L. ferriphilum strain IESL25 was isolated from an industrial bioheap process located in northern Chile. For this, pregnant leach solution (PLS, with 30 g/L sulfate concentration) containing microorganisms and dissolved copper from the heap ore was directly streaked onto iron overlay plates containing autotrophic basal salt, trace elements (ABS/TE); (Nancucheo et al., 2016) and ferrous iron incubated at 37°C for 15 days under oxic conditions, as previously described by Rawlings et al. (2007). After the incubation period, colonies with deposition of red-brownish coloration were transferred into corresponding liquid media containing 50 mM ferrous iron, supplemented with ABS/TE pH 1.7 and incubated under same conditions (37°C for 15 days). Liquid cultures were scaled up to 50 mL and DNA was extracted for PCR-cloning and sequencing of 16S rRNA gene by vector cloning, as described previously (Okibe et al., 2003). The resulting sequence data were visualized using Chromas Lite version 2.01 and compared with gene sequences deposited in NCBI GeneBank database,¹ using the nucleotide BLAST tool. ARB software² was used in order to determine phylogenetic affiliation of the strain based on 16S rRNA sequence (Supplementary Figure S2). The 16S rRNA gene sequence of L. ferriphilum IESL25 is available from NCBI nucleotide database (Accession Number HQ902070).

L. ferriphilum IESL25 was grown in ABS/TE medium supplemented with increasing concentrations of magnesium sulfate (MgSO₄∙7H₂O, Merck) and using an inoculum equivalent to 5% of the total culture volume. Since the culture medium contains sulfate salts, total sulfates concentration was analyzed by analytical gravimetric analysis, yielding total sulfate contents of 8 g/L for the control and 40, 80, and 100 g/L for the experimental cultures. Bacterial growth was conducted in triplicate and followed by microscopy using a Neubauer cell counting chamber and a Leyca MDLS microscope equipped for fluorescence and contrast phase techniques. DAPI staining technique was also used for cells counting at the lag and early exponential growth phases. Iron oxidation by L. ferriphilum IESL25 during the growth experiments was determined by periodical titration of ferrous iron with potassium dichromate (Vogel, 1961).

Cultures containing cells grown in the presence of 8 g/L (control) and 80 g/L of sulfate were first fixed in 4% (v/v) glutaraldehyde (Fluka). Then, one milliliter of culture containing fixed cell samples was filtered through 0.22 μm pore size polycarbonate filtration membranes (Millipore). The cells were dehydrated by immersing the filter membranes for one minute in ethanol at increasing concentrations (20, 40, 60, 80, and 100%). Subsequently, the filters were prepared for microscopic observation by standard procedures including CO₂ critical drying and gold coating. Finally, the prepared filters were observed in a JEOL JSM 6360LV scanning electronic microscope, at acceleration voltages ranging from 10 to 20 kV.

A matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF) equipment (Bruker Daltonics, Germany) was used to analyze small molecular weight compounds released by bacteria subjected to hypo-osmotic shock approach (HSA). Briefly, for hypoosmotic shock approach, cells from cultures grown at different osmotic strength were harvested by centrifugation and cells pellet was exposed to low osmotic strength by resuspending in 5.0 mL of acidified MilliQ-grade water (pH 1.7, adjusted with sulfuric acid) for 10 min at 37°C. Then, bacterial debris were removed by centrifugation and the supernatants were filtered through 0.2 μm pore-sized membranes obtaining the extract. Each Extract was later mixed either with TiO₂ (10 mg/mL) or CsI (30 mg/mL) as matrix, in a 1:1 ratio and spotted onto a target plate MTP384 (Bruker Daltonics, Germany). The spectral peaks obtained from extracts control (8 g/L) and high sulfate culture (80 g/L) were compared to identify differential mass/charge peaks. Each differential peak detected was identified by comparing with those of known common organic osmolytes produced by prokaryotes.

Protein extracts were prepared from L. ferriphilum cells harvested at exponential phase of growth (8 and 80 g/L of sulfate), according to a modified methodology reported by Li et al. (2009). Cells were harvested by centrifugation (1 L culture, at 10,000 g, 20 min), the mass of each cell pellet was estimated after centrifugation using the weight of the empty microcentrifuge tube and the weight of the same tube containing the pellet. The tubes containing the pellets were subjected to sonication in ice bath (6 times, 60% amplitude, constant 30 s) in buffer UTCTEDP (6 M Urea, 2 M Thiourea; 4% CHAPS; 2% Trirón X-100; 40 mM Tris; 1 mM EDTA; 5 mM MgCl2; 60 mM DTT; 1 mM PMSF, Protease inhibitors Roche, cOmplete^™, Mini Protease Inhibitor Cocktail and 100 mg/L DNAse I). Later, the salts were removed using purification centrifugal filters (Amicon Ultra-0.5, Millipore cat. UFC500308), followed by a double precipitation using 5% TCA/Acetone. Protein final concentration was determined by using PierceTM Coomassie plus kit (Thermo Fisher cat. 23236) or 2D Quant kit (Cytiva 80-6483-56).

