Biodiversity and Biotechnological Potential of Dunaliella sp. Isolates from Kalloni Solar Saltworks (Lesvos, Greece)

Abstract

Hypersaline solar saltworks represent unique ecological niches that harbor extremophilic microalgae with considerable biotechnological potential. Within these environments, members of the genus Dunaliella are particularly noteworthy due to their remarkable metabolic plasticity and ability to accumulate high-value biomolecules. In the present study, we investigated the biodiversity of Dunaliella in hypersaline saltworks by isolating and identifying autochthonous strains and assessing their growth kinetics and biomass biochemical composition in the context of potential biotechnological applications. Specifically, sixteen strains of Dunaliella were isolated from evaporation and crystallizer ponds of the Kalloni saltworks in Lesvos, Greece, and subjected to an integrative characterization combining morphological observations, molecular phylogenetics, growth kinetics, and biochemical profiling. Phylogenetic analyses based on four genetic markers (18S, ITS, rbcL, tufA) consistently resolved the isolates into three distinct clades: one corresponding to Dunaliella salina/D. minutissima, one to D. parva, and a third representing a clearly divergent lineage. Growth assays revealed marked variability in cell density, biomass productivity and specific growth rate, with certain strains exhibiting enhanced proliferation under controlled conditions. Biochemical analyses demonstrated distinct allocation patterns, with evaporation pond isolates comparatively enriched in proteins (up to 60.8% DW), whereas crystallizer pond isolates accumulated higher levels of carbohydrates (up to 19.0% DW), carotenoids (up to 7.34% mg g⁻¹ DW) and phenolic compounds (up to 8.68% mg GAE g⁻¹ DW). Antioxidant assays (FRAP, TEAC) further indicated significantly elevated reducing and radical scavenging activities among crystallizer isolates. These findings expand current knowledge on the biodiversity of autochthonous Dunaliella strains and support their potential as sustainable sources of bioactive compounds for applications in the agri-food, nutraceutical, pharmaceutical, and cosmeutical sectors.

1. Introduction

The escalating environmental pressures associated with climate change and global population growth necessitate the development of sustainable and resilient solutions [1]. Sectors such as food production, healthcare, and energy are increasingly turning to innovative biotechnological strategies that support resource efficiency while minimizing ecological degradation. Within this context, blue biotechnology has emerged as a critical scientific field, leveraging marine biodiversity to advance environmentally sustainable and economically viable technologies [2,3]. Microalgae, in particular, play a fundamental role in maintaining planetary health, contributing substantially to global oxygen production through photosynthesis serving as the base of the aquatic food chain [4,5,6]. Beyond their primary role as oxygenic photoautotrophs, microalgae exhibit exceptional ecological and metabolic plasticity, which has driven the evolution of complex biosynthetic pathways and the production of diverse bioactive compounds [7,8]. Their ability to colonize a wide range of aquatic and extreme environments, including hypersaline lakes, acidic springs, and polar habitats, highlights their adaptive potential [9,10]. Furthermore, microalgae represent a largely unexploited reservoir of photosynthetic organisms with considerable biotechnological promise. They are capable of synthesizing a wide spectrum of high-value metabolites, such as proteins, lipids, polysaccharides, and pigments, with applications across the agri-food, nutraceutical, pharmaceutical, cosmetic, and bioenergy sectors [11,12].

Solar saltworks, also known as salinas or salterns, are artificial systems composed of a series of shallow, interconnected ponds where seawater is gradually concentrated through solar and wind evaporation, ultimately leading to the precipitation and harvesting of high-purity sodium chloride [13]. In addition to their industrial role, solar saltworks function as complex coastal aquatic ecosystems, characterized by strong salinity gradients (ranging from marine levels up to 300 ppt), habitat heterogeneity, and ecological richness [14]. Their shallow ponds, nutrient dynamics, and microbial diversity support unique biological communities, including halophilic microalgae, benthic microorganisms, and bird populations, that contribute to both salt production and broader ecosystem functioning [15]. Kalloni saltworks is a modern mechanized facility that covers a surface area of 2.63 km² and has a potential production capacity of 40,000 metric tons of salt [16]. It consists of shallow (up to 0.5 m deep) interconnected ponds with leveled bottoms, separated by dykes and arranged in a sequential manner. Approximately 90% of the total surface area of the saltworks is covered by evaporation ponds (saltpans). Salinity progressively increases along this sequence due to solar radiation and wind-driven evaporation, reaching values above 300 ppt in the terminal ponds (crystallizers), where salt precipitation takes place [17]. The saltworks are part of the Kalloni Gulf coastal system, a semi-enclosed embayment that covers an area of approximately 110 km² [18]. The gulf is ecologically significant, supporting diverse habitats and serving as an important site for avifauna conservation. It is a proposed “Site of Community Interest” within the Natura 2000 network and plays a crucial role in preserving nesting, wintering, and resting grounds for various bird species [19]. Sampling from extreme habitats such as solar saltworks offers a valuable opportunity for isolating resilient microalgal strains. The pronounced salinity gradients, intense solar radiation, and substantial diurnal fluctuations characteristic of these environments impose strong selective pressures favoring microalgae with exceptional physiological plasticity and stress tolerance.

