP or astexin-1-(GSSG)5-Thb-sfGFP at 2.5 μM, 5 μM and 10 μM, in elution buffer (50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, pH 8) was added to a Costar 96-well black plate (Fisher Scientific) and equilibrated at 24 °C for 5 min. Protein concentrations were determined using the BCA assay according to the manufacturer’s recommendations (Pierce). Maximum emission at 510 nm was measured upon excitation with 485 nm light. The fluorescence level was measured in a BioTek Synergy 4 plate reader. Three technical replicates were performed for each concentration.

Supplementary Material

Supp Info

Acknowledgments

We thank T. Muir for the use of the ESI-MS instrument. This work was supported by NIH grant GM107036. AJL is a Sloan Research Fellow in Chemistry.

Footnotes

Accession numbers

The coordinates for the astexin-1 structure have been deposited to the PDB (accession number 2N68) and the BMRB (accession number 25754).

References

  • 1.Maksimov MO, Pan SJ, Link AJ. Lasso peptides: structure, function, biosynthesis, and engineering. Natural Product Reports. 2012;29:996–1006. doi: 10.1039/c2np20070h. [DOI] [PubMed] [Google Scholar]
  • 2.Hegemann JD, Zimmermann M, Xie X, Marahiel MA. Lasso Peptides: An Intriguing Class of Bacterial Natural Products. Accounts of Chemical Research. 2015;48:1909–1919. doi: 10.1021/acs.accounts.5b00156. [DOI] [PubMed] [Google Scholar]
  • 3.Arnison PG, Bibb MJ, Bierbaum G, Bowers AA, Bugni TS, Bulaj G, Camarero JA, Campopiano DJ, Challis GL, Clardy J, Cotter PD, Craik DJ, Dawson M, Dittmann E, Donadio S, Dorrestein PC, Entian K-D, Fischbach MA, Garavelli JS, Goransson U, Gruber CW, Haft DH, Hemscheidt TK, Hertweck C, Hill C, Horswill AR, Jaspars M, Kelly WL, Klinman JP, Kuipers OP, Link AJ, Liu W, Marahiel MA, Mitchell DA, Moll GN, Moore BS, Muller R, Nair SK, Nes IF, Norris GE, Olivera BM, Onaka H, Patchett ML, Piel J, Reaney MJT, Rebuffat S, Ross RP, Sahl H-G, Schmidt EW, Selsted ME, Severinov K, Shen B, Sivonen K, Smith L, Stein T, Sussmuth RD, Tagg JR, Tang G-L, Truman AW, Vederas JC, Walsh CT, Walton JD, Wenzel SC, Willey JM, van der Donk WA. Ribosomally synthesized and post-translationally modified peptide natural products: overview and recommendations for a universal nomenclature. Natural Product Reports. 2013;30:108–160. doi: 10.1039/c2np20085f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pomares MF, Salomon RA, Pavlova O, Severinov K, Farias R, Vincent PA. Potential Applicability of Chymotrypsin-Susceptible Microcin J25 Derivatives to Food Preservation. Applied and Environmental Microbiology. 2009;75:5734–5738. doi: 10.1128/AEM.01070-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Pan SJ, Link AJ. Sequence Diversity in the Lasso Peptide Framework: Discovery of Functional Microcin J25 Variants with Multiple Amino Acid Substitutions. Journal of the American Chemical Society. 2011;133:5016–5023. doi: 10.1021/ja1109634. [DOI] [PubMed] [Google Scholar]
  • 6.Knappe TA, Manzenrieder F, Mas-Moruno C, Linne U, Sasse F, Kessler H, Xie XL, Marahiel MA. Introducing Lasso Peptides as Molecular Scaffolds for Drug Design: Engineering of an Integrin Antagonist. Angewandte Chemie-International Edition. 2011;50:8714–8717. doi: 10.1002/anie.201102190. [DOI] [PubMed] [Google Scholar]
  • 7.Hegemann JD, De Simone M, Zimmermann M, Knappe TA, Xie X, Di Leva FS, Marinelli L, Novellino E, Zahler S, Kessler H, Marahiel MA. Rational Improvement of the Affinity and Selectivity of Integrin Binding of Grafted Lasso Peptides. Journal of Medicinal Chemistry. 2014;57:5829–5834. doi: 10.1021/jm5004478. [DOI] [PubMed] [Google Scholar]
