Transcription activation by the resistance protein AlbA as a tool to evaluate derivatives of the antibiotic albicidin

The rising numbers of fatal infections with resistant pathogens emphasizes the urgent need for new antibiotics. Ideally, new antibiotics should be able to evade or overcome existing resistance mechanisms. The peptide antibiotic albicidin is a highly potent antibacterial compound with a broad activity spectrum but also with several known resistance mechanisms. In order to assess the effectiveness of novel albicidin derivatives in the presence of the binding protein and transcription regulator AlbA, a resistance mechanism against albicidin identified in Klebsiella oxytoca, we designed a transcription reporter assay. In addition, by screening shorter albicidin fragments, as well as various DNA-binders and gyrase poisons, we were able to gain insights into the AlbA target spectrum. We analysed the effect of mutations in the binding domain of AlbA on albicidin sequestration and transcription activation, and found that the signal transduction mechanism is complex but can be evaded. Further demonstrating AlbA's high level of specificity, we find clues for the logical design of molecules capable of avoiding the resistance mechanism.


Bacteria, media and antibiotics
E. coli DSM1116, E. coli BW25113, S. Typhimurium TA100, B. subtilis DSM10, M. luteus DSM1790, M. phlei DSM750 were used for minimal inhibitory concentration (MIC) assays. E. coli strain DSM 1116 was used for all susceptibility testing. E. coli TOP10 or BL21 Star (DE3) cells (Invitrogen) were used for cloning and protein expression. The plasmids used in this study are listed in table S1. LB medium (Lysogeny broth: 10 g/L peptone, 5 g/L yeast extract, 5 g/L NaCl) was used as broth or in agar plates for all experiments with the exception of MIC assays where MHBII (BBLTM Mueller-Hinton Broth II, Becton, Dickinson and Company, New Jersey, USA) was used. Incubation steps took place at 37°C and 180 rpm shaking if not stated otherwise. Throughout the study, aza-His albicidin (compound 2) was used as albicidin standard due to its better stability and similar activity compared to natural albicidin 1 . All oligonucleotides and restriction enzymes were purchased from Thermo Scientific and all DNA sequencing reactions were performed by Microsynth AG.
[P] is the protein concentration used in the reaction mixture and n was set to 1. The uncertainty of each measurement point was determined from the normalized background signals in each gel lane.

Cloning of the transcription reporter system
For the reporter plasmid backbone, we chose the vector pCS-PesaRlux 2 which was a gift from Cynthia Collins (Addgene plasmid #47640). Using Gibson assembly 3 the PesaR promoter was removed and the lux reporter gene cassette was replaced with the improved ilux 4 operon from the vector pGEX-iluxCDABE-frp, which was generous gift from Carola Gregor. The ilux operon contains a FMN reductase to faster recycle the necessary co-factor to generate a strongly enhanced luminescence signal 4 . The pCS vector backbone was amplified by PCR using the primers pCS-fw and pCS-rev (Table S5) and the ilux cluster was amplified using the primer pair Ilux-fw and Iluxrev at annealing temperatures of 64.5 °C and 63.6 °C respectively. The two resulting fragments were purified from gel and incubated at a 1:4 ratio (backbone to ilux cluster fragment) in Gibson mix for one hour at 50°C before transformation of TOP10 cells and selection on LB agar plates with 50 µg/mL kanamycin. A cloning site and ribosomal binding site (RBS) were inserted in the new vector pCS-ilux via site-directed ligase-independent mutagenesis 5 using the primer pairs pCS-MCS-RBS-fwt/pCS-MCS-RBS-rev and pCS-MCS-RBS-fw/pCS-MCS-RBS-revt (annealing temperatures of 61.9 °C and 67.3 °C, respectively), yielding the construct pCS-MCS-RBSilux which contained restriction sites for the enzymes BamHI and XhoI upstream of the RBS and ilux cassette (vector map Figure S2A). To generate the reporter plasmids, we then inserted the pAlbA promoter or, as a test system, the T7 promoter. The promoters were obtained as oligonucleotides with BamHI and XhoI restriction sites adjacent to the promoter regions (see Table S5). The promoter oligonucelotides were dissolved in buffer (10 mM Tris, 50 mM NaCl, 1 mM EDTA, pH 7.5) to reach a concentration of 100 µM and annealed at 95 °C for 5 min followed by slow cooling to RT to yield double-stranded DNA. The promoter fragments and pCS-MCS-RBS-ilux were treated with BamHI and XhoI according to the manufacturer's protocol, purified using a DNA purification kit (Thermo) and ligated with T4 ligase to yield the plasmids pCS-pAlbA-ilux and pCS-T7-ilux ( Figure S2B). The ligation mixtures were added to TOP10 cells and positive clones were selected on LB agar plates with 50 µg/mL kanamycin. For co-expression with the reporter plasmid, AlbA was cloned into pET15b (Novagen) with an N-terminal His-tag. The AlbA gene was amplified from the expression vector pET28a-AlbAL 6 using the primers AlbA_XhoI_fw and AlbA_BamHI_rev containing the corresponding restriction sites. Restriction cloning was performed as described above and positive clones were on LB agar plates with 100 µg/ml ampicillin. After test measurements suggested that the N-terminal His-tag might affect DNA binding of AlbA, site-directed ligase-independent mutagenesis with the primer pairs deltaHis-AlbA-fwt/deltaHis-AlbA-rev and deltaHis-AlbA-fw/deltaHis-AlbA-revt was performed to yield the expression construct pET15b-ΔHisAlbA. The correct sequence of all constructs was confirmed by plasmid sequencing.