The green microalga Dunaliella is a halophilic genus commonly found in hypersaline environments, including salt lakes, lagoons, and solar saltworks, where it often serves as the dominant or sole oxygenic phototroph. Notably, species such as D. salina and D. viridis thrive in conditions of extreme salinity, tolerating environments with sodium chloride concentrations approaching saturation [20,21,22]. A key physiological feature that underlies this remarkable halotolerance is the absence of a rigid cell wall, which allows Dunaliella cells to osmotically adjust their volume in response to rapid changes in external salinity [23]. To maintain cellular homeostasis under such hyperosmotic stress, Dunaliella synthesizes high intracellular concentrations of glycerol as a compatible solute and accumulates large amounts of β-carotene, a potent antioxidant pigment that protects the photosynthetic apparatus against photooxidative damage under intense light and saline conditions [24,25,26]. These adaptive mechanisms not only confer ecological competitiveness but also have significant biotechnological relevance, as glycerol and carotenoids are commercially valuable metabolites. Furthermore, the extreme conditions required for Dunaliella cultivation inherently minimize the risk of contamination by other microorganisms, facilitating more controlled and stable biomass production. Owing to these properties, Dunaliella has attracted extensive scientific attention as a model for stress physiology and as a candidate for large-scale biotechnological exploitation [27]. Numerous studies have examined its biodiversity, biochemical potential, and metabolic plasticity across various regions, including Greece, where Dunaliella populations have been reported in multiple saltworks [28,29,30,31]. Despite the documented presence of Dunaliella spp. in Greek solar saltworks, existing studies have primarily focused on occurrence and limited strain isolation, while comprehensive molecular characterization, comparative phylogenetic resolution, and systematic evaluation of growth performance, biochemical traits, and biotechnological potential of autochthonous Greek isolates remain scarce. This limited knowledge constrains our understanding of their genetic diversity, physiological adaptations, and potential for biotechnological exploitation.

In this study, sixteen Dunaliella isolates were identified from the Kalloni solar saltworks in northeastern Greece. To comprehensively evaluate their diversity and biotechnological potential, a polyphasic approach was employed, integrating morphological characterization, molecular phylogenetics, growth kinetics, and biochemical profiling. This multi-dimensional assessment aimed to uncover phenotypic and physiological variability among the isolates.

2. Materials and Methods

2.1. Sampling and Study Area

The microalgal strains examined in this study were isolated from water samples obtained from the Kalloni solar saltworks (39°13′ N 26°14′ E) located on Lesvos Island, NE. Aegean Sea, Greece (Figure 1).

The study area, where the water samples were obtained, included two different ponds within the saltworks, each characterized by distinct salinity levels. Samplings were carried out during October 2021 at 11:00 AM using sampling bottles. The first sample was collected from the evaporation ponds (39°12′46″ N, 26°14′39″ E) with a salinity of 23°Bé (Specific Gravity 1.1885), pH 7.8, and a temperature of 29.7 °C, while the second was taken from the crystallizer ponds (39°12′48″ N, 26°14′46″ E) with a salinity of 40°Bé (Specific Gravity 1.3809), pH 7.6, and a temperature of 30.6 °C. In each case, the sample represented a pool of three spatially independent samplings. The water was filtered through a 100 μm mesh to remove meso- and macro-phytoplankton, as well as zooplankton, ensuring that only nanoplankton (<20 μm) and microplankton (<100 μm) remained in the samples. The samples were then transferred to the lab, where they were stored under controlled conditions (temperature: 20 °C, light intensity: 100 µmol photons·m⁻²·s⁻¹, light period: 12 h light–12 h dark) and prepared for further analysis.

2.2. Strain Isolation and Cultivation

The samples were designated based on the pond of origin, resulting in two processing groups. Each group comprised two samples: one that underwent exclusive filtration and another that was filtered and then enriched with Guillard f/2 medium (Supplementary Table S2) [32]. The four samples were cultivated in 1 L flasks and incubated in the laboratory under illumination to promote growth and proliferation. The resulting cultures encompassed a range of distinct microalgal taxonomic assemblages. The strains were isolated using a combination of two different methods: (a) serial dilution and (b) agar plating [33]. For the isolation stages, autoclaved natural seawater enriched with f/2 medium was used. Sixteen Dunaliella sp. strains were analyzed, all maintained as uni-algal cultures, with the first eight originating from the evaporation ponds and the other eight from the crystallizer ponds. The strains were designated as SKE01–SKE08 and SKC09–SKC16, where “S” denotes saltworks, “K” refers to Kalloni, “E” and “C” correspond to evaporation and crystallizer ponds, respectively, and the trailing number indicates the specific strain. The single-strain, non-axenic cultures were preserved in sterile saltwater with 40‰ salinity, enriched with f/2 medium, in the Ecology and Systems Dynamics laboratory at the University of the Aegean in a culture chamber with steady conditions (temperature: 20 °C, light intensity: 100 µmol photons·m⁻²·s⁻¹, light period: 12 h light–12 h dark) and were sub-cultured on a monthly basis under sterile conditions. An aliquot of fresh subculture of each strain at the early stationary phase was examined under light microscopy. Micrographs were captured using an OLYMPUS BX50 microscope equipped with an OLYMPUS DP71 camera. Prior to experimental analyses, all strains were subjected to multiple successive acclimation and cultivation cycles under the standardized laboratory conditions.