  • 8.Ducasse R, Yan KP, Goulard C, Blond A, Li YY, Lescop E, Guittet E, Rebuffat S, Zirah S. Sequence Determinants Governing the Topology and Biological Activity of a Lasso Peptide, Microcin J25. Chembiochem. 2012;13:371–380. doi: 10.1002/cbic.201100702. [DOI] [PubMed] [Google Scholar]
  • 9.Maksimov MO, Pelczer I, Link AJ. Precursor-centric genome-mining approach for lasso peptide discovery. Proceedings of the National Academy of Sciences of the United States of America. 2012;109:15223–15228. doi: 10.1073/pnas.1208978109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Zimmermann M, Hegemann Julian D, Xie X, Marahiel Mohamed A. The Astexin-1 Lasso Peptides: Biosynthesis, Stability, and Structural Studies. Chemistry and Biology. 2013;20:558–569. doi: 10.1016/j.chembiol.2013.03.013. [DOI] [PubMed] [Google Scholar]
  • 11.Petka WA, Harden JL, McGrath KP, Wirtz D, Tirrell DA. Reversible hydrogels from self-assembling artificial proteins. Science. 1998;281:389–392. doi: 10.1126/science.281.5375.389. [DOI] [PubMed] [Google Scholar]
  • 12.Tang Y, Tirrell DA. Biosynthesis of a highly stable coiled-coil protein containing hexafluoroleucine in an engineered bacterial host. Journal of the American Chemical Society. 2001;123:11089. doi: 10.1021/ja016652k. [DOI] [PubMed] [Google Scholar]
  • 13.Cheung WL, Pan SJ, Link AJ. Much of the Microcin J25 Leader Peptide is Dispensable. J Am Chem Soc. 2010;132:2514–2515. doi: 10.1021/ja910191u. [DOI] [PubMed] [Google Scholar]
  • 14.Burkhart BJ, Hudson GA, Dunbar KL, Mitchell DA. A prevalent peptide-binding domain that guides ribosomal natural product biosynthesis. Nature Chemical Biology. 2015;11:564–570. doi: 10.1038/nchembio.1856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Pan SJ, Rajniak J, Cheung WL, Link AJ. Construction of a Single Polypeptide that Matures and Exports the Lasso Peptide Microcin J25. Chembiochem. 2012;13:367–370. doi: 10.1002/cbic.201100596. [DOI] [PubMed] [Google Scholar]
  • 16.Yan KP, Li YY, Zirah S, Goulard C, Knappe TA, Marahiel MA, Rebuffat S. Dissecting the Maturation Steps of the Lasso Peptide Microcin J25 in vitro. Chembiochem. 2012;13:1046–1052. doi: 10.1002/cbic.201200016. [DOI] [PubMed] [Google Scholar]
  • 17.Iwatsuki M, Tomoda H, Uchida R, Gouda H, Hirono S, Omura S. Lariatins, antimycobacterial peptides produced by Rhodococcus sp K01-B0171, have a lasso structure. Journal of the American Chemical Society. 2006;128:7486–7491. doi: 10.1021/ja056780z. [DOI] [PubMed] [Google Scholar]
  • 18.Hegemann JD, Zimmermann M, Xie XL, Marahiel MA. Caulosegnins I-III: A Highly Diverse Group of Lasso Peptides Derived from a Single Biosynthetic Gene Cluster. Journal of the American Chemical Society. 2013;135:210–222. doi: 10.1021/ja308173b. [DOI] [PubMed] [Google Scholar]
  • 19.Pedelacq JD, Cabantous S, Tran T, Terwilliger TC, Waldo GS. Engineering and characterization of a superfolder green fluorescent protein. Nature Biotechnology. 2006;24:79–88. doi: 10.1038/nbt1172. [DOI] [PubMed] [Google Scholar]
  • 20.Hegemann JD, Zimmermann M, Zhu SZ, Klug D, Marahiel MA. Lasso Peptides From Proteobacteria: Genome Mining Employing Heterologous Expression and Mass Spectrometry. Biopolymers. 2013;100:527–542. doi: 10.1002/bip.22326. [DOI] [PubMed] [Google Scholar]
  • 21.Maksimov MO, Link AJ. Discovery and Characterization of an Isopeptidase That Linearizes Lasso Peptides. Journal of the American Chemical Society. 2013;135:12038–12047. doi: 10.1021/ja4054256. [DOI] [PubMed] [Google Scholar]