Mutagenesis
All AlbAL mutants were generated using the site-directed ligase-independent mutagenesis strategy 5 with either pET28a-AlbAL 6 or pET15b-ΔHis-AlbA as PCR template. Primer pairs with overhangs containing the mutations were designed for each mutant (Table S6). The ligation mixtures were added to TOP10 cells and positive clones were selected on LB agar plates with 50 µg/mL kanamycin for pET28a constructs and 100 µg/mL ampicillin in case of pET15b based constructs. The correct sequence of all constructs was confirmed by plasmid sequencing.
Protein expression and purification N-terminally His-tagged AlbA was expressed using a pET28a vector as described previously 6 . In brief, E. coli BL21(DE3) cells were transformed with pET28-AlbA plasmid and grown in LB medium with kanamycin. LB broth supplemented with 50 µg/mL kanamycin was inoculated from over-night cultures and grown at 37 °C until an OD600 of 0.6 was reached. After induction with 0.2 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) the proteins were expressed at 18 °C for 20 hrs. The cells were harvested by centrifugation (4,000 x G, 20 min), resuspended in lysis buffer (50 mM Tris, pH 8.0, 500 mM NaCl, 10% glycerol) and processed in a cell disruptor (Constant Systems Ltd) at 25 kPsi before centrifugation at 50,000 x G, at 4 °C for 45 min. The lysate was incubated with Pure Cube Ni-NTA agarose (Cube Biotech, Germany) at 4 °C for 20 min while shaking. The slurry was transferred to empty gravity-flow columns (ThermoFisher) and after a washing step with lysis buffer, the protein was eluted in two steps with 250 mM imidazole and 500 mM imidazole in the lysis buffer, respectively. His6-AlbA was buffer exchanged on PD10 Desalting Columns in assay buffer (50 mM sodium phosphate, pH 7.0, 100 mM NaCl) and the purity of the proteins was assessed on SDS-PAGE ( Figure S7). All AlbA mutants were purified following the same protocol. Concentrations were determined with a nano-photometer P330 (Implen, Munich, Germany) spectrophotometer using extinction coefficients calculated with the ProtParam tool 7 before storage at -80 °C.
Transcription activation assay E. coli BL21(DE3) were transformed with pET15b-AlbA or pET15b-ΔHisAlbA and pCS-pAlbA-ilux. An over-night culture was grown in LB medium with kanamycin and ampicillin to inoculate three 50 mL cultures of BL21(DE3):pET15b-AlbA + pCS-pAlbA-ilux or BL21(DE3):pET15b-ΔHisAlbA + pCS-pAlbA-ilux in LB medium with 50 µg/ml kanamycin and 100 µg/ml ampicillin. The cultures were grown to OD600 0.4-0.6 and transferred to white, clear bottom 96-well plates (Corning Inc.). To 190 µL culture in each well, 10 µl DMSO with albicidin or one of its derivatives were added. In assays to test concentration dependence of the promoter, aza-His albicidin (2) was added to reach final concentrations of 0, 0.5, 1.0, 1.5, 2.0 and 5 µM. In transcription response assays with albicidin fragments and albicidin derivatives, 1.5 µM or 2.0 µM final concentration was added of each compound. In assays with ciprofloxacin, novobiocin, acridin orange, Azid MegaStokes, tunicamycin, tetracycline and reserpine, each compound was dissolved in DMSO and added to cultures at final concentrations of 15 µM. Bisbenzimide (Hoechst 33342) was dissolved in DMSO and added to cultures at a final concentration of 5 µM. Sixty points were measured on a TECAN Infinite 200 microplate reader, after orbital shaking for 10 min, with an amplitude of 3 mm, and a wait time of one minute between each measurement. At each time point, the absorbance at 600 nm (9 mm bandwidth and 25 flashes) and the luminescence (500 ms integration time, no attenuation) were measured. Each assay was measured in triplicate at 30 °C. The luminescence data was normalized to the OD before the background signal (wells with no albicidin added) was subtracted and the mean and standard deviation of each measurement point was calculated. Each measurement series was normalized to 2 as a standard, yielding relative luminescence curves of which the maximum within the first five hours of measurement was extracted to compare the luminescence output for each compound. To compare the transcription activation of AlbA mutants, pET15b-ΔHis-constructs of each mutant together with pCS-pAlbA-ilux were used to transform E. coli BL21(DE3). BL21(DE3):pET15b-ΔHisAlbA-mutants + pCS-pAlbAilux were grown over-night in 250 µL cultures in 96-well round bottom plates with LB medium with kanamycin and ampicillin (4 replica per mutant per plate) to inoculate an equal number of 250 µL cultures with a starting OD600 of 0.02. After six hours, 2 x 100 µL of each culture were transferred to a clear bottom plate and 2 dissolved in 100% DMSO or DMSO was added to a concentration of 1.5 µM (1% DMSO final concentration). Luminescence and absorbance were measured as described above but with 40 time points instead of 60 time points. The measurement was repeated with four more plates, yielding data of 20 biological replica. For each well, the luminescence data was normalized to the OD before the background signal (wells with no albicidin added) was subtracted. The maxima of all twenty luminescence curves of each mutant were determined and differences between the transcription activity of the AlbA mutants and AlbA wild-type were analysed in Origin2022, academic version 9.9.0.225. Wilcoxon-Mann-Whitney tests for statistical evaluation of the data were performed using the Triola Statdisk 13.0 package using the Wilcoxon test for independent samples with a significance of p ≤ 0.01 or p ≤ 0.05.