2.3. Growth Kinetics

The sixteen strains were inoculated in 2 L of rigorously aerated autoclaved natural seawater (40‰ salinity) enriched with f/2 medium. Cultivation was carried out in a temperature- and light- controlled culture chamber under artificial illumination from cool daylight fluorescent lamps (6500 K) at an irradiance of 100 μmol photons·m⁻²·s⁻¹, with a photoperiod of 12:12 h light: dark and a constant temperature of 23 °C. Cell density was monitored daily using a Motic AE31 inverted light microscope (Motic Instruments Inc., Xiamen, China) and a Palmer–Maloney Nannoplankton counting chamber (PhycoTech, Inc., St. Joseph, MI, USA) [34] to generate growth curves for each strain. Dry weight (DW) biomass was determined by filtering a known culture volume of predetermined cell density onto pre-weighed GF/C filters (Whatman, Maidstone, UK) under vacuum. Filters were rinsed with isotonic ammonium formate to remove residual salts and dried overnight at 60 °C in a Thermo Electron Heraeus T12 oven (Thermo Fisher Scientific, Waltham, MA, USA). The final DW was obtained gravimetrically using a KERN ALJ 220-4M analytical balance (KERN & SOHN GmbH, Balingen, Germany). Microalgae biomass was harvested in the stationary phase by centrifugation at 3000 rpm for 20 min at 4 °C using a Thermo Scientific Heraeus Megafuge 16R centrifuge (Thermo Fisher Scientific, Waltham, MA, USA). The collected biomass was washed twice and lyophilized (VaCo2, Zirbus Technology GmbH, Bad Grund, Germany).

Growth parameters were calculated for each strain, including maximum specific growth rate (µ_max), maximum cell density (N_max), maximum cell yield (Y_max), maximum dry weight biomass productivity (P_b,max), maximum cell productivity (P_c,max), and maximum dry weight biomass (X_max). The N_max and X_max were determined by identifying the highest measured values of cell density and dry biomass, respectively, across all replicates for each strain, and calculating their mean values. The µ_max was calculated as:

$${\mathrm{\mu }}_{\mathrm{m}\mathrm{a}\mathrm{x}}={\displaystyle \frac{\ln (N_t/N_0)}{t-t_0}}$$

(1)

where N₀ is the initial cell density at time t₀ and N_t is the cell density at time t [35].

Additionally, P_b,max was calculated according to the equation:

$${\mathrm{P}\mathrm{b},}_{\mathrm{m}\mathrm{a}\mathrm{x}}={\displaystyle \frac{X_t-X_0}{t- t_0}}$$

(2)

where X₀ and X_t are the initial and highest biomass concentration at time t₀ and t, respectively. Moreover, P_c,max was determined as:

$${\mathrm{P}\mathrm{c},}_{\mathrm{m}\mathrm{a}\mathrm{x}}={\displaystyle \frac{N_t-N_0}{t- t_0}}$$

(3)

where N₀ and N_t are the initial and highest cell densities at time t₀ and t, respectively [36]. Finally, Y_max was determined as the difference between the initial and maximum cell densities [37].

2.4. Taxonomic Classification

To comprehensively characterize the isolated strains, both classical morphological approaches and molecular taxonomy techniques were utilized. Light microscopy was used to assess basic morphological characteristics, including cell length, volume and shape, as well as observations of the cell wall, chloroplast number, and pigmentation, to obtain an initial indication of the taxonomic classification of the strains. Morphological observations were conducted using a Zeiss AX10 optical microscope (Carl Zeiss, Oberkochen, Germany). Genomic DNA was extracted from all isolates with the Genomic DNA from Soil kit (Macherey-Nagel, Düren, Germany), according to the manufacturer’s instructions. For molecular taxonomic classification, four genetic markers were targeted: the nuclear 18S rRNA gene, the internal transcribed spacer (ITS) region, and two chloroplast markers, rbcL (ribulose-1,5-bisphosphate carboxylase/oxygenase) and tufA (elongation factor Tu) (Supplementary Table S1). These regions were amplified using polymerase chain reaction (PCR), using the reagents and cycling conditions described by Ntzouvaras et al. (2023) [38]. Amplicons were purified and sequenced bidirectionally according to the same protocol.

