Abstract
The invention relates to a system for expressing nonribosomal peptide synthetases (NRPSs), polyketide synthases (PKS) or NRPS/PKS hybrid synth(et)ases. NRPS, PKS or hybrids thereof are large multi-domain proteins or multi-domain complexes, the expression of which for the production of peptides often causes difficulties. The invention correspondingly relates to a system for expressing portions of the enzymes which can be assembled post-translationally via protein-protein interactions, introduced in a targeted manner, to form multi-enzyme complexes. The invention discloses protein fragments of such an assembly, and the nucleic acids coding therefor. The invention also relates to a vector system for the protein fragments of the invention and its use for producing functional NRPS/PKS enzyme complexes.
Claims
1. A protein or a protein fragment comprising at least a first domain or partial domain of a non-ribosomal peptide synthetase (NRPS), a polyketide synthase (PKS) or an NRPS/PKS hybrid synth(et)ase (first PKS-NRPS domain), wherein the protein or the protein fragment has an N-terminus or a C-terminus comprising a first binding domain and wherein this first binding domain preferably represents the N-terminus or C-terminus, respectively, of the protein or the protein fragment, and wherein the first binding domain is characterised by the property of being able to enter into a specific protein-protein binding with at least one corresponding second binding domain.
2. The protein or protein fragment of claim 1, further comprising at least one, preferably two, three or four or more, further PKS-NRPS domain(s), wherein the further PKS-NRPS domain(s) is/are arranged in a direct functional arrangement next to the first PKS-NRPS domain.
3. The protein or protein fragment of any one of claim 1 or 2, wherein the first PKS-NRPS domain, or partial domain, is selected from an A domain, a C domain, a C/E domain, an E domain, a C.sub.start domain, an FT domain, or a T domain.
4. The protein or protein fragment according to any one of claims 1 to 3, comprising at least an A domain, a C domain and a T domain, preferably wherein the protein or protein fragment has at least one NRPS-PKS elongation module, an initiation module or a termination module.
5. The protein or protein fragment of any one of claims 1 to 4, wherein the binding domain comprises a synthetic coiled-coil domain (SYNZIP), preferably wherein the SYNZIP is selected from a 1-23 SYNZIP.
6. The protein or protein fragment of any one of the preceding claims, wherein the term opposite the first binding domain comprises a third binding domain, and wherein the third binding domain is characterised by the property of being able to enter into a specific protein-protein binding with at least one corresponding fourth binding domain.
7. The protein or protein fragment of claim 6, wherein the first and second binding domains are selectively capable of binding to the third or fourth binding domain.
8. The protein or protein fragment of any one of claims 1 to 7, wherein the first binding domain is linked to the first PKS-NRPS domain by a linker.
9. An isolated nucleic acid construct comprising a first coding region having a nucleic acid sequence encoding a protein or protein fragment according to any one of claims 1 to 8.
10. A vector system for producing a functional NRPS or PKS, wherein the vector system comprises at least one nucleic acid construct according to claim 9, and wherein the at least one nucleic acid construct is suitable for expressing at least two proteins or protein fragments according to any one of claims 1 to 8, and wherein the at least two proteins or protein fragments are different and together form a functional NRPS, PKS or NRPS/PKS hybrid.
11. The vector system of claim 10, wherein the at least two proteins or protein fragments form the functional NRPS, PKS or NRPS/PKS hybrid through the binding of the first and second binding domains.
12. The vector system according to any one of claim 10 or 11, wherein the vector system comprises nucleic acid constructs suitable for the expression of at least three or more proteins or protein fragments according to any one of claims 1 to 8, or wherein at least two of the three or more proteins or protein fragments together form a functional NRPS, PKS or an NRPS/PKS hybrid.
13. The vector system of claim 23, wherein at least three proteins or protein fragments can together form a functional NRPS, PKS or NRPS/PKS hybrid, wherein the functional NRPS or PKS is formed by binding the proteins or protein fragments to one another by means of binding a first binding domain to a second binding domain and binding a third binding domain to a fourth binding domain.
14. A process for preparing a functional (complete) NRPS or PKS comprising connecting at least a first protein or protein fragment of any one of claims 1 to 8 with a second protein or protein fragment of any one of claims 1 to 8, wherein the first protein or protein fragment has a terminal first binding domain, and wherein the second protein or protein fragment has the terminal second binding domain instead of the terminal first binding domain.
Description
BRIEF DESCRIPTION OF THE FIGURES AND SEQUENCES
[0065] The figures show
[0066] FIG. 1: shows the SYNZIP interaction partners and possible networks. A) Protein microarray assay results of 26 peptides forming specific interactive pairs. Peptides which are immobilised on the surface of the microarray are shown in series. Fluorescence-labelled peptides in solution are listed in row. According to the array score (shown on the right), black spots show a strong (0-0.2) and white spots a weak fluorescence signal (>1.0). The absence of homospecific interactions is indicated by the red diagonal line. Interactions that showed an array score of <0.2 are highlighted in green. The number of strong interaction partners is shown in the lower column (Reinke et al., 2010). B) Possible SYNZIP interaction networks: 1. linear 2. annular 3. branched and 4. orthogonal networks with the corresponding SYNZIP numbers are indicated. Dashed lines indicate a weak and solid lines a strong interaction. The star highlights the antiparallel interaction between SYZIP17 and SYNZIP18 (Thompson et al., 2012).