For the isoelectric focusing (IEF), 17 cm Immobilized pH Gradient (IPG) strips with a non-linear pH range of 5–8 (Bio-Rad) were used. These strips were rehydrated overnight with protein samples previously extracted and quantified using a rehydration buffer [6 M Urea; 2 M thiourea; 4% CHAPS; 0.27% (v/v) ampholytes 3–10; 0.13% (v/v) ampholytes 5–8, and 0.001% (w/v)]. The IEF process involved the following steps: an initial voltage of 500 V for 1 h, followed by an increase to 1,000 V for another hour, with a gradual ramp-up to 8,000 V, maintaining this voltage for a total of 32,000 volt-hours. These procedures were carried out in a Bio-Rad system at a constant temperature of 20°C. Subsequently, the rehydrated strips were equilibrated first with 0.5% Dithiothreitol (DTT), followed by re-equilibration in a buffer containing 4.5% iodoacetamide, which replaced DTT. Each equilibration step took 20 min. The second-dimension electrophoresis was conducted at a constant voltage of 100 V for 18 h.

Delta 2D 4.0 imaging software (Decodon, Greifswald, Germany) was used to perform gel image densitometric analysis. To create a unified master gel, the acquired gel images were merged, and common spots were identified and adjusted across all images. Normalization was then carried out based on the total spot density. Statistical analysis was conducted using the Delta 2D-based student’s t-test to assess the significance and statistical variations in protein expression. A p-value less than 0.05 was considered to indicate statistical significance. Only those spots exhibiting a significant differential level between control and high sulfate concentration were selected for identification purposes. The selected spots were excised of the gel and separately placed in tubes previously washed with Methanol (Merck) or Acetonitrile (Merck) and washed twice for 10 min with 50% Acetonitrile/Water (v/v). The tubes containing gel pieces were stored at −20°C until they were submitted to liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) analyses.

The partial prediction of the sequence of the proteins digested with trypsin was performed by LC-MS/MS with a 3D ion trap spectrometer (Thermo Electron). After the LC-MS/MS analyses, mass spectra were compared with the theoretical mass spectra against a Leptospirillum database using the SEQUEST program.³ Subsequently, the statistical program ProteinProphet⁴ was used for data filtering. This whole protein prediction procedure was performed at the Proteomic Services of the Cancer Center of the University of Maryland, United States, and is described in greater detail by Matthiesen and Bunkenborg (2020). Using the peptide sequences from each protein, a search was carried out with the NCBI BLASTp program (see text footnote 1) to identify the protein (Altschul et al., 1997). The Compute pI/Mw tool⁵ was used to obtain the theoretical isoelectric point and molecular size values.

After an incubation period (between 15–20 days), several colonies of autotrophic iron oxidizers (red-brownish colour and between 0.2–0.5 mm) were obtained by directly streaking PLS onto overlay plates (Figure 1A). Figure 1B shows a second stage where a single colony was transferred an overlay plate. The partial 16S rRNA sequence of the isolate has been identified and assigned the name L. ferriphilum IESL25. This sequence has been deposited in the NCBI GenBank nucleotide database (accession number HQ902070).

The effect of increasing sulfate on culture growth of L. ferriphilum IESL25 is shown in Figure 2. Initially, the isolate was grown in ferrous iron-ABS/TE medium containing 8 g/L of sulfate concentration and 50 mM of ferrous iron, obtaining a culture doubling time of 10.8 h. In the presence of higher sulfate concentration, the culture growth profile was affected producing a longer lag phase with higher doubling times. The doubling times in the presence of 40 g/L, 80 g/L and 100 g/L of sulfate were 11.0 h, 13.5 h, and 14.2 h, respectively. For proteomic analyses, the cells were harvested in the exponential growth phase, at the point when 90% of ferrous iron was oxidized to obtain the highest number of active cells. The harvesting points (90% of oxidized Fe²⁺) occurred at approximately 40, 60, and 75 h for the control condition culture, and in the presence of 80 g/L and 100 g/L of sulfate, respectively.

Based on the observations of growth profiles, it can be deduced that sulfate exerts a suppressive influence on the growth of L. ferriphilum IESL25 as the concentration increases. In the range of sulfate used in the experiments (8–100 g/L), the doubling time values increased following a linear correlation. In light of this information, it is evident that growth inhibition occurs, as indicated by a 24% increase in doubling time between the control condition and the highest sulfate concentration used in this study (100 g/L). Regarding industrial conditions, a previous report using acidophilic culture to recycle water from a low-grade ore heap process indicated between 110 and 130 g/L as the that the upper limits of tolerance by microorganisms to sulfate (Adaos, 2013).