E. coli growth inhibition assay
To test the antibacterial activity of albicidin derivatives in presence and absence of AlbA, susceptibility assays in liquid culture were conducted. An overnight culture of E. coli DSM 1116 was used to inoculate 50 mL LB which was grown to an OD ~1.0. The culture was diluted in LB medium to using 0.5 McFarland standard (1.5 x 10 8 cells/mL) and added to a clear round bottom 96-well plate (microdilution tray) with LB to obtain 1 x 10 6 cells/mL. Each well contained 85 µL inocolum to which either 10 µL AlbA (final concentration 10 µM) or buffer (50 mM sodium phosphate pH 7.0, 100 mM NaCl) were added. The albicidin derivatives were dissolved in 100% DMSO and 5 µL were added to the wells to obtain a final concentration of 10 µM compound and 5% (v/v) DMSO in a final culture volume of 100 µL. For the assays with mutants, 10 µM of the mutant proteins were added instead of wild-type AlbA and E. coli growth in presence of 10 µM aza-His albicidin (2) was measured. All assays were set up in triplicates. The plates were incubated at 37 °C in the dark without shaking for 20 h before documentation using a biostep scanner (Epson) with argusX.

MIC determination
Minimal inhibitory concentration (MIC) values were determined according to the ninth edition of the Approved Standard M07-A9 in microdilution assays. The test was carried out for six different bacterial strains (Gram-negative: E. coli DSM1116, E. coli BW25113, S. Typhimurium TA100; Gram-postive: B. subtilis DSM10, M. luteus DSM1790, M. phlei DSM750). 20 mL LB medium were inoculated with 20 μL of cryo stock of each strain and incubated overnight at 37 °C, 200 rpm shaking. The test inoculum was adjusted by the 0.5 McFarland Standard (OD625 from 0.08 to 0.1). Within 15 min of preparation, the adjusted inoculum suspension was diluted in MHBII medium so that each well contained approximately 5 × 10 5 cells/mL in a final volume of 100 μL. 95 μL of the inoculum were applied per well and 5 μL of the albicidin derivative solutions were added. Previously, the dry antibiotic compounds were dissolved in DMSO (100%) with a concentration of 2560 μg/mL and then further diluted in DMSO (100%) for testing. 5 μL of each antibiotic dilution were applied to the microdilution tray to reach final concentrations of 8 μg/mL to 0.016 μg/mL. One row of each 96-well plate served as a growth control without antibiotic substances and another row of the microdilution tray served as sterility control (only MHB II-media). The antimicrobial effect of the solvent (DMSO) was tested by adding 5 μL DMSO to several wells. Purity checks and cell titer controls were performed according to International Standard M07-A9 on Mueller-Hinton II Agar. Both microdilution trays and agar plates were incubated at 37 °C for 20 h and subsequently analyzed by naked eye and documented using a biostep scanner (Epson) with argusX.

DNA-Gyrase inhibition testing