[0067] FIG. 2: shows the construction of an AmbS hybrid for the production of novel peptides. A: Schematic representation of the NRPS hybrids (NRPS-3a and NRPS-3b) from XUs of the AmbS (black) and GxpS (red). The associated relative peptide production of peptides 7, 8 and 9 from triplicate measurements is shown in %. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18. B: Structure of the peptides produced.
[0068] FIG. 3: shows the construction of an SzeS hybrid for the production of novel peptides. A: Schematic representation of the NRPS hybrids (NRPS-4a and NRPS-4b) and the covalently linked hybrid (NRPS-4c) from XUs of the SzeS (green) and GxpS (red). The associated relative peptide production of peptides 10 and 11 from triplicate measurements is shown in %. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain or FT domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18. B: Structure of the peptides produced.
[0069] FIG. 4: shows the construction of an XldS hybrid for the production of novel peptides. A: Schematic representation of the NRPS hybrids (NRPS-5a and NRPS-5b) from XUs of the XldS (turquoise) and GxpS (red). The associated relative peptide production of peptides 12, 13, 14 and 15 from triplicate measurements is shown in %. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18. B: Structure of the peptides produced.
[0070] FIG. 5: shows the proof of concept of various interfaces and SZ oligomerisation status based on the XtpS. Schematic representation of the XtpS (light green) divided in the T-C (NRPS-13), A-T (NRPS-14) and C-A (NRPS-15 and NRPS-16) as well as constructs with different SZ oligomerisation status (NRPS-15 and NRPS-16). The WT-XtpS (NRPS-1) was used as a reference. The relative production of peptides 1 and 2 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18, yellow: SZ19.
[0071] FIG. 6: shows the influence of the SZs on the production of the A-T divided XtpS. Schematic representation of the three control experiments without N-terminal SZ (NRPS-14b), C-terminal SZ (NRPS-14c) and both SZs (NRPS-14d), as well as representation of the construct with both SZs (NRPS-14a). The WT-XtpS (NRPS-1) was used as a reference. The relative production of peptides 1 and 2 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18.
[0072] FIG. 7: shows the influence of the SZs on the production of the C-A (SZ19/18) divided XtpS. Schematic representation of the three control experiments without N-terminal SZ (NRPS-16b), C-terminal SZ (NRPS-16c) and both SZs (NRPS-16d), as well as representation of the construct with both SZs (NRPS-16a). The WT-XtpS (NRPS-1) was used as a reference. The relative production of peptides 1 and 2 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: yellow, SZ19; green, SZ18.
[0073] FIG. 8: shows the influence of GS linkers on the production of the C-A (SZ17/18) divided XtpS. Schematic representation of the construct without GS linkers (NRPS-15a) and with a to AS long (NRPS-15b), 8 AS long (NRPS-15c) and 4 AS long GS linker (NRPS-15d), which was introduced between the C-terminal end of the first XtpS section and SZ 17. The WT-XtpS (NRPS-1) was used as a reference. The relative production of peptides 1 and 2 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18.
[0074] FIG. 9: shows the productivity of the three-part XtpS. Schematic diagram of the XtpS divided into the T-C(NRPS-17a) and A-T (NRPS-18a) linkers and corresponding negative control (NRPS-18b). The WT-XtpS (NRPS-1) was used as a reference. The relative production of peptides 1 and 2 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2.
[0075] FIG. 10: shows the productivity of the three-part GxpS. A: Schematic representation of the GxpS (NRPS-20) shared in the A-T linkers. The WT-GxpS (NRPS-2) was used as a reference. The relative production of peptides 3, 4, 5 and 6 from triplicate measurements is given in % of the WT level. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2. B: Structure of the peptides produced.
[0076] FIG. 11 shows the reprogramming of the XtpS for the production of novel peptides. A: Schematic representation of the hybrids NRPS-23b and NRPS-23c, which were produced by substituting the XtpS (light green) tridomain with a GxpS (red) and SzeS (green) tridomain. The relative production of peptides 16a/b, 17a, 18 and 19a/b from triplicate measurements is shown in % as normalised in comparison to WT (NRPS-1). The tripartite division of XtpS (NRPS-18) is also shown. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2. B: Structure of the peptides produced.
[0077] FIG. 12 shows the reprogramming of the XtpS for the production of novel peptides. A: Schematic representation of the hybrids NRPS-24a and NRPS-24c, which were produced by the substitution of the GxpS (red) tridomain with an XtpS, (light green) and SzeS (green) tridomain. The relative production of peptides 20, 21a/b, 22, 23a/b, 24, 25, 3 and 2 from triplicate measurements is shown in % in comparison with WT (NRPS-2). The tripartite division of GxpS (NRPS-20) is also shown. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2 B: structure of the peptides produced.
[0078] FIG. 13: shows the design of XtpS hybrids for the production of novel peptides. A: Schematic representation of the hybrids NRPS-26 and NRPS-27, which were each produced from parts of the GxpS (red) and SzeS (green) as well as XtpS (light green). The associated relative peptide production of peptides 20, 22 and 26 from triplicate measurements is shown in %. The tripartite division of XtpS (NRPS-18) is also shown. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain or FT domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2. B: Structure of the peptides produced.