The nucleotide sequences obtained in this study were processed using DNA Baser v5.08 software (Heracle BioSoft SRL, Lilienthal, Germany). Sequence similarity was assessed through BLAST (version 2.17.0) searches against the GenBank database and reference sequences closely related to those generated in this study were retrieved and included in subsequent phylogenetic analyses. Sequence alignment was carried out using the Clustal W algorithm implemented in MEGA X version 10.2.2 [39]. Phylogenetic trees were constructed separately for each DNA marker using both the Neighbor-Joining (NJ) method, based on Kimura’s two-parameter model, and the Maximum Likelihood (ML) method, employing the best-fit nucleotide substitution model for each marker: the Jukes-Cantor model [40] for the 18S rRNA gene, the Kimura-2 model [41] for the ITS region, and the Tamura-3 model [42] for the rbcL and tufA genes. Bootstrap support values were estimated based on 1000 replicates. All nucleotide sequences generated for the selected markers were submitted to the NCBI database with the accession numbers PX114591 to PX114606, PX118316 to PX118331, PX318481 to PX318508, PX352695, PX352696, PX422755 and PX422756.

2.5. Analysis of Biomass Composition

To achieve a comprehensive biochemical profile of the isolated microalgae biomass, established protocols were employed to quantify both primary and secondary metabolites. For primary metabolite analysis, the total nitrogen (Total-N) content was determined using the Kjeldahl method (Foss Tecator Kjeltec 8200 Distillation Unit) according to Sáez-Plaza et al. 2013 [43]. Crude protein content was finally estimated by multiplying the total nitrogen value by the conventional nitrogen-to-protein conversion factor of 6.25 [44,45]. The total lipid content was quantified using the sulfuric acid–phosphovanillin method, as described by Farinacci and Laurent (2023) [46] using a standard calibration curve prepared with cholesterol (10–100 µg). In addition, total polysaccharide content was determined using a phenol–sulfuric acid colorimetric method [47] while polysaccharide concentrations were calculated using glucose for the standard curve (0.05–1.00 mg/mL).

Secondary metabolites were analyzed using methanolic extracts prepared from freeze-dried biomass [48]. Briefly, 25 mg of each microalgal isolate was suspended in 0.5 mL of pure methanol in a water bath sonicator (40 kHz, Witeg Labortechnik GmhB, Wertheim, Germany) for 20 min at 90% amplitude in darkness. The lysate was centrifuged at 13,000 rpm for 10 min at 4 °C, and the supernatant was collected. The extraction was repeated twice, and the pooled supernatants were used for analysis. Throughout extraction, the temperature was maintained below 10 °C to minimize sample degradation. The same procedure was employed for pigment extraction, with the exception that 80% (v/v) acetone was used as the extraction solvent. Methanolic extracts were used for the determination of antioxidant capacity and total phenolic content, while acetonic extracts were employed for pigment quantification.

Total antioxidant capacity of microalgae isolates was assessed through both Ferric Reducing Antioxidant Power (FRAP) assay following the method described by Johnson et al. (2023) [49] and ABTS radical scavenging activity assay as described by Safafar et al. (2015) [50]. Antioxidant capacity from both assays was determined using a Trolox standard curve and expressed as Trolox equivalents (TE) milligrams per gram of dry biomass (mg TE/g DW). The total phenolic content (TPC) of the microalgal isolates was quantified using the Folin–Ciocalteu colorimetric method, based on the protocol described by Wang et al. (2023) [51] and expressed as gallic acid equivalents (GAE) per gram of dry biomass (mg GAE/g DW). Pigment concentrations (Chl a, Chl b, carotenoids) were calculated using correction equations established by Lichtenthaler and Buschmann (2001) [52]. Chlorophylls and total carotenoids were expressed as milligrams of pigment per gram of dry biomass (mg/g DW).

2.6. Statistical Analysis

The data were expressed as the mean ± standard error (SE). Statistical analysis was performed using the Statistica 12 software package (StatSoft Inc., Tulsa, OK, USA). One-way analysis of variance (ANOVA) was conducted to determine significant differences among the sixteen Dunaliella strains. When the ANOVA result was significant (p < 0.05), Tukey’s Honest Significant Difference (HSD) post hoc test was applied to perform pairwise comparisons. Effect sizes were additionally quantified using partial eta squared (ηp²) to assess the magnitude of pond- and lineage-related effects. Statistical significance was considered at p < 0.05. Principal Component Analysis (PCA) was performed in RStudio (Version 4.2.1) and the ggplot2, ggrepel, and ggforce packages for visualization. Principal component analysis (PCA) was performed using all quantified biochemical variables, including carbohydrates, total lipids, Chl a, Chl b, carotenoids, total phenolics, TEAC and FRAP assays. All variables were standardized using z-score transformation and PCA was conducted using unrotated, orthogonal components. Figures were created with SigmaPlot 12.0 (Systat Software Inc., San Jose, CA, USA).