DNA-supercoiling experiments with DNA-gyrase were performed using an E. coli gyrase supercoiling kit (Inspiralis Limited, Norwich, UK) following the manufacturer's instructions. A total volume of 30 μL gyrase buffer contained 0.5 μg relaxed pBR322 plasmid DNA (inspiralis Limited), 1 U DNA-gyrase (6 U/μL) (Inspiralis Limited) and the albicidin derivatives 1, 2 and 10−22 at a final concentration of 45 nM. The final DMSO concentration was 3%. Samples were incubated at 37 °C for 30 min and subsequently loaded on an agarose gel. Electrophoretic analysis was performed using a 1% agarose gel (100 V, 90 min). DNA bands were stained with ethidium bromide, visually analyzed and documented with a Biostep BioView transilluminator and INTAS GelDoc.

CD spectroscopy
The native fold of the purified AlbA mutants was evaluated using CD spectroscopy. The measurements were conducted on a J-815 CD spectrometer (Jasco, Groß-Umstadt, Germany). AlbA and its mutants were diluted in buffer (25 mM sodium hydrogen phosphate, pH 7.0) to reach a concentration of 5 µM. Far-UV spectra were acquired at 20 °C and 30 °C between 190-260 nm with a path length of 0.1 cm and a bandwidth of 1 nm at a scanning speed of 50 nm/min in 5 accumulations. The data pitch was set to 0.1 nm. The spectra were processed with Spectra Manager (JASCO) and the mean residue ellipticity (MRE) was calculated Where θ is the ellipticity and l, c, and n denote the path length, molar concentration and number of amino acids.

Tryptophan fluorescence quenching
Purified His-tagged AlbA and mutant proteins were diluted to a concentration of 50 nM in 200 µl binding assay buffer (50 mM sodium phosphate pH 6.8, 150 mM NaCl). Aza-His albicidin (2) solutions with concentrations of 78 nM, 156 nM, 312.5 nM, 625 nM, 2.5 µM, 5 µM, 20 µM and 40 µM were prepared in 100% DMSO. Of each dilution, 2 µl were added to the protein samples (1% DMSO final concentration) before incubation for 20 min at RT in the dark. The samples were transferred to a Hellma quartz fluorescence cuvette with a path length of 10 mm and tryptophane fluorescence was measured on a PerkinElmer LS 55 fluorescence spectrometer. The excitation was set to λex = 282 nm and emission was recorded between 300 nm and 450 nm with a scanning speed of 60 nm s -1 . The slit width was set to 8 nm for excitation and 4 nm for emission, respectively. All measurements were performed in triplicate and standard deviations are given for the Kd and Hill factor n. The quenching was determined from the normalized integrated emission band (300-400 nm) of each titration step subtracted from the emission band of the protein without ligand. The normalized tryptophan fluorescence emission was fitted to the equation (3) using Excel Solver 21 and Origin2022: [L] is the total ligand concentration, n is the Hill coefficient, and Bmax is the maximum binding capacity of the protein.
To test the binding of the compounds 10-22 to AlbA, 50 nM compound were combined with 50 nM AlbA in 200 µl binding assay buffer (with 1% DMSO final concentration). Compound 2 was measured as a control. Tryptophan fluorescence quenching was measured in triplicate as described above after incubation for 20 min at RT in the dark. The quenching was determined from the integrated emission bands (300-400 nm) and calculated as the ratio between free AlbA (I0) and liganded AlbA (I).