[0079] FIG. 14: shows the design of GxpS hybrids for the production of novel peptides. A: Schematic representation of the hybrids NRPS-28 and NRPS-29, which were each produced from parts of the XtpS (light green) and SzeS (green) and GxpS (red). The associated relative peptide production of peptides 16a/b, 27, 17a/b, 28a/b, 18, 19, 29, 30, 31a/b and 32 from triplicate measurements is shown in %. The tripartite division of GxpS (NRPS-20) is also shown. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain or FT domain; diamond, C/E domain; small circle at the C-terminus, TE domain. Helices represent SZs: orange, SZ17; green, SZ18; dark blue: SZ1; light blue: SZ2. B: Structure of the peptides produced.
[0080] FIG. 15: shows the division of a hybrid of NRPS and PKS modules. The substance produced by the hybrid is glidobactin A (see structure). The NRPS GlbD was shared between the A and T domains with SZs. Symbols represent domains: circle, A domain; rectangle, T domain; triangle, C domain; PKS domains in GlbB are named according to their functions. Helices represent SZs: light grey, SZ17; dark grey, SZ18.
[0081] FIG. 16 shows a preferred embodiment in which the sequence of the Synzpi variants SZ1 and SZ2 (A) or SZ2 and SZ19 (B) were shortened in each case at the N terminus. The shortened but still fully functional syncip results in improved peptide production within the NRPS.
[0082] The sequences show:
[0083] SEQ ID Nos 1 and 2: synzip sequences
[0084] SEQ ID Nos 3 and 5: preferred sequence motifs for inserting the binding domains according to the invention
[0085] SEQ ID Nos 6 to 30: peptide sequences of the NRPS peptides produced in this application
EXAMPLES
[0086] Certain aspects and embodiments of the invention will now be illustrated by way of example and with reference to the descriptions, figures and tables set forth herein. Such examples of the substances, processes, uses and other aspects of the present invention are only representative and should not be understood as limiting.
[0087] The examples show:
Example 1: SYNZIP Mediated De Novo Design of NRPSs Based on the XU Concept
[0088] This work initially dealt with the de novo construction of NRPSs and the production of novel peptides based on the XU concept. By introducing the SYNZIP pair 17/18 into the conserved WNATE motif of the C-A linker, hybrid NRPs were to be constructed from two systems. The antiparallel SZ pair should serve as a non-covalent mediator between the various synthetases. With a dissociation constant (Kd) of <10 nM, SZ17 and SZ18 have a strong affinity to one another (Thompson et al., 2017), so that almost all properties of a covalent linkage are present. The NRPS hybrids were to be generated by combining the first two XUs of the AmbS, SzeS and XldS with the last three XUs of the GxpS SZ17 was added to the C-terminal end of the AmbS, SzeS and XldS and SZ18 to the N-terminal end of the GxpS. With regard to the rule established by Bozhüyük et al., which requires the consideration of C-domain specificity, the specificities for the first two hybrids (AmbS-GxpS and Sze-GxpS) were observed, but not for the last one (XldS-GxpS).
Example 2 Plasmid Construction and Heterologous Expression of GxpS Hybrids in E. coli DH10B::mtaA
[0089] The first two XUs of AmbS (A1-C3), SzeS (C1-C3) and XldS (C1-C3) were first amplified using the gDNA from X. miraniensis DSM 17902, X. szentirmaii DSM 16338 and X. indica DSM17382. For this purpose, the primers listed in Table 3 were used. These contained matching overhangs to a pACYC_ara_araE vector which already contained the sequence of SZ17. After linearisation of the vector, the plasmids pJW91 (ambS_A1-C3_SZ17), pJW92 (szeS_C1-C3_SZ17), and pJW93 (xldS_C1-C3_SZ17) were cloned from plasmid backbone and inserts by a hot fusion reaction (see 2.3.7). After screening and verification of the plasmids, they were each transformed together with a further plasmid pJW76 (SZ18_gxpS_A3-TE) or pJW83 (gxpS_A3-TE) into E. coli DH10B::mtaA. In doing so, pJW76 contained the sequence of the last three XUs of the GxpS and the sequence of the SZ18. In contrast, the transformation of pJW83, which lacks the sequence of SZ18, served as a negative control. Protein production was carried out by induction with L(+)-arabinose in triplicates at 22° C. for 72 hours.
[0090] In the subsequent analysis by means of HPLC-MS (see 2.5.2), a search was carried out for the masses of the peptides which would result from the hybrid systems. Accordingly, m/z values of 607.23 [M+H.sup.+] for 7 (linear peptide) and 589.33 [M+H.sup.+] 8 (cyclic peptide) were sought for the hybrid NRPS-3a, which was composed of parts of the AmbS and GxpS. These masses could be calculated from the peptide sequence (sQflL). Due to the promiscuity of the third GxpS A domain, which is capable of incorporating leucine in addition to phenylalanine, m/z values of 573.36 [M+H.sup.+] (linear peptide) and 555.35 [M+H.sup.+] (9) (cyclic peptide) were also searched for. In this case, the masses resulted from the sequence (sQllL). The peptides 7, 8 and 9, which eluted at a retention time of 6 min, 7.1 min and 7 min, could be identified on the basis of their mass. The linear peptide with the sequence (sQllL), on the other hand, could not be detected. Since no standard was available at the time of data acquisition, the quantification of the results was carried out relatively and was calculated from the mean value of the peak area (FIG. 2). This showed that 8 was the most frequently detected peptide. 7 and 9 were produced at 8.1% and 21.2% relative to 8. All peptides could be verified on the basis of their MS.sup.2 spectrum (Annex FIG. 2). Furthermore, the measurement data of the negative control (NRPS-3b) showed that the production of 7, 8 and 9 is also possible without N-terminal SZ (FIG. 7). However, the production of the peptides was significantly lower than in the comparison system with both SZs (NRPS-3a). Thus, 7 and 8 were only produced to ˜50% and peptide 9 to ˜18%.