As observed in the profiles, the total cells numbers at the end of the growth experiments decreased, as higher concentrations of sulfate were included in the culture medium (Figure 2). On the other hand, in the control condition (8 g/L), the cells showed average sizes of approximately 1 μm, meanwhile, in cultures growth at higher sulfate concentration (80 g/L), the average cell size was approximately 4 μm. These size differences were observed by contrast phase and scanning electron microscopy (SEM) (Figure 3).

Obtaining cells from the different conditions, at the same exponential growth phase point allows performing proteomic analyses of cells at similar physiological phase, to ensure the reproducibility of the results and validating the comparison of the effect of different concentrations of sulfate (Matthiesen and Bunkenborg, 2020). Previous studies have suggested that cells in the stationary growth phase are structurally, physiologically, and functionally different from those in the exponential growth phase (Bathke et al., 2019). However, it has been shown that genes expressed in response to osmotic shock could also be involved in adaptation to the stationary phase, thus when cells enter the stationary phase, the general metabolism is reorganized and reserve compounds such as polyphosphates accumulate and glycogen, in addition to osmolytes such as trehalose, glycine betaine, and other compatible solutes (Jakowec et al., 2023). The latter justifies the choice for cell collection point, during the exponential phase growth and the importance of not exceeding the time for cell collection in order to avoid false positive results in proteomic analyses. It should also be noted that at the chosen collection time, the ferrous ion is still present in the solution, which ensures that cells have a source of energy and are still growing in such a way that the risk of the expression of related proteins and other types of stress are minimized.

A relevant change observed in L. ferriphilum IESL25 cells from a culture grown in the presence of 80 g/L of sulfate was the increase in the length of the cells by approximately 4-fold, compared to the control culture. Morphological changes seek to maintain the consistency of the cell under extreme conditions, such as increased osmotic pressure, and in this case, environmental stress due to elevated sulfate concentration. Perhaps the most frequent change resulting from environmental stress is filamentation (Nepal and Kumar, 2018). Filamentation of L. ferriphilum in response to sulfate could be a strategy that enables bacteria to survive in challenging environmental conditions by temporarily halting cell division and elongating their shape. This adaptation would allow them to better handle osmotic imbalances until conditions become more favourable for optimal growth and cell division (Nepal and Kumar, 2020).

MALDI-TOF analyses after hypo-osmotic shock approach showed peaks of mass/charge ratios similar to molecular weights of ectoine and hydroxyectoine as indicated in Figure 4. Similar findings were observed using two different ionization matrixes. For TiO₂⁺ CsI matrix (Figure 4A), Hydroxyectoine (H⁺ adducts) has a fold change (treated/control) of 6.17 and Hydroxyectoine-Cs⁺ a fold change of 10.41. For NALDI⁺ CsI matrix (Figure 4B), Ectoine (H⁺) has a fold change of 8.75.

During the bioleaching process of low-grade copper ore and the generation of AMD, microorganisms are subjected to different environmental stressing conditions, such as changes in temperature, pH, concentration of sulfate, toxic metals, and lack of nutrients, among others (Jia et al., 2019). In response to some of these environmental changes, specifically, salt stress, bacteria modify gene expression or the production and accumulation of compatible solutes, also known as osmoprotectants or osmolytes (Gumulya et al., 2018). Here, two osmolytes ectoine and hydroxyectoine were identified. Because the matrixes used to carry out this were TiO₂⁺ CsI and NALDI⁺ CsI, it must be emphasized that the detection of peaks and their intensities depend on the ionizability of the compounds in the matrix used. Regarding the spectra obtained by using each matrix, it was observed that adduct peaks associated with the molecular weight of hydroxyectoine were detected by using TiO₂ as a matrix, in addition to perceiving fewer background signals (blank) than with the NALDI matrix. Despite the latter, NALDI matrix allowed the detection of several entities with intense signals, including a signal associated with the molecular weight of the osmolyte ectoine. Ectoine and hydroxyectoine act as compatible solutes by stabilizing cellular structures, enzymes, and macromolecules, preventing their denaturation, and ensuring cell survival in extreme environments (Corbett et al., 2022). Growing L. ferriphilum IESL25 requires a minimal number of total salts for its development. In the basic salt-based media used, no organic substances or osmolyte precursors were supplemented. Therefore, the identification of these osmolytes suggests the production of these compounds from scratch in response to environmental challenges such as high sulfate concentration. There have been only a limited number of studies conducted on microorganisms with acid and salt tolerance and most of it is related to the NaCl stress (Gumulya et al., 2018). Ectoine accumulation has been shown to increase halotolerance in Leptospirillum and Acidihalobacter species (Khaleque et al., 2019). Comparative genomic analyses have suggested trehalose as another potential compatible solute produced by Leptospirillum spp. (Rivera-Araya et al., 2020) and it has been previously detected in cultures of Leptospirillum ferriphilum grown in the presence of elevated concentrations of sulfate (Galleguillos et al., 2018).