Molecular docking
Using UCSF Chimera 8 and AutoDock Vina 9 , AlbAS 6 (PDB-ID: 6et8) was docked with the bisbenzimide structure calculated by pubchem (CID:1464, simulated according to Kim et al. 10 ). A monomer molecule of AlbAS without albicidin was prepared as the receptor and docking was conducted using standard parameters with the receptor search volume covering the albicidin binding pocket. The docking results were manually inspected and the highest scoring complex was selected for comparison to the albicidin-AlbA complex.
AlbA structure modelling AlphaFold v2.1.0 11,12 was used to predict the structure of full-length AlbA. The program was run in the ColabFold 13 notebook (AlphaFold.ipynb) with MMseqs2 and HHsearch to generate sequence alignments and templates and the implemented Amber software to relax the best-ranked structure. For modelling of full-length AlbA monomers, default parameters were used. For modelling dimers, the program was run using template_mode pdb70, the multiple sequence alignments were set to unpaired_paired mode and alphafold2_multimer_v3 was chosen as modelling type with auto settings on recycling parameters.

Synthesis of albicidin derivatives
Details of the synthesis routes for albicidin and the described albicidin analogues and the identifying MS and NMR spectra are given in the following publications and patents. The total synthesis of albicidin has been described by Kretz et al. 14 . The syntheses of compounds 2, and 18 have been described by Grätz et al. 15 and compounds 12, 14, 15, 19, 21 have been described by Behroz et al. 1 . Procedures for the synthesis of albicidin fragments 3, 4, 6, 7 and 9 have been described in Vieweg et al. 16 , and compounds 10, 11 and 13 are described in Michalczyk et al. 17 . The syntheses of the AB-PCP ester and C-terminally allyl-protected tripeptide are described in Behroz et al 1 . The dipeptide 5b (101 mg, 250 µmol, 1.00 eq.) and PCP activated AB building block 1 (5a, 150 mg, 275 µmol, 1.10 eq.) were dissolved in anhydrous DMF and Et3N (174 µL, 5.0 eq.) was added. The reaction mixture was stirred at room temperature for 16 h. All volatiles were removed in vacuo and the residue was taken up in a mixture of THF and MeOH (1:1 v/v), and 3 M KOH(aq.) was added dropwise at 0 °C. The ice bath was removed and after 15 min of stirring the suspension was acidified to pH ≈ 2 by the addition of 3 M HCl(aq.). The resulting suspension was concentrated under reduced pressure and the crude material was dissolved in DMSO and purified by HPLC (PLRP-S column, CH3CN in H2O). BCD fragment 5 (11.2 mg, 7 % over two steps) was obtained as white powder.

Boc(C, Pom)-D-OMe (8a)
To a solution of the amino acid 8a (800 mg, 2.16 mmol, 1.50 eq.) in dry THF was added EEDQ (570 mg, 2.30 mmol, 1.60 eq.) and the reaction mixture was stirred at room temperature for 15 min. A premixed solution of methyl 4-aminobenzoate (218 mg, 1.44 mmol, 1.00 eq.) in dry THF was added slowly and the reaction mixture was stirred at room temperature for 72 h. The organic solvent was evaporated under reduced pressure and the residue was partitioned between EtOAc and 1 M HCl(aq.). The organic layer was washed with 1 M HCl(aq.) (2x), brine, dried over anhydr. Na2SO4, and concentrated in vacuo to afford the crude product, which was purified by column chromatography (SiO2, Hexane/EtOAc -5:1). Compound 8b (310 mg, 43 %) was obtained as yellow solid.