[0091] For the second hybrid NRPS-4a, which was composed of the first two XUs of SzeS and the last three XUs of GxpS, data was searched in the recorded HPLC-MS for m/z values of 634.38 [M+H.sup.+] for the phenylalanine derivative 10 (formyl-1TflL) and 601.36 [M+H.sup.+] for the leucine derivative 11 (formyl-1TllL). The construct without N-terminal SZ (NRPS-4b) and the covalently linked system (NRPS-4c) served as comparison systems. Both the mass of 10 and the mass of 11 could be detected in the extracted ion chromatogram (EIC) of the measured data, which eluted in each case at a retention time of 8 minutes and 7.8 minutes. Peptide 10 was most frequently determined and peptide 11 was determined in relative values of 8% (FIG. 3). Furthermore, the measurement data of the negative control (NRPS-4b) showed a significant decrease of 10 and 11 by ˜80%. The covalently linked construct (NRPS-4c) also produced the peptides in significantly smaller amounts. Thus, only ˜60% of 10 and ˜30% of ii were detected relative to NRPS-4a.
[0092] Ultimately, the HPLC-MS data of the third hybrid (NRPS-5a), which was composed of parts of the XldS and GxpS, showed the production of most expected derivatives (FIG. 4). Since the C1 domain of the XldS permits the incorporation of a C13, C14 or C15 FS at the N-terminal end of the peptide, the promiscuity of the third GxpS A domain results in six possible derivatives with m/z values of 830.54 [M+H.sup.+] 12 (13:0-qNflL), 844.55 [M+H.sup.+] 13 (14:0-qNflL), 858.12 [M+H.sup.+] (15:0-qNllL), 796.55 [M+H.sup.+] (13:0-qNllL), 810.57 [M+H.sup.+] 15 (14:0-qNllL) and 824.59 [M+H.sup.+] (15:0-qNllL). Four of them were detected. The retention times of the C13, C14 or C15 derivatives were in each case 11.3 min, 11.8 min and 12.3 min. While 13 was the most frequently produced peptide, the remaining peptides were detected in relative amounts between 2.3% (12) and 14.3% (15) (FIG. 9). Furthermore, the signal intensity of the EICs was low, indicating low overall production. In addition, the negative control (NRPS-5b) showed no significant difference in peptide production from the NRPS-5a construct (FIG. 4).
Example 3: Strategies for the SYNZIP-Mediated Reconstruction of NRPSs
[0093] Above, the conserved WNATE motif (SEQ ID No 3) of the C-A linker was chosen as the preferred cleavage site on the basis of the XU concept. This cleavage site was postulated by Bozhüyük et al. from sequence alignments of NRPS linker regions from Photorhabdus and Xenorhabdus and from published NRPS structural data of other organisms as an ideal fusion point. Also mentioned was that the A-T and T-C linker regions are less suitable for the reprogramming of NRPS because of their low conserved sequence compared with the C-A linker. With the introduction of the SZ pair 17/18 into the T-C and T-A linker regions, a comparative test should nevertheless be carried out to determine whether these insertion sites are not also equally suitable fusion points. This hypothesis was checked using the XtpS model system. To this end, XtpS was aligned with the structural data of bacillibactin synthetase from Bacillus subtilis (Tarry et al., 2017), which was published in 2017, in order to obtain conclusions about possible secondary structures. Subsequently, on the basis of this, cleavage sites in the T-C and A-T linker region were defined, which ultimately related to the sequence motifs RV|LP (SEQ ID No 4) of the T-C linker and VY|AAP (SEQ ID No 5; vertical line illustrates cleavage site) of the A-T linker.
[0094] Furthermore, two different SYNZIP oligomerisation statuses should be compared, which mean that the XtpS subunits bound to the SZs are either in spatial proximity or further apart. In principle, both orientations can be implemented with both parallel and antiparallel SZ. However, only the conformation in which the proteins are further apart is practicable. The reason for this is that after dividing the NRPS system in two, the SZs can only be attached to two (C-terminus of the first and N-terminus of the second NRPS part), instead of four possible terms (N- and C-terminus of the first and N- and C-terminus of the second protein part) of the proteins. By introducing SZ19 and the functional reverse form of SZ18, however, a close orientation seems possible. Nevertheless, this pair is not characterised by Thompson et al. (only SZ19 with the forward form of SZ18), and accordingly data on the Kd value, interaction partners, etc., is missing. Ultimately, the antiparallel SZ pair 17/18 was used for wide conformation and the parallel SZ pair 19/18 for close conformation.