3. Results

3.1. Taxonomic Classification of Microalgal Isolates

Sixteen microalgae strains were isolated from the Kalloni solar saltworks located on Lesvos Island, Greece. Light microscopy analysis of the isolated strains revealed many differences but also basic similarities between these strains. Cells displayed notable morphological variability within and between cultures, with cell length ranging from 7.7 µm to 18.2 µm, while width varied from 5.2 µm to 16.7 µm. Also, SKC strains in general exhibited increased cell volume compared to SKE strains, a pattern consistent with the nearly two-fold higher salinity of the crystallizer pond. However, this response appeared to be also strain-dependent and influenced by underlying genetic and environmental factors. The general cell shape ranged from ovoid to broadly ellipsoidal, and all observed cells possessed two anteriorly inserted flagella of approximately equal length. The absence of a rigid cell wall and the presence of a single, cup-shaped chloroplast were also noted, while for some strains, cells with distinct orange-red pigmentation were observed, strongly indicative of intracellular β-carotene accumulation—a characteristic commonly associated with Dunaliella species under stress conditions such as high salinity or intense light exposure. Collectively, these morphological and pigmentation traits are consistent with diagnostic features of the genus Dunaliella and support the preliminary classification of the isolates within this group (Figure 2). To clarify the evolutionary relationships among the isolated microalgal strains, a DNA barcoding strategy was employed using four commonly utilized molecular markers: the nuclear 18S rRNA gene, the internal transcribed spacer (ITS) region, and the plastid genes rbcL and tufA [53,54,55]. Partial sequences for each marker were obtained from the isolates, aligned, and trimmed to the same length. Phylogenetic trees were constructed separately for each marker using both the Neighbor-Joining (NJ) and Maximum Likelihood (ML) approaches to assess evolutionary relationships. The resulting tree topologies were generally consistent across all markers and methods, with only slight discrepancies observed (Supplementary Figures S1–S4). Due to their higher bootstrap support values, the ML trees were chosen for further analysis.

The nuclear markers 18S rRNA gene and ITS region placed all sixteen isolated strains within the Dunaliella sp. group. Due to the difficulties in obtaining clean and complete 18S rRNA gene sequences for all isolates, phylogenetic analysis was performed using a shorter fragment (551 bp). The resulting phylogeny revealed that all microalgal isolates, regardless of their sampling origin, clustered closely together with previously characterized Dunaliella species, including D. minutissima TAU-MAC 1220, D. pseudosalina MAH, D. primolecta CCAP 11/34, and D. salina (Supplementary Figure S1). Pairwise comparisons of 18S rRNA gene sequences showed high similarity (99.82–100%) among most of the isolated strains and 98.91–100% with the reference Dunaliella species in the phylogeny. Within the group, isolate SKC15 exhibited the greatest divergence, sharing 98.55 to 98.73% similarity with the other isolates and 97.82% to 98.73% with the Dunaliella reference strains. The ITS region (563 bp) provided greater phylogenetic resolution, grouping the isolates into three distinct and well-supported lineages (Supplementary Figure S2). In the first group (I) strains SKE01, SKE02, SKC09, and SKC11 were 100% identical and closely related to SKC10, which shared 99.64% identity with them. SKC10 was also 100% identical to the reference strain Dunaliella sp. ABRIINW M1/1 and showed 98.73–99.82% similarity with other related reference strains within the group. Isolates SKE03 and SKE05, both isolated from the same site, formed a separate lineage (II) displaying complete identity with each other and high similarity to Dunaliella sp. MBTD-CMFRI-S121 (98.91%) and MBTD-CMFRI-S086 (98.73%). They were moderately related to the first group, sharing 97.27–97.45% sequence similarity. The third cluster (III) consisted of eight identical strains, collected from both sampling sites, which exhibited high similarity (99.68–99.82%) to reference strains including D. salina RR102, Dunaliella sp. ABRIINW U1/1, and D. minutissima TAU-MAC 1220. Although slightly divergent, SKC13 also clustered within this group, sharing 99.82% identity. Notably, this group shared only 92.77–93.66% similarity with the other two, indicating clear genetic divergence among the lineages.