BCD fragment (8)
The dipeptide 8b (250 mg, 497 µmol, 1.00 eq.) was dissolved in 4 M HCl in 1,4-dioxane and stirred at room temperature for 2.5 h. Subsequently, all volatiles were removed under reduced pressure to obtain the crude product, which was taken up in H2O/CH3CN and freeze-dried to afford the Boc-deprotected product as slightly yellow powder. The unprotected dipeptide (114 mg, 284 µmol, 1.00 eq.) and Fmoc-protected acid chloride (150 mg, 397 µmol, 1.40 eq.) were dissolved in anhydrous THF and Et3N (395 µL, 10.0 eq.) was added. The reaction mixture was stirred at room temperature for 16 h. All volatiles were removed in vacuo and the residue was taken up in a mixture of THF and MeOH (1:1 v/v), and 3 M KOH(aq.) was added dropwise at 0 °C. The ice bath was removed and after 15 min of stirring the suspension was acidified to pH ≈ 2 by the addition of 3 M HCl(aq.). The resulting suspension was concentrated under reduced pressure and the crude material was dissolved in DMSO and purified by HPLC (PLRP-S column, CH3CN in H2O). BCD fragment 8 (7.53 mg, 7 % over three steps) was obtained as white powder.

B(Pyr)-C(azaHis)-D(Pyr)-E(iPr)-F(pABA) (16)
The Tetrapeptide 16b 17 (100 mg, 135 µmol, 1.00 eq.) and PCP activated AB building block 18 (16a, 81.3 mg, 149 µmol, 1.10 eq.) were dissolved in anhydrous DMF and Et3N (132 µL, 7.0 eq.) was added. The reaction mixture was stirred at room temperature for 16 h. All volatiles were removed in vacuo and the residue was taken up in a mixture of THF and MeOH (1:1 v/v), and 3 M KOH(aq.) was added dropwise at 0 °C. The ice bath was removed and after 15 min of stirring the suspension was acidified to pH ≈ 2 by the addition of 3 M HCl(aq.). The resulting suspension was concentrated under reduced pressure and the crude material was dissolved in DMSO and purified by HPLC (PLRP-S column, CH3CN in H2O). Derivative 16 (9.98 mg, 8% over two steps) was obtained as white powder.

tert-Butyl-(S)-(1-hydroxypent-4-yn-2-yl)carbamate (17b)
Lithium aluminium hydride (3.56 g, 93.8 mmol, 2.00 eq.) was added to dry THF (100 mL) at 0°C under a N2atmosphere. To the reaction mixture was added a solution of (S)-2-((tert-butoxycarbonyl)amino)pent-4-inic acid (17a, 10.0 g, 46.9 mmol, 1.00 eq.) in THF (20 mL) over a period of 20 min. The mixture was stirred for 12 h at room temperature until the reaction was completed. The reaction mixture was diluted with ethyl acetate (50 mL), quenched by the addition of water (50 mL) and filtrated over Celite ® . The mixture was extracted by ethyl acetate (3 x 30 mL) and washed by water (30 mL). The combined organic phases were dried over Na2SO4, filtered and the solvent was removed by rotary evaporation. The product 17b (7.78 g, 77.9 mmol, 83 %) was obtained as white solid and used in the next step without further purification. The protocol was based on a literature known procedure used for the synthesis of a similar product 19 . Triethylamine (9.12 mL, 65.2 mmol, 2.00 eq.) and p-toluenesulfonyl chloride (7.46 g, 39.1 mmol, 1.20 eq.) were added to a solution of alcohol 17b (6.50 g, 32.6 mmol, 1.00 eq.) in CH2Cl2 (50 mL) at 0 °C. The reaction mixture was stirred for 3 h at room temperature and after completion of the reaction the volatiles were removed by rotary evaporation. The residue was taken up by CH2Cl2 (50 mL) and washed by an aqueous solution of HCl (1 M, 2 x 30 mL) and a saturated aqueous solution of NaCl (30 mL). The organic phase was dried over Na2SO4, filtered and concentrated by rotary evaporation. The crude product was purified by column chromatography (SiO2, 20:1 → 7:1, cyclohexane:ethyl acetate) and the product (17c, 7.11 g, 20.2 mmol, 62 %) was obtained as a yellow solid. The protocol was based on a literature known procedure used for the synthesis of a similar product 20 .