[0095] For the XtpS divided in the T-C and A-T linkers (NRPS-13 and NRPS-14, cleavage site see 3.2), in each case two plasmids, pNA2 (xtpS_A1-T2_SZ17) and pNA3 (SZ18_xtpS_C3-TE), and pNA4 (xtpS_A1-A2_SZ17) and pNA5 (SZ18_xtpS_T2-TE), were assembled and together transformed in E. coli DH10B::mtaA. In contrast, the plasmids pJW61 (xtpS_A1-C3_SZ17) and pJW62 (SZ18_xtpS_A3-TE) were used to represent the cleavage site in the C-A linker. Furthermore, for the NRPS-16 divided in the C-A linker, the C-terminal SZ17 was replaced by SZ19, while the N-terminal SZ18reverse remained unchanged. The production by the wild-type XtpS (NRPS-1) served as a reference. Production cultures of all constructs were simultaneously prepared as triplicates and the synthesis of 1 and 2 was tested by means of HPLC-MS.
[0096] Since an absolute quantification was not possible due to a missing standard, a relative evaluation of the peak area was carried out (FIG. 5). Different from the relative values in FIG. 5, the linear peptide 1 is not formed in virtually the same amounts as the cyclic peptide 2, based on the absolute peptide yield, but only at about 0.1%. This result comes about merely because of the better ionisation of the linear peptide and must be taken into account when considering the relative values. The measurement results showed that the production of the cyclic product 2, with an m/z value of 411.31 [M+H+], and in most cases also the production of the linear peptide 1, with an m/z value of 429.31 [M+H+], could be demonstrated for all constructs (FIG. 5). 2 was best produced with about 80% relative to the WT by the constructs divided in the T-C(NRPS-13) and A-T (NRPS-14) linkers, whereas the two constructs divided in the C-A, NRPS-15 and NRPS-16, showed a significantly lower production with 27% (NRPS-15) and 13% (NRPS-16). The linear peptide (1) was produced in negligibly smaller amounts better by NRPS-14 instead of NRPS-13 and NRPS-16 showed no production of 1.
Example 5: Influence of the SYNZIPs on the Production of the A-T Divided XtpS
[0097] The influence of the SZs on the production of 1 and 2 was examined for the A-T divided construct NRPS-14 a (FIG. 6). After assembling pNA11 (xtpS_A1-A2) and pNA12 (xtpS_T2-TE), control experiments were carried out in which, in the first case, the N-terminal (NRPS-14b), in the second case the C-terminal (NRPS-14c) and in the third case both SZs (NRPS-14c) (FIG. 6) were left out. For this purpose, the plasmids pNA12 (xtpS_T2-TE) and pNA11 (XtpS_A1-A2) were cloned, each of which lacked the sequence of SZ17 and SZ18. The results of the HPLC-MS data are shown in FIG. 6. The negative controls of the A-T divided construct (NRPS-14b, NRPS-14c and NRPS-14d) showed a very low production of the cyclic product 2 and absolutely no production of the linear peptide 1 at ˜3-10% of the WT level. Overall, the controls showed a decrease in productivity of 90% compared to NRPS-14a (FIG. 6). Furthermore, NRPS-14a produced the peptides 2 and 1 with 104% and 81% at WT level.
Example 6: Influence of the SYNZIPs on the Production of the A-C(5Z19/18) Shared XtpS
[0098] The same control experiments were likewise carried out for the A-C divided construct with SZ pair 19/18. Since this construct already showed low production with both SZs (FIG. 5), it was possible to detect only a very low or no production of 2 and 1 for the three control experiments carried out (FIG. 7). The relative analysis of the HPLC-MS measurement data showed no peptide production for control experiments NRPS-16b and NRPS-16d; NRPS-16c showed only a very low production of 2 with 3.2% of the WT level. In relation to the construct with both SZs (NRPS-16a), control NRPS-16c shows a decrease in the production of cyclic peptide 2 by 80%.
Example 7: Influence of GS Linkers on the Production of A-C(SZ17/18) Shared XtpS
[0099] Since the C-A (NRPS-15, FIG. 5), compared to the T-C(NRPS-13, FIG. 5) and A-T (NRPS-14, FIG. 5) divided XtpS construct, showed considerably poorer productivity, GS linkers of different lengths were introduced between the C-terminal end of the first XtpS section and SZ17, with the aim of increasing productivity. Since, according to the XU concept published by Bozhüyük et al. for the construction of reprogrammed NRPS, ten AS of the conserved WNATE motif were deleted and the same happened to the C-A construct (NRPS-15) shown in FIG. 8, the introduction of a ten AS long GS linker was started. This was achieved by assembling the plasmid pNA8 (xtpS_A1-C3_GS(10)_SZ17), a plasmid derived from pJW61 (A1-C3_SZ17). In addition, two further plasmids, pNA9 (xtpS_A1-C3_GS(8)_SZ17) and pNA10 (xtpS_A1-C3_GS(4)_SZ17), were constructed, which code for an eight and four AS long GS linker. The evaluation of the HPLC-MS measurement data showed a better production of all constructs (NRPS-15b, NRPS-15c, NRPS-15d) with GS linker compared to the construct without it. Overall, the introduction of a linker resulted in an average increase in productivity of ˜37% for cyclic 2 and ˜26% for linear peptide 1. Furthermore, the cyclic product 2 with almost WT level was produced for all constructions with GS linker.