  • 22.Silverman AP, Levin AM, Lahti JL, Cochran JR. Engineered Cystine-Knot Peptides that Bind alpha(v)beta(3) Integrin with Antibody-Like Affinities. Journal of Molecular Biology. 2009;385:1064–1075. doi: 10.1016/j.jmb.2008.11.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Getz JA, Cheneval O, Craik DJ, Daugherty PS. Design of a Cyclotide Antagonist of Neuropilin-1 and-2 That Potently Inhibits Endothelial Cell Migration. Acs Chemical Biology. 2013;8:1147–1154. doi: 10.1021/cb4000585. [DOI] [PubMed] [Google Scholar]
  • 24.Getz JA, Rice JJ, Daugherty PS. Protease-Resistant Peptide Ligands from a Knottin Scaffold Library. Acs Chemical Biology. 2011;6:837–844. doi: 10.1021/cb200039s. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Kehoe JW, Kay BK. Filamentous phage display in the new millennium. Chemical Reviews. 2005;105:4056–4072. doi: 10.1021/cr000261r. [DOI] [PubMed] [Google Scholar]
  • 26.Chen W, Georgiou G. Cell-surface display of heterologous proteins: From high-throughput screening to environmental applications. Biotechnology and Bioengineering. 2002;79:496–503. doi: 10.1002/bit.10407. [DOI] [PubMed] [Google Scholar]
  • 27.Sergeeva A, Kolonin MG, Molldrem JJ, Pasqualini R, Arap W. Display technologies: Application for the discovery of drug and gene delivery agents. Advanced Drug Delivery Reviews. 2006;58:1622–1654. doi: 10.1016/j.addr.2006.09.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Jiang J, Abu-Shilbayeh L, Rao VB. Display of a PorA peptide from Neisseria meningitidis on the bacteriophage T4 capsid surface. Infection and Immunity. 1997;65:4770–4777. doi: 10.1128/iai.65.11.4770-4777.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lindqvist BH, Naderi S. Peptide presentation by bacteriophage P4. FEMS Microbiology Reviews. 1995;17:33–39. doi: 10.1111/j.1574-6976.1995.tb00185.x. [DOI] [PubMed] [Google Scholar]
  • 30.Vaccaro P, Pavoni E, Monteriu G, Andrea P, Felici F, Minenkova O. Efficient display of scFv antibodies on bacteriophage lambda. J Immunol Methods. 2006;310:149–158. doi: 10.1016/j.jim.2006.01.003. [DOI] [PubMed] [Google Scholar]
  • 31.Nomellini JF, Duncan G, Dorocicz IR, Smit J. S-layer-mediated display of the immunoglobulin G-binding domain of streptococcal protein G on the surface of Caulobacter crescentus: Development of an immunoactive reagent. Applied and Environmental Microbiology. 2007;73:3245–3253. doi: 10.1128/AEM.02900-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Guntert P, Mumenthaler C, Wuthrich K. Torsion angle dynamics for NMR structure calculation with the new program DYANA. Journal of Molecular Biology. 1997;273:283–298. doi: 10.1006/jmbi.1997.1284. [DOI] [PubMed] [Google Scholar]
  • 33.Pronk S, Pall S, Schulz R, Larsson P, Bjelkmar P, Apostolov R, Shirts MR, Smith JC, Kasson PM, van der Spoel D, Hess B, Lindahl E. GROMACS 4.5: a high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics. 2013;29:845–854. doi: 10.1093/bioinformatics/btt055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Spronk C, Linge JP, Hilbers CW, Vuister GW. Improving the quality of protein structures derived by NMR spectroscopy. J Biomol NMR. 2002;22:281–289. doi: 10.1023/a:1014971029663. [DOI] [PubMed] [Google Scholar]
  • 35.Spronk C, Nabuurs SB, Krieger E, Vriend G, Vuister GW. Validation of protein structures derived by NMR spectroscopy. Prog Nucl Magn Reson Spectrosc. 2004;45:315–337. [Google Scholar]
  • 36.Linge JP, Williams MA, Spronk C, Bonvin A, Nilges M. Refinement of protein structures in explicit solvent. Proteins. 2003;50:496–506. doi: 10.1002/prot.10299. [DOI] [PubMed] [Google Scholar]
  • 37.Spronk C, Nabuurs SB, Bonvin A, Krieger E, Vuister GW, Vriend G. The precision of NMR structure ensembles revisited. J Biomol NMR. 2003;25:225–234. doi: 10.1023/a:1022819716110. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Info
NIHMS766857-supplement-Supp_Info.docx (6.1MB, docx)

RESOURCES