Allyl-5-(allyloxy)picolinate (17e)
To a suspension of potassium carbonate (2.61 g, 21.6 mmol, 3.00 eq.) in DMF (10 mL) was added allyl bromide (1.86 mL, 21.6 mmol, 3.00 eq.) and a solution of 5-hydroxypicolinic acid (17d, 1.00 g, 7.19 mmol, 1.00 eq.) in DMF (5 mL) was added dropwise at 0 °C. The reaction mixture was stirred for 16 h at room temperature, after completion of the reaction diluted by water (50 mL) and extracted by diethyl ether (4 x 30 mL). The combined organic phases were concentrated by rotary evaporation and the product (17e, 1.23 g, 16.8 mmol, 78 %) was obtained as orange oil and used without further purification in the next step.

5-(Allyloxy)picolinic acid (17f)
To a solution of allylester (17e, 1.23 g, 5.63 mmol, 1.00 eq.) in THF (10 mL) was added a solution of lithium hydroxide (943 mg, 39.4 mmol, 7.00 eq.) in water (10 mL) and the reaction mixture was stirred for 16 h at room temperature. After completion of the reaction the organic solvent was removed by rotary evaporation and the residue was extracted by CH2Cl2 (5 x 10 mL). The combined organic phases were concentrated by rotary evaporation and the product (17f, 757 mg, 4.22 mmol, 75 %) was obtained as orange solid.

Allyloxy-D(N)-E(OBn)-F(OBn, OBn) (17h)
Carboxylic acid (17f, 525 mg, 2.93 mmol, 1.20 eq.) was suspended in thionyl chloride (5 mL) and stirred for 2 h at 90 °C. The reaction mixture was allowed to cool down to room temperature and the volatiles were removed by rotary evaporation. The residue was taken up in dry THF (5 mL) and added to a cooled solution of benzyl protected dipeptide 17g 1 (1.51 g, 2.44 mmol, 1.00 eq.) and triethylamine (680 µL, 4.88 mmol, 2.00 eq.) in dry THF (5 mL) at 0 °C and stirred for 16 h at room temperature. To the reaction mixture was added diethyl ether (70 mL), the precipitated product was filtered and the solid was dissolved in CH2Cl2 (30 mL). The organic phase was washed by an aqueous solution of HCl (1 M, 2 x 30 mL) and a saturated aqueous solution of NaCl (30 mL). The organic phase was dried over Na2SO4, filtered and concentrated by rotary evaporation. Tripeptide (17h, 1.80 g, 2.78 mmol, 95 %) was obtained as yellowish solid. To a solution of allyl-protected alcohol (17h, 1.81 g, 2.32 mmol, 1.00 eq.) in dry THF (10 mL), morpholine (4.00 mL, 46.2 mmol, 20.0 eq.) and Pd(PPh3)4 (536 mg, 464 µmol, 0.20 eq.) were added and the reaction was stirred for 3 h at room temperature. After completion of the reaction the volatiles were removed by rotary evaporation and the residue was taken up with ethyl acetate. The organic phase was washed by an aqueous solution of HCl (1 M, 2 x 30 mL) and a saturated aqueous solution of NaCl (30 mL). The organic phase was dried over Na2SO4, filtered and concentrated by rotary evaporation. The crude product was purified by column chromatography (SiO2, 10 % → 40 % ethyl acetate in cyclohexane) alcohol (17i, 1.13 g, 1.53 mmol, 66 %) was obtained as yellowish solid.

Boc-azaHis(POM)-D(N)-E-F (17l)
Tetrapeptide (17k, 442 mg, 410 µmol, 1.00 eq.) was dissolved in THF (10 mL) and a stream of N2-gas was bubbled through the solution for 5 min. Palladium on charcoal (10 wt. %, 44.2 mg) was added to the reaction mixture which was stirred for 16 h at room temperature. After completion of the reaction the mixture was filtered over Celite® and the filtrate was concentrated by rotary evaporation. The product (17l, 311 mg, 385 µmol, 94 %) was obtained as colorless solid.