Example 7: Proof of Concept: Productivity of a Three-Part XtpS System
[0100] Since the division of the XtpS into two parts was successful for each of the positions mentioned above, the next step was to divide the system into two parts. For this purpose, a further SYNZIP pair, SZ1 and SZ2, was introduced which does not communicate with SZ17 and SZ18 and thus forms a so-called orthogonal network. With a Kd value of <10 nM, SZ1 and SZ2, as well as SZ17 and SZ18, show a very strong affinity for one another. Furthermore, only the A-T and T-C linker regions were selected as positions for the three-part division, which proved to be the most favourable positions with the best production through NRPS-13 and NRPS-14 (FIG. 5). Accordingly, two constructs, NRPS-17a and NRPS-18a, were produced, which were each divided into the linker regions T-C and A-T, respectively, with the introduction of the two SZ pairs. In detail, the SZs were introduced into the second and third T-C linkers or second and third A-T linkers. A total of four further plasmids, namely pNA17 (SZ18_xtpS_C3-T3_SZ1) and pNA18 (SZ2_xtpS_C4-TE), were cloned for the division in the T-C, and pNA15 (SZ18_xtpS_T2-A3_SZ1) and pNA16 (xtpS_SZ2_T3-TE) were cloned for the division in the A-T region. In addition, the plasmids were assembled without SZs as negative controls. This resulted in pNA19 (xtpS_T2-A3) and pNA20 (xtpS_T3-TE) for the negative control of the A-T split (NRPS-18b, FIG. 9).
[0101] For both three-part constructs, NRPS-17a and NRPS-18a, the production of the linear 1 and cyclic 2 peptide could be identified (FIG. 9). In comparison to NRPS-18a, NRPS-17a produced the peptides in two (2) to more than three times the amount (1). Accordingly, 2 was identified as 71.7% and 32.2%, respectively, and 1 as 25.6% and 7.3%, respectively. Furthermore, the negative control of the A-T divided system (NRPS-18b) showed no production of the peptides.
Example 8: SYNZIP-Mediated Tridomain Exchange for the Construction of Hybrid NRPs
[0102] In addition to XtpS, GxpS (NRPS-20) were also divided into three parts (FIG. 10) and tridomain sections of XldS and SzeS were produced. The resulting tridomain sections of the systems should then be combined with one another in a further experiment for the production of novel peptides. From sequence alignments of all A-T and T-C linker regions of the four systems, the cleavage site of the A-T linker (see 3.2, slight variation of the sequence motif within and between the systems) turned out to be a more favourable cleavage site (sequence motif of the cleavage site more conserved). Accordingly, only the interface in the A-T linker was used for all other constructs shown. This means that the substrate specificity of the downstream C domain should no longer be taken into account, as in the XU concept, but that of the upstream C domain. In addition to XtpS, NRPS-23 also consists of parts of the GxpS and SzeS (FIG. 11). After checking the productivity of the three-part NRPS-18, NRPS-23b and NRPS-23c, the tridomains were interchanged.
Example 9: Productivity of Further Systems Divided into Three Parts
[0103] For the construction of the GxpS (NRPS-18) and SzeS (NRPS-19) systems divided into three parts, a set of three plasmids was cloned in each case. This resulted in the plasmids pNA26 (gxpS_A1-A2_SZ17), pNA27 (SZ18_gxpS_T2-A3_SZ1) and pNA28 (SZ2_gxpS_T3-TE) as well as pNA29 (szeS_C1-A2_SZ17), pNA30 (szeS_T2-A3_SZ1) and pNA31 (SZ2_szeS_T3-TE), which were each transformed jointly in E. coli DH10B::mtaA.
[0104] For the NRPS-20 (three-part GxpS), all four derivatives 3, 4, 5 and 6 with m/z values of 586.40 [M+H+], 600.41 [M+H+], 552.41 [M+H+] and 566.43 [M+H+] in each case could be determined (FIG. 10). Compared to the WT-NRPS (NRPS-2), however, productivity was greatly reduced. For example, only 5.2% of 3 was produced, only 10.8% of 4 and only ˜14% of 5 and 6 (FIG. 10).
Example 10: The Exchange of Tridomains for the Production of Novel Peptides
[0105] The division of the described NRPSs based on three plasmids makes manipulation of the systems simple. With the experimental implementation, instead of the original plasmids, one or two plasmids are replaced and transformed together in a new constellation into the respective expression strain. The post-translational communication between the various NRPSs is then mediated by the artificial leucine zippers. Thus, for example, the second plasmid of the XtpS set can be replaced by the second plasmid of the GxpS set, thereby constructing a new hybrid system. Overall, the plasmids produced in this work permit the construction of 50 hybrid synthetases. In the following examples, 8 of them are described.
[0106] The first tridomain exchange was intended to allow the substitution of the second valine of Xtp for a phenylalanine. In the experimental implementation, instead of the pNA15 plasmid, the plasmids pNA27 and pNA30 were transformed together with pNA4 and pNA16 in E. coli DH10B::mtaA, thereby enabling the production of the hybrids NRPS-23b and NRPS-23c (FIG. 11).
