*2.1. Detection of Homodolastatin 16, Dolastatin 16, and Antanapeptin A in Kenyan Isolates of a Filamentous Cyanobacterium, Moorea producens*

Besides morphological appearance, the identity of the filamentous Kenyan marine cyanobacterium *M. producens* [21] was supported by the detection of a number of natural products known to be associated with this species. Following organic extraction of the sample and fractionation on C-18 silica, comparative HRFABMS against a dolastatin 15 standard, showed the presence of homodolastatin 16 (*m*/*z* [M + Na]<sup>+</sup> C48H72O10N6Na (calcd. 915.5202)), dolastatin 16 ([M + Na]<sup>+</sup> C47H70O10N6Na (calcd. 901.5046)), and antanapeptin A (C41H60O8N4 (calcd. 759.4314)) (Figure 1).

**Figure 1.** Electrospray mass spectrometry showing molecular ions of dolastatin 16, homodolastatin 16, and antanapeptin A from LCMS analysis of Kenyan *M. producens*' extracts at retention times of 11.39 min, 12.06 min, and 12.53 min, respectively. Note: dolastatin 15 included as standard. LCMS conditions: gradient of 10% 0.1% formic acid in water/90% 0.1% formic acid in acetonitrile to 5% 0.1% formic acid in water/95% formic acid in acetonitrile at 0.3 mL min<sup>−</sup>1.

The detection of molecular ions corresponding to homodolastatin 16 from *M. producens* in the current study was consistent with that of the HRFABMS *m*/*z* [M+Cs]<sup>+</sup> 1025.4364 (calcd for C48H73N6O10Cs, 1025.4364) obtained earlier [21]; however, this work represents the first report of the detection of dolastatin 16 in the Kenyan *M. producens*. The dolastatins are a group of antineoplastic pseudopeptides initially isolated from the sea hare *Dolabela auricularia* but that originate from the marine cyanobacterium *M. producens* distributed pan-tropically worldwide. Debromoaplysiatoxin (DAT) that is associated with *M. producens*' swimmers' itch was not detected by LCMS in the Kenyan strain of *M. producens*. The non-axenic nature of the Kenyan *M. producens* has been reported earlier [19] and was consistent with the isolation of the filament bacteria *Pseudoalteromonas carrageenovora* and *Ochrobactrum anthropi* from detached *M. producens*' filaments on marine agar 2216 (10% *w*/*v*).

#### *2.2. Phylogenetic Divergence of M. producens from L. majuscula*

*L. majuscula* CCAP 1446/4 strain from the Culture Collection, Oban, Scotland, United Kingdom, and 16S rDNA analysis were used to delineate the Kenyan *M. producens* from other *L. majuscula* species elsewhere. Treatment of the cyanobacteria with cycloheximide, several rinses with phosphate buffered saline (PBS), and nitrogen liquefaction ensured that detached filaments were free from eukaryotic cells, protozoa, and fungi. Further, the cyanobacteria were treated with copper sulphate powder (CuSO4·5H2O) and left to stand for 10 min, 30 min, and 60 min, respectively, to kill any available bacteria. The control was untreated. Pooled detached filaments of both *L. majuscula* CCAP 1446/4 strain and the Kenyan *M. producens* biomass, free from eukaryotic cells, protozoa, and fungi, respectively, but with dead bacteria afforded sufficient biomass for gDNA extraction and genome sequencing. Crushing and homogenization of the biomass using a bench top bead-based homogenizer (PowerLyzerTM, 5 min) (MO BIO LABORATORIES Inc., California, USA) produced a homogenate from which bacterial gDNA was exhaustively extracted and the residue retained. CuSO4·5H2O assisted differential bacterial gDNA isolation of the residue of pre-extracted non-axenic filamentous cyanobacteria biomass followed by nitrogen liquefaction and sonication generated quality gDNA of *M. producens* for genome sequencing. The gDNA isolated from the residue provided high-purity DNA with 260/280 and 260/230 ratios of between 1.90 and 2.29 in the qubit assay, respectively. Treatment of the cyanobacteria biomass for times greater than 60 min denatured the *M. producens* DNA (Figure S1). The untreated *M. producens* control did not generate any pure DNA. Sequences of the Kenyan *M. producens* at the aforementioned times of 10 min, 30 min, and 60 min were synonymous with Kenyan *Moorea producens* Rep 1, Kenyan *Moorea producens* Rep 2, and Kenyan *Moorea producens* Rep 3, respectively, shown on the phylogeny diagram (Figure 2).

Initial NCBI blasts matched the cyanobacterium with *Aminobacterium colombiense*. However, there was a distinct difference in the sizes of DNA for bacteria and cyanobacteria observable in an electrophoresis gel. Moreover, mismatches arising from the primer CYA 106F were not unusual [29]. To validate the method, it was used to identify *L. majuscula* CCAP 1446/4. Further NCBI blasts of the generated sequence of the Kenyan *M. producens*, without restricting organism identity, matched the sequence at 99% identity with an uncultured *Aminanaerobia bacterium*. With the blast restriction to cyanobacteria, the sequences matched with an uncultured cyanobacterium at 98% identity. The 16S rDNA sequences obtained from *L. majuscula* strain CCAP 1446/4 by the aforesaid method for five replicates did not show any matches to *Aminanaerobia bacteria* but consistently matched with *L. majuscula* at 100% identity. The phylogenetic divergence of Kenya *M. producens* from *L. majuscula* and its clones is shown in Figure 2 and the blasted sequences are shown in Figure S2.

**Figure 2.** Phylogeny of Kenyan *M. producens* relative to *Lyngbya majuscula* and *Lyngbya majuscula* CCAP 1446. Both *L. majuscula* and *L. majuscula* CCAP 1446 are phylogenetically far distant from the Kenyan *M. producens*. Kenyan *M. producens* Rep 1 (CuSO4·5H2O, 10 min), Kenya *M. producens* Rep 2 (CuSO4·5H2O, 30 min), Kenyan *M. producens* Rep 3 (CuSO4·5H2O, 60 min). Rep 1 and Rep 2 of *L. majuscula* CCAP 1446 are sequence replicates of the biomass treated with CuSO4·5H20 at 10 min.