AzaHis-C(POM)-D(N)-E-F (17m)
Tetrapeptide (17l, 331 mg, 410 µmol, 1.00 eq) was suspended in a solution of 4 M HCl in 1,4-dioxane (5 mL) and the reaction mixture was stirred for 3 h at room temperature. After completion of the reaction, the volatiles were removed by rotary evaporation, the residue was taken up by water and acetonitrile and lyophilized. The amine (17m, 281 mg, 398 µmol, 97 %) was obtained as white solid.  To a solution of the amino acid 20a (100 mg, 296 µmol, 2.00 eq.) in dry THF was added EEDQ (73.1 mg, 296 µmol, 2.00 eq.) and the reaction mixture was stirred at room temperature for 15 min. A premixed solution of the tripeptide 20b 18 (87.0 mg, 148 µmol, 1.00 eq.) in dry THF was added slowly and the reaction mixture was stirred at room temperature for 72 h. The organic solvent was evaporated under reduced pressure and the residue was partitioned between EtOAc and 1 M HCl(aq.). The organic layer was washed with 1 M HCl(aq.) (2x), brine, dried over anhydr. Na2SO4, and concentrated in vacuo to afford the crude product, which was purified by column chromatography (SiO2, Hexane/EtOAc -8:1).

Cbz/Allyl-protected amino tetrapeptide precursor (20d)
A solution of the Boc-protected tetrapeptide 20c (500 mg, 550 µmol, 1.00 eq.) in 4 M HCl in 1,4-dioxane (10 mL) was stirred at room temperature for 2 h. Subsequently, all volatiles were removed under reduced pressure to obtain the crude product, which was taken up in H2O/CH3CN and freeze-dried to afford the Boc-deprotected amine 20d (465 mg, quant) as yellow/orange powder.

Unprotected guanidino tetrapeptide (20f)
The tetrapeptide 20e (150 mg, 205 µmol, 1.00 eq.) was dissolved in 4 M HCl in 1,4-dioxane and stirred at room temperature for 3 h. Subsequently, all volatiles were removed under reduced pressure to obtain the crude product, which was taken up in H2O/CH3CN and freeze-dried to afford the Boc-deprotected product as yellow powder. HRMS (ESI): m/z calculated for C34H35N8O11 (M+H) + 731.2420, found 731.2417. Boc deprotected tetrapeptide (115 mg, 124 µmol, 1.00 eq.) was dissolved in dry THF (5 mL) and the solution was purged with argon for 10 min. Palladium (10 wt. % on activated carbon, 17 mg) was added, the reaction mixture was saturated with H2 and stirred at room temperature for 5 h under hydrogen atmosphere. The suspension was filtered over Celite®, concentrated under reduced pressure, and freeze-dried to afford the Cbz-Boc-deprotected tetrapeptide 20f (90 mg, quant over two steps) as yellow powder.

HCl*H-Hyp-pABA-HMpABA-HMpABA-OH (22g)
The protected amine (22f, 35.0 mg, 0.051 mmol, 1.00 eq) was dissolved in absolute DCM (5 mL), trifluoroacetic acid (2 mL) was added and the reaction mixture was stirred for 3 h at room temperature. The volatiles were removed by rotary evaporation and the residue was taken up with acetonitrile and an aqueous solution of HCl (1 M). The product (22g, 31.2 mg, 0.051 mmol, 98 %) was obtained after freeze drying as yellowish solid and used in the next step without further purification.
To a solution of tetrapeptide (22g, 31.2 mg, 0.051 mmol, 1.00 eq) in absolute DMF (2 mL) was added triethylamine and activated AB building block 1 (5a, 30.7 mg, 0.056 mmol, 1.11 eq) and the reaction mixture was stirred for 16 h at room temperature. The resulting mixture was diluted with acetonitrile and filtered. The solid was dissolved in a mixture of THF/H2O/acetonitrile (3:1:3) and concentrated under reduced pressure. After freeze-drying and preparative RP-HPLC the final product (22,25.0 mg, 0.029 mmol, 57 %) was obtained.