**3. Results**

#### *3.1. Bacterial Growth and Fe-CAS-Plate Reactions*

Comparative phenotypic analysis was achieved by allowing each strain to develop for 5 days on their respective agarized media under conditions that promoted the best growth. This elapsed time ensured stationary phase was reached for each representative, which resulted in the formation of sufficient colonies for analysis. Triplicate CAS-supplemented and CAS-free controls were simultaneously plated to identify the viability of inoculum. In all cases, growth occurred on CAS plates, but was marginally reduced in comparison to controls. Both after 3 and 5 days, the average zone of colour change (blue to yellow) was recorded, revealing several phenotypic attributes. As some strains grew slower than others, 5 days' period was chosen to analyze and compare all simultaneously. We identified siderophore production based on zone of diffusion/colour change around colonies as follows (Figure A1): no zone (−), a zone <1 mm around colonies (+), a moderate diffusion <10 mm (++), and considerable diffusion >10 mm (+++). Fe-chelating CAS reactions after 5 days are listed in Tables 1 and 2, as well as shown in Figure 2 beside each strain name.

#### *3.2. Substitute Cation CAS Assays*

Once Fe-chelating siderophore production was confirmed using the standard Fe-CAS assay, all strains were tested on CAS supplemented agar plates that contained one of 9 other metal(loid) cations: Mg2+, V3+, Mn2+, Co2+, Ni2+, Cu2+, Zn2+, Se4+ and Te2+. Results for varied CAS assays are listed along-side Fe2+ data in Tables 1 and 2 using the same zone distinctions as described. While some strains only produced siderophores that reacted with Fe2+, others had secondary metabolites capable of chelating additional metal(loid) cations. Most strains capable of chelating metals other than Fe could also chelate iron, with the exception of two strains P4 and SS56 (Figure A2), which were found to produce metallophore capable of acting predominantly on other tested metal(loid)s, rather than Fe.

#### *3.3. Phylogenetic Diversity of Siderophore Producing AAP*

Isolates were chosen to represent AAP from a variety of environments as well as embody a host of phylogenetically diverse species. In this way, the 101 representatives listed in Tables 1 and 2 were cultivated, and 16S rRNA gene sequences acquired either from repositories, or decoded in this work. Phylogenetic relation to sequences of known type strains was determined by BLAST search (Tables A1 and A2). In addition, these sequences were used to create a phylogenetic tree (Figure 2), that also included some previously described type species not tested for siderophore production, but were included as key placeholders of phylogenetic groups. The evolutionary analysis performed on Mega X using Maximum Likelihood method involved 132 nucleotide sequences and had a total of 1717 positions in the final data set. The chosen AAP diversely spread throughout *Erythrobacteraceae* and *Sphingomonadaceae* relating to known AAP type species. While some aligned to reported AAP in *Acetobacteraceae*, and *Rhodobacteraceae*, many others aligned to organisms previously undescribed as AAP within these clades, as well as to some species within *Hyphomonadaceae* and *Methylobacteraceae*.

**Figure 2.** Phylogenetic tree of AAP tested for siderophore production. Isolates hailed from α-proteobacterial families *Erythrobacteraceae*, *Sphingomonadaceae*, *Acetobacteraceae*, *Rhodobacteraceae*, *Hyphomonadaceae* and *Methylobacteraceae*, as well as a representative within the γ- subclass of *Proteobacteria* (titled sections around circumference of circle). Fe-chelation siderophore activities are listed between strain names and phylogenetic position. Strain names in bold are confirmed AAP, red highlighted are not.


**Table 1.** Freshwater and saline AAP analyzed for metal(loid)-chelation via CAS assays with varied cations.


**Table 2.** Meromictic lake, saline spring and biological soil crust AAP analyzed for metal(oid)-chelation via CAS assays with varied cations.

#### *3.4. C. halotolerans Pigment Purification and Identification*

When purified via 16.5% tris-tricine gel electrophoresis, the brown pigment migrated further than the loading buffer's running dye, CBB, after 1 h (Figure 3A). This gel-shift revealed the pigment under study to be smaller than CBB, which has a known molecular weight of 856.03 g/mol. Since well 1 contained a standard ladder, the measurement of migration distance for the siderophores' brown band, and of each protein in the ladder allowed for the rough estimation of pigment size to be near ~341 Da (Figure S1). Gel staining and destaining revealed that the siderophore sample collected after resin concentration and diluted in methanol at 200, 20, 2, and 0.2 μg/mL loaded into wells 2 through 5, respectively, had some contaminating small proteins as expected (Figure 3B). In addition, samples that received the subsequent removal of proteins larger than 3 kDa via spin-column were also run on the same gel. Wells 6 through 8 contained brown pigment, which passed through the <3 kDa spin column and then diluted in 60% methanol to 100, 10, and 1 μg/mL, respectively. This step purified the brown pigment of any contaminating proteins (Figure 3A,B). Often, small molecules below 1 kDa are lost from the gel during destaining step [40], which likely caused the small brown pigment to escape similarly to CBB, (Figure 3B). Regardless, testing the crude dried pigment (Disk 1), as well as fractions <3 kDa and >3 kDa (Disks 2 and 3), confirmed the smaller fraction containing brown pigment acted as a siderophore, while the larger proteins did not (Figure 3C). Since the <3 kDa fraction had no contaminants (Figure 3B), but contained the brown pigment prior to destaining (Figure 3A), the small ~341 Da molecule produced by *C. halotolerans* acted as a siderophore.

**Figure 3.** Gel purification of *C. halotolerans* brown pigment, and confirmation of siderophore activity. ( **A**) Unstained and (**B**) stained tris-tricine gel electrophoresis performed on: siderophore from resin concentration, wells 2–5; siderophore sample smaller than 3 kDa, wells 6–8; proteins larger than 3 kDa and remaining in solution well 9; standard ladder, wells 1 and 10. (**C**) Siderophore activity observed for: (1) Crude resin extract positive reaction visible as yellowing area; (2) <3 kDa dark brown fraction produced positive yellowing reaction; (3) Proteins >3 kDa negative results observed as darkening of blue due to alkaline pH, without siderophore activity present.