The plastid markers rbcL and tufA provided complementary phylogenetic resolution, delineating the isolates into three well-supported clades. In the rbcL phylogeny, which was congruent with the ITS topology (Supplementary Figure S3), isolates SKE06, SKE07, SKC12, SKC13, SKC14, SKC15, and SKC16 shared 100% pairwise sequence identity and were identical to well-characterized strains such as D. salina OUC21 and D. bardawil UTEX 2538. Isolates SKE04 and SKE08, sharing a 99.69% similarity, were also clustered in this clade, showing 99.84% similarity to the rest group members. A second set of identical isolates (SKE01, SKE02, SKC09, SKC10 and SKC11) formed a distinct lineage, together with the also identical D. parva TAU-MAC 0120. The first two clusters in the rbcL phylogeny were moderately related, showing 97.97–98.75% similarity. Clade III included SKE03 and SKE05, identical to each other but distantly related to all other strains in the phylogenetic tree, with pairwise similarities ranging from 49.0% to 51.0%. The tufA gene marker resolved the isolates into three lineages with varying degrees of relatedness (Supplementary Figure S4). Isolates SKE04, SKE06, SKC12, SKC13, SKC14, SKC15, SKC16 and reference strains such as D. salina PLY DF41 were identical, while SKE07 showed 99.82% similarity to the group. The second group included six isolates (SKE01, SKE02, SKC10, SKC11, SKE08, and SKC09), sharing 99.47–100% similarity. Members of the first two groups in the tufa phylogeny were closely related, showing 99.29–99.82% sequence identity. Identical isolates SKE03 and SKE05 formed a third, distinct group. Their closest relative was SKE08 (99.82% similarity), and they shared 99.29–99.65% identity with other clade II members. The relationship between groups I and III was comparatively more distant, with pairwise sequence similarities ranging from 99.29% to 99.47%. Overall, the tufA phylogeny highlights high sequence conservation among most isolates, with SKE03 and SKE05 representing a clearly separate lineage. To obtain more in-depth taxonomic resolution and better delineate relationships within the phylogeny, we decided to combine two markers: rbcL + ITS and rbcL + tufA, using the reference strains available in the database.

By combining the rbcL and ITS gene markers into a concatenated phylogenetic tree, three distinct clades were revealed (Figure 3). One lineage included isolates SKE04, SKE06, SKE07, SKE08, SKC12, SKC14, SKC15, and SKC16 which were 100% identical to each other and to the reference strain D. minutissima TAU-MAC 1220. Isolate SKC13 was closely related showing 99.82% similarity to the other members of this clade. A second group comprised SKE01, SKE02, SKC09, and SKC11, which were identical to each other and shared 99.82% identity with D. parva TAU-MAC 0120. Isolate SKC10 was also placed within this cluster, showing 99.64% similarity to its group members and 99.45% to the reference strain. This group was divergent from the first lineage, sharing only 92.75–92.93% identity. The third lineage consisted of SKE03 and SKE05, which were identical to each other but more distantly related to the other isolates. They showed 97.27–97.45% similarity to the second group and 93.48–93.66% similarity to the first, highlighting their distinct position within the phylogeny.

When combining the tufA and rbcL gene markers, the isolates were again resolved into three main lineages (Figure 4), confirming the pattern observed in the previous analyses. SKE04, SKE06, SKC12–SKC16 formed a highly conserved group, sharing 100% sequence identity with each other and with D. salina. SKE07 clustered closely with this group (99.82% similarity), while SKE08 was slightly more divergent, showing 99.47–99.65% identity with the other members. Another set of isolates, SKE01, SKE02, SKC10, and SKC11, displayed complete identity among themselves and 99.65–99.82% similarity to the first lineage and the reference D. salina strain. SKC09 also grouped with these isolates, showing 99.65% similarity to its cluster members and 99.24–99.47% similarity with the first lineage.

As shown in the other phylogenies, isolates SKE03 and SKE05 formed a distinct and more distant lineage. These isolates were identical to each other and shared 99.29–99.47% similarity with the first lineage and 99.29–99.65% with the second group, with SKE08 identified as their closest relative (99.82%).

Overall, the phylogenetic analyses consistently resolved the sixteen isolates into three main lineages across all four markers, with only minor variations in placement. Nuclear markers (18S rRNA gene and ITS) placed all isolates within the Dunaliella genus, with 18S rRNA gene showing high overall similarity (98.91–100%) and ITS providing greater resolution. The majority of isolates clustered closely with D. salina and D. minutissima, while five isolates (SKE01, SKE02, SKC09, SKC10, SKC11) were closely related to D. parva. SKE03 and SKE05 consistently formed a distinct lineage, distant from the other isolates. Plastid markers (rbcL and tufA) provided complementary support, showing high sequence conservation within the main groups and confirming the three-lineage pattern observed with nuclear markers. Most isolates shared 100% identity within their respective clusters, with minor divergence in a few strains. Moreover, the combined analyses of concatenated markers (rbcL + ITS and rbcL + tufA) confirmed the general relationships and reinforced the grouping into three lineages. Overall, the isolated strains consistently grouped into: (I) a clade closely related to D. salina and D. minutissima (lineage A), (II) a D. parva–related cluster (SKE01, SKE02, SKC09–SKC11) (lineage B), and (III) a clearly divergent lineage composed of SKE03 and SKE05 (lineage C). These results confirm the general relationships among the isolates across multiple markers and reliably delineated the isolates into distinct clusters (Supplementary Table S3). However, additional molecular markers will be required to achieve finer phylogenetic resolution and more precise taxonomic assignment.