[0107] The relative evaluation of the HPLC-MS analysis showed the peptides to be expected for NRPS-23b and NRPS-23c (FIG. 11). Thus, both the phenylalanine derivative and the leucine derivative in linear form (16, 17) and cyclic form (18, 19) were detected by NRPS-23b. In the EIC of peptides 16 and 19, there were also double peaks in each case which had different retention times, but showed identical fragmentation of the MS2 spectrum. From this, it was concluded that the peptides occurred as stereoisomers and eluted accordingly at different times. Since a non-natural protein-protein interface exists exclusively for the last C/E domain of the hybrid NRPS-23b, it was deduced that the upstream AS occurs in two conformations. In the example of 16 and 19, these are in each case the AS phenylalanine and leucine. However, which AS is actually affected has not been studied and therefore remains unresolved. Furthermore, the peptides 16 and 18 were produced by NRPS-23c, as already produced by NRPS-23b. A double peak occurred again for peptide 16, for which reason stereoisomers 16a and 16b were assumed. For the relative evaluation of the isomers, the peak areas were added.
[0108] The most frequently detected peptides were the linear peptides 16a/b (in both hybrids) and 17, all of which were produced in virtually identical amounts of 94.4%-100% (FIG. 11). The influence of ionisation on the frequency of the linear peptide is discussed in section 4. Overall, both hybrids, NRPS-23b and NRPS-23c, showed a similar production of the peptides 16a/b and 18, after which 16a/b were produced with 94.4% and 100% respectively and 18 with 19.1% and 24.8% respectively. Furthermore, very similar values could also be determined for the phenylalanine derivatives (16a/b and 18) and leucine derivatives (17a and 19a/b) which were produced by NRPS-23b.
[0109] In the second tridomain exchange, the substitution of the phenylalanine of the Gxps by the valine (from Xtp) or phenylalanine (from Sze) should take place. By replacing pNA27 with pNA15 and pNA30 in each case, the hybrids NRPS-24a and NRPS-24c were produced (FIG. 12).
[0110] The relative analysis of the measured data showed that peptide production could be determined for NRPS-22a and NRPS-22c. Both hybrids produced both the valine derivative (20, 22, 24 and 3) and the leucine derivative (21a/b, 23, 25 and 4) in linear and cyclic form (FIG. 12). The most commonly produced peptide of the hybrid NRPS-24a was the cyclic valine derivative 22. The linear shape (20) was detected at 34.7% relative to 22. Furthermore, the leucine derivative was produced in cyclic form (23) at 62.9% in smaller amounts than 22%. The linear peptide 21 occurred with relative values of 20.9% and was detected as a stereoisomer (21a and 21b). NRPS-24c showed an overall poorer production compared to NRPS-24a. Thus, the cyclic valine derivative 3 was produced to the extent of 66.1% compared with 22, and the cyclic leucine derivative 4 was detected to the extent of only 16.3% in relative values. Furthermore, the linear peptides 24 and 25 were detected in NRPS-22c at 13.4% and 2.7% in smaller amounts than the cyclic peptides. The structures of the peptides are shown in FIG. 12B. Further novel peptides could correspondingly be obtained by further combinations of the corresponding plasmids and are shown in FIGS. 14 and 14.
[0111] Furthermore, the hybrid of PKS and NRPS modules shown in FIG. 15 was generated, which leads to a complete synthesis of the glidobactin peptide. For this purpose, GlbD was divided between A-T and SZ17/SZ18. In the negative controls without in each case one or both SZs, there is virtually no glidobactin A production.
[0112] The following plasmids were used in the context of the examples:
TABLE-US-00001 Name Genotype Reference pACYC_ara_araE ori p15A, cm.sup.R, araC-P.sub.BAD, tacI- AK Bode araE, MCS pCOLA_ara_tacI ori ColA, kan.sup.R, araC-P.sub.BAD, tacI, AK Bode MCS pCDF_ara_tacI ori ColDF13, spek.sup.R, araC-P.sub.BAD AK Bode tacI, MCS pACYC_ara_XtpS ori p15A, cm.sup.R, araC-P.sub.BAD XtpS Watzel, 2019 and tacI-araE (unpublished) pACYC_ara_GxpS ori p15A, cm.sup.R, araC-P.sub.BAD GxpS Watzel, 2019 and tacI-araE (unpublished) pA22.3 ori ColA, kan.sup.R, araC-P.sub.BAD Bozhüyük et szeS_C.sub.1A.sub.1T.sub.1C/ al., 2018 E.sub.2A.sub.2T.sub.2C.sub.3.sub.