*2.4. Production of Ncps by N. edaphicum CCNP1411*

Apart from the structural analysis, we also made attempts to determine the relative amounts of the individual Ncps produced by *N*. *edaphicum* CCNP1411. To exclude the effect of the extraction procedure on the amounts of the detected peptides, different solvents and pH were applied. As the process of Ncp linearization during long storage of the freeze-dried material was suggested [16], both the fresh and lyophilized biomasses were analyzed. Regardless of the extraction procedure, Ncp-A2-L with Phe in C-terminus was always found to be the main Ncp analogue (Figure 10A–D). In addition, when MePro and Pro-containing peptide were compared separately, the peak intensity of the linear Ncps with Phe in C-terminus (i.e., Ncp-A2-L and Ncp-E1-L) was higher than the Ncps with Leu. These results might indicate preferential incorporation of Phe into the synthesized peptide chain.

**Figure 10.** Relative cell contents of nostocyclopeptides extracted from 10 mg of lyophilized (**A**,**B**,**D**) or 500 mg fresh (**C**) biomass of *N*. *edaphicum* CCNP1411 with MilliQ water (**A**), 50% methanol in water (**B**), and 20% methanol in water (**C**,**D**). The cell content was expressed as peak intensity in LC-MS/MS chromatogram.

The study also showed that the cell contents of the linear Ncps are higher than the cyclic ones. (Figure 10A–D and Figure S8). The release of Ncps from the synthetase as linear aldehydes is catalyzed by a reductase domain, located in the C-terminal part of the NcpB [17]. This reductive release triggers the spontaneous, and enzyme independent, macrocyclization of the linear peptide [19,20]. The reaction leads to the formation of a stable imino bond between the C-terminal aldehyde and the N-terminal amine group of Tyr [19,20]. In *N*. *edaphicum* CCNP1411 cells, depending on the Ncp analogue, the analyzed material (fresh or lyophilized) and extraction solvent, the cyclic Ncps constituted from even less than 10% (Ncp-A2, fresh biomass) to over 90% of the linear peptide (Figure 10A–D and Figure S8). In case of Ncp-A1, with MePro-Leu in C-terminus, the contribution of the cyclic form was always most significant, and at pH 8 it reached up to 91.7% of the linear peptide aldehyde (Ncp-A1-L) (Figure 10A–D and Figure S8). The cyclic analogues, Ncp-E1 and Ncp-E3 were produced in trace amounts and their spectra were sporadically recorded. It was proven that the macrocyclization process of Ncps is determined by the geometry of the linear peptide aldehyde and the conformation of D-Gln and Gly is crucial for the folding and formation of the imino bond [19]. As these two residues are present in all detected Ncps, then, probably other elements of the structure affect the cyclization process, as well. We hypothesize that due to the steric hindrances, the cyclization of Ncp-A1 with Leu in C-terminal position is easier than the cyclization of Ncp-A2 with Phe. As a consequence, the proportion of the cyclic Leu-containing Ncp-A1 to the linear form of the peptide is higher.

Thus far, Ncps synthesis was reported in few *Nostoc* strains, and the structural diversity of the peptides was found to be low. Other classes of NRPs were detected in cyanobacteria representing different orders and genera, and within one class of the peptides numerous analogues were detected. For example, the number of naturally produced cyclic heptapeptide microcystins (MCs), is over 270 [40,41] and in one cyanobacterial strains even 47 MCs analogues were detected [40]. Cyanopeptolins are produced by many cyanobacterial taxa and so far more than 190 structural analogues of the peptides have been discovered [41]. In this work, cyanopeptolin gene cluster was identified in *N*. *edaphicum* CCNP1411 and thirteen products of the genes were previously reported [6]. These peptides contain seven amino acids and a short fatty acid chain, and only one element of the structure, 3-amino-6-hydroxy-2-piperidone (Ahp), is conserved [6]. The structural diversity of NRPs is generated by frequent genetic recombination events and point mutations in the NRP gene cluster. The changes in gene sequences affect the structure and substrate specificity of the encoded enzymes. The tailoring

enzymes can further modify the product, leading to even higher diversity of the synthetized peptides [42]. In case of Ncps, both the number of the producing organisms and the structural diversity of the peptides are limited. Ncp-M1 from *Nostoc* living in symbiosis with gastropod [16] is the only Ncp with structure distinctly different from Ncp-A1, Ncp-A2 and other Ncps described in this work.

The diversity within one class of bioactive metabolites offers a good opportunity for structure-activity relationship studies, without the need to synthesize the variants. The studies are of paramount importance when the efficacy and safety of a drug candidate are optimized. Therefore, in our future work, when sufficient quantities of pure Ncps are isolated, the activity of individual analogues against different cellular targets will be tested and compared, in order to select the lead compound for further studies.