3.2. Culture Kinetics

Growth kinetics of the sixteen Dunaliella isolates cultivated under identical conditions are presented in

Table 1. Maximum specific growth rate (µ_max), maximum cell density (N_max), maximum cell yield (Y_max), maximum dry weight biomass productivity (P_b,max), maximum cell productivity (P_c,max), and maximum dry weight biomass (X_max) of sixteen Dunaliella isolates cultivated under identical conditions. Values are expressed as mean ± standard error (SE), n = 3. Group means (bottom rows) represent average values for isolates by pond origin and lineage. Different superscript letters within each column indicate statistically significant differences among isolates or groups (Tukey’s HSD, p < 0.05). Significance levels for group comparisons: * p < 0.05, ** p < 0.01, *** p < 0.001. Partial eta squared (ηp²) values are reported for pond- and lineage-level comparisons.

Isolate	μmax (day−1)	Nmax (106 × cells mL−1)	Ymax (106 × cells mL−1)	Pb,max (g L−1 day −1)	Pc,max (106 × cells mL−1 day −1)	Xmax (g L−1)
SKE01	0.633 ± 0.019 abc	3.11 ± 0.26 cd	2.87 ± 0.26 cd	0.029 ± 0.007 abcd	0.288 ± 0.003 b	0.24 ± 0.01 de
SKE02	0.568 ± 0.028 abcd	4.47 ± 0.26 b	4.24 ± 0.26 b	0.021 ± 0.005 abcd	0.413 ± 0.037 a	0.22 ± 0.01 e
SKE03	0.567 ± 0.019 abcd	3.07 ± 0.1 cd	2.83 ± 0.1 cd	0.026 ± 0.002 abcd	0.258 ± 0.008 bc	0.27 ± 0.01 de
SKE04	0.564 ± 0.042 bcd	3.75 ± 0.08 bc	3.52 ± 0.08 bc	0.024 ± 0.005 abcd	0.304 ± 0.023 b	0.28 ± 0.01 de
SKE05	0.661 ± 0.026 a	3.62 ± 0.21 bc	3.41 ± 0.21 bc	0.038 ± 0.003 a	0.276 ± 0.010 bc	0.43 ± 0.03 c
SKE06	0.247 ± 0.008 e	2.01 ± 0.11 ef	1.8 ± 0.11 ef	0.013 ± 0.000 d	0.090 ± 0.006 d	0.43 ± 0.00 c
SKE07	0.488 ± 0.001 d	5.62 ± 0.41 a	5.42 ± 0.41 a	0.016 ± 0.001 cd	0.275 ± 0.017 bc	0.45 ± 0.02 c
SKE08	0.543 ± 0.005 cd	2.20 ± 0.13 de	2 ± 0.13 de	0.019 ± 0.002 bcd	0.201 ± 0.013 c	0.20 ± 0.01 e
SKC09	0.601 ± 0.021 abc	3.24 ± 0.20 c	3.02 ± 0.2 c	0.036 ± 0.009 ab	0.293 ± 0.016 b	0.27 ± 0.02 de
SKC10	0.606 ± 0.006 abc	3.13 ± 0.29 cd	2.91 ± 0.29 cd	0.025 ± 0.004 abcd	0.323 ± 0.016 b	0.19 ± 0.01 e
SKC11	0.641 ± 0.018 ab	5.68 ± 0.20 a	5.45 ± 0.2 a	0.029 ± 0.000 abcd	0.297 ± 0.008 b	0.32 ± 0.02 d
SKC12	0.227 ± 0.011 e	2.14 ± 0.13 def	1.93 ± 0.13 def	0.023 ± 0.001 abcd	0.098 ± 0.008 d	0.65 ± 0.02 b
SKC13	0.257 ± 0.011 e	2.14 ± 0.08 def	1.94 ± 0.08 def	0.031 ± 0.002 abcd	0.100 ± 0.005 d	0.81 ± 0.04 a
SKC14	0.235 ± 0.003 e	1.18 ± 0.03 f	0.98 ± 0.03 f	0.033 ± 0.002 abc	0.075 ± 0.003 d	0.57 ± 0.02 b
SKC15	0.273 ± 0.003 e	1.94 ± 0.05 ef	1.75 ± 0.05 ef	0.030 ± 0.002 abcd	0.106 ± 0.008 d	0.65 ± 0.02 b
SKC16	0.296 ± 0.003 e	1.54 ± 0.06 ef	1.33 ± 0.06 ef	0.027 ± 0.001 abcd	0.080 ± 0.005 d	0.56 ± 0.01 b
Evaporation ponds	0.534 ± 0.026 **	3.49 ± 0.24 *	3.26 ± 0.24 *	0.023 ± 0.002	0.263 ± 0.019 **	0.314 ± 0.021
Crystallizer ponds	0.392 ± 0.037	2.63 ± 0.28	2.41 ± 0.28	0.029 ± 0.001 *	0.171 ± 0.022	0.501 ± 0.043 ***
Ponds ηp2	0.179	0.106	0.103	0.181	0.120	0.252
Lineage A	0.351 ± 0.027 b	2.52 ± 0.27 b	2.31 ± 0.27 b	0.024 ± 0.001 a	0.149 ± 0.017 b	0.513 ± 0.037 a
Lineage B	0.610 ± 0.010 a	3.93 ± 0.29 a	3.70 ± 0.29 a	0.028 ± 0.003 a	0.323 ± 0.015 a	0.248 ± 0.013 b
Lineage C	0.564 ± 0.054 a	3.19 ± 0.21 ab	2.96 ± 0.21 ab	0.030 ± 0.004 a	0.243 ± 0.024 a	0.360 ± 0.035 b
Lineage ηp2	0.535	0.227	0.223	0.524	0.060	0.413

(means ± SE) and Supplementary Figure S5.