—gxpS_A.sub.3T.sub.3C/ E.sub.4A.sub.4T.sub.4TE und tacI pJW61 ori p15A, cm.sup.R, araC-P.sub.BAD Watzel, 2019 xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- (unpublished) SYNZIP17 and tacI-araE pJW62 ori ColA, kan.sup.R, araC-P.sub.BAD Watzel, 2019 SYNZIP18-xtpS_A.sub.3T.sub.3C/ (unpublished) E.sub.4A.sub.4T.sub.4TE and tacI pJW63 ori p15A, cm.sup.R, araC-P.sub.BAD Watzel, 2019 xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3 and (unpublished) tacI-araE pJW64 ori ColA, kan.sup.R, araC-P.sub.BAD Watzel, 2019 xtpS_A.sub.3T.sub.3C7E.sub.4A.sub.4T.sub.4TE and (unpublished) tacI pJW75 ori p15A, cm.sup.R, araC-P.sub.BAD Watzel, 2019 gxpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- (unpublished) SYNZIP17 and tacI-araE pJW76 ori ColA, kan.sup.R, araC-P.sub.BAD Watzel, 2019 SYNZIP18-gxpS_A.sub.3T.sub.3C/ (unpublished) E.sub.4A.sub.4T.sub.4TE and tacI pJW82 ori p15A, cm.sup.R, araC-P.sub.BAD Watzel, 2019 gxpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3 and (unpublished) tacI-araE pJW83 ori ColA, kan.sup.R, araC-P.sub.BAD Watzel, 2019 gxpS_A.sub.3T.sub.3C/E.sub.4A.sub.4T.sub.4TE and (unpublished) tacI pJW91 ori p15A, cm.sup.R, araC-P.sub.BAD This study ambS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- SYNZIP17 and tacI-araE pJW92 ori p15A, cm.sup.R, araC-P.sub.BAD This study szeS_C.sub.1A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- SYNZIP17 and tacI-araE pJW93 ori p15A, cm.sup.R, araC-P.sub.BAD This study xldS_C.sub.1A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- SYNZIP17 and tacI-araE pNA1 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3- SYNZIP19 and tacI-araE pNA2 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2-SYNZIP17 and tacI-araE pNA3 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-xtpS_C.sub.3A.sub.3T.sub.3C/ E.sub.4A.sub.4T.sub.4TE and tacI pNA4 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2-SYNZIP17 and tacI-araE pNA5 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-xtpS_T.sub.2C.sub.3A.sub.3T.sub.3C/ E.sub.4A.sub.4T.sub.4TE and tacI pNA6 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2 and tacI- araE pNA7 ori ColA, kan.sup.R, araC-P.sub.BAD This study xtpS_C.sub.3A.sub.3T.sub.3C/E.sub.4A.sub.4T.sub.4TE and tacI pNA8 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3-GS(10)- SYNZIP17 and tacI-araE pNA9 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3-GS(8)- SYNZIP17 and tacI-araE pNA10 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2T.sub.2C.sub.3-GS(4)- SYNZIP17 and tacI-araE pNA11 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/E.sub.2A.sub.2 and tacI- araE pNA12 ori ColA, kan.sup.R, araC-P.sub.BAD This study xtpS_T.sub.1C.sub.3A.sub.3T.sub.3C/E.sub.4A.sub.4T.sub.4TE and tacI pNA14 ori p15A, cm.sup.R, araC-P.sub.BAD This study xtpS_A.sub.1T.sub.1C/ E.sub.2A.sub.2T.sub.1C.sub.3A.sub.3T.sub.3C/E.sub.4A.sub.4T.sub.4- gxpS_TE and tacI-araE pNA15 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-xtpS_T.sub.2C.sub.3A.sub.3- SYNZIP1 and tacI pNA16 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study SYNZIP2-xtpS_T.sub.3C/E.sub.4A.sub.4T.sub.4TE and tacI pNA17 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-xtpS_C.sub.3A.sub.3T.sub.4- SYNZIP1 and tacI pNA18 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study SYNZIP2-xtpS_C/E.sub.4A.sub.4T.sub.4TE and tacI pNA19 ori ColA, kan.sup.R, araC-P.sub.BAD This study xtpS_T.sub.2C.sub.3A.sub.3 and tacI pNA20 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study xtpS_T.sub.3C/E.sub.4A.sub.4T.sub.4TE and tacI pNA21 ori ColA, kan.sup.R, araC-P.sub.BAD This study xtpS_C.sub.3A.sub.3T.sub.4 and tacI pNA22 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study xtpS_C/E.sub.4A.sub.4T.sub.4TE and tacI pNA26 ori p15A, cm.sup.R, araC-P.sub.BAD This study gxpS_A.sub.1T.sub.1C/E.sub.2A.sub.2-SANZIP17 and tacI-araE pNA27 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-gxpS_T.sub.2C.sub.3A.sub.3- SYNZIP1 and tacI pNA28 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study SYNZIP2-gxpS_T.sub.3C/E.sub.4A.sub.4T.sub.4 C/E.sub.5A.sub.5T.sub.5TE and tacI pNA29 ori p15A, cm.sup.R, araC-P.sub.BAD This study szeS_C.sub.1A.sub.1T.sub.1C/E.sub.2A.sub.2-SYNZIP17 and tacI-araE pNA30 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-szeS_T.sub.2C.sub.3A.sub.3- SYNZIP1 and tacI pNA31 ori CloDF13, spec.sup.R, araC-P.sub.BAD This study SYNZIP2-szeS_T.sub.3C/E.sub.4A.sub.4T.sub.4 C/E.sub.5A.sub.5T.sub.5C.sub.6A.sub.6T.sub.6TE and tacI pNA34 ori ColA, kan.sup.R, araC-P.sub.BAD This study SYNZIP18-xtpS_A.sub.3T.sub.3C/ E.sub.4A.sub.4T.sub.4-gxpS_TE and tacI