Two-dimension protein electrophoresis gels (2-DE) in the pH range 5-8NL of protein extracts obtained from cultures grown under the control condition of 8 g/L and 80 g/L of sulfate are represented in Figure 5 (full gel images in Supplementary Figure S1). These protein electrophoresis assays were performed by triplicate and the expression analyses were subsequently carried out in accordance with point 2.5 of Materials and Methods section. Spots 1 to 4 represent the proteins differentially expressed and identified by LC-MS/MS. The mean normalized volume of spot 1 was 20.65 ± 0.18; spot 2, 9.59 ± 0.18; spot 3, 5.20 ± 0.15, and spot 4, 0.03 ± 0.02 (Supplementary Table S1).

The protein prediction results were based on the analyses of peptides obtained for each excised gel spot (protein). Table 1 shows the BLASTp analyses for each peptide obtained from gel spots sorted from highest to lowest score. In all cases, the peptides corresponded to the same protein in different species of the genus Leptospirillum.

Proteomic investigations of Acidihalobacter aeolianus DSM 14174^T, indicated a remarkable 422-fold increase in ectoine synthase expression under conditions of elevated salt stress (Khaleque et al., 2018). Rivera-Araya et al. (2019) concluded that chloride stress up-regulated genes for the synthesis of potassium transporters (kdpC and kdpD), and biosynthesis of the compatible solutes (hydroxy)ectoine (ectC and ectD) and trehalose (otsB). On the other hand, Acidihalobacter prosperus DSM 5130, A. prosperus strain F5, A. prosperus strain V6, and A. ferrooxidans strain V8 represent a group of halotolerant, iron- and sulfur-oxidizing acidophiles whose genomes have been sequenced. The analysis of their genomes revealed the presence of genes associated with the production of osmoprotectants like ectoine and proline, as well as genes responsible for synthesizing periplasmic glucans and osmolyte transporters (Ossandon et al., 2014; Khaleque et al., 2017a,b,c).

This background is closely related to the differentially identified proteins represented in Figure 5. The first protein of interest corresponds to Aspartate-semialdehyde dehydrogenase (ASD). ASD is an enzyme that plays a crucial role in the biosynthesis of certain amino acids and compatible solutes in microorganisms. It is highlighted that one of the key functions of ASD involves the production of the compatible solute ectoine (Becker and Wittmann, 2020). ASD catalyses the conversion of L-aspartate semialdehyde to L-aspartate-4-semialdehyde, which is further processed by other enzymes in the ectoine biosynthesis pathway. These enzymes convert L-aspartate-4-semialdehyde into ectoine and its precursor, N-γ-acetyl-L-2,4-diaminobutyric acid (Becker and Wittmann, 2020). Other enzymes that increased their expression correspond to Isocitrate dehydrogenase (IDH) and Succinyl-CoA synthetase (SCoA). The relationship between IDH, SCoA, and ectoine production lies in the fact that the TCA cycle, in which both enzymes produce metabolites that can be utilized as precursors or intermediates in the biosynthesis of ectoine. During cellular stress TCA cycle plays an important role by modulating NADH/NADPH homeostasis, scavenging reactive oxygen species, producing ATP by substrate-level phosphorylation, signalling, and supplying metabolites to cope with a wide range of cellular disruptions (MacLean et al., 2023). For instance, alpha-ketoglutarate, which is produced by the action of IDH, can serve as a precursor for ectoine biosynthesis. On the other hand, the energy generated in the TCA cycle by SCoA may also play a role in providing the necessary ATP or GTP for ectoine biosynthesis. Thus, the TCA cycle and its associated enzymes could indirectly induce the production of ectoine in L. ferriphilum (MacLean et al., 2023). Finally, according to expression analysis, Hsp20 protein was inhibited under osmotic stress by magnesium sulfate. A wide range of microorganisms, including extremophiles, have at least one homolog of Hsp. Izquierdo-Fiallo et al. (2023) analysed Hsp20-encoding genes are widely distributed and highly redundant in acidophiles, however, their low level of expression suggests that they are proteins not specialized to act under conditions of cellular stress. Although functions that help the correct folding of proteins are attributed, this does not mean that they are proteins that are always expressed in stress situations.