Pronounced strain-specific variability was observed across different growth parameters among the Dunaliella isolates. The maximum specific growth rate (μ_max) varied from 0.227 d⁻¹ (SKC12) to 0.661 d⁻¹ (SKE05), indicating substantial differences in intrinsic cell division capacity. Maximum cell density (N_max) ranged from 1.18 × 10⁶ cells mL⁻¹ (SKC14) to 5.68 × 10⁶ cells mL⁻¹ (SKC11), while maximum cell yield (Y_max) followed a similar pattern, spanning 0.98 × 10⁶ (SKC14) to 5.45 × 10⁶ cells mL⁻¹ (SKC11). Biomass-related indices also showed marked heterogeneity with maximum dry-weight biomass productivity (P_b,max) ranging from 0.013 (SKE06) to 0.038 (SKE05) g L⁻¹d⁻¹, and maximum cell productivity (P_c,_max) from 0.075 × 10⁶ (SKC14) to 0.413 × 10⁶ (SKE02) cells mL⁻¹d⁻¹. Final biomass concentrations (X_max) reached 0.19–0.81 g L⁻¹, with SKC13 producing the highest value.

Moreover, isolates originating from the evaporation ponds exhibited statistically significantly higher μ_max (0.534 d⁻¹), N_max (3.49 × 10⁶ cells mL⁻¹) and Y_max (3.26 × 10⁶ cells mL⁻¹), accompanied by greater P_c,_max (0.263 × 10⁶ cells mL⁻¹d⁻¹), collectively reflecting enhanced proliferative capacity, supported by moderate pond-level effect sizes. In contrast, crystallizer isolates attained significantly higher X_max (0.501 g L⁻¹) and P_b,_max (0.029 g L⁻¹d⁻¹), with comparatively stronger effect sizes, indicating improved biomass accumulation efficiency due to larger cell size and/or greater accumulation of intracellular compounds, highlighting strain-specific growth strategies. Within-lineage comparisons revealed significant among-strain variability across most kinetic parameters, underscoring substantial intraspecific physiological diversity. When grouped by phylogenetic lineage, lineage A displayed higher X_max values (0.513 g L⁻¹), presenting significantly greater final biomass accumulation compared to other lineages, with a large lineage effect size. In contrast, lineages B and C displayed significantly higher μ_max (0.610 and 0.564 d⁻¹, respectively) and P_c,_max (0.323 and 0.243 × 10⁶ cells mL⁻¹d⁻¹, respectively), denoting superior cell proliferation potential, although both attained statistically lower X_max compared to lineage A. Lineage-level differences were characterized by large effect sizes for μ_max and Pb,_max, moderate effect sizes for N_max and Y_max, and comparatively small effect sizes for Pc,_max. Overall, these findings demonstrate that Dunaliella growth performance is governed by both genetic lineage and environmental origin, with evaporation pond strains optimized for rapid proliferation and crystallizer pond isolates for maximal biomass accumulation under standardized culture conditions.

3.3. Biochemical Analyses

To assess the metabolic diversity and biotechnological potential of the sixteen Dunaliella isolates, a comprehensive biochemical characterization was conducted. The analysis encompassed primary macromolecular constituents and secondary metabolites including phenolic compounds and photosynthetic pigments. Substantial variability was observed both among individual isolates and across groups defined by pond origin and lineage.

The distribution of primary macromolecular constituents (proteins, carbohydrates, and lipids) among the sixteen Dunaliella isolates is presented in Figure 5. Protein content was highest in strains SKE08 (60.8% DW) and SKC11 (51.7% DW), whereas SKE04 exhibited the lowest protein allocation (22% DW). Carbohydrate levels ranged from ~7% to 19% DW, with SKC13 (19% DW) and SKC14 (16.9% DW) showing the greatest accumulation, in contrast to SKE05 and SKE03 (both ~7% DW). Lipid content varied between ~5% and 15% DW, reaching maximal values in SKC13 (14.7% DW) and SKE02 (13.7% DW), while SKC09 contained the lowest abundance (4.9% DW). Overall, among primary metabolites, only carbohydrate levels were significantly affected by pond origin. Isolates from crystallizer ponds exhibited a higher proportional allocation to carbohydrates, whereas those from evaporation ponds showed relatively greater, yet not statistically significant, protein content. Moreover, taxonomic affiliation further shaped macromolecular allocation, with lineage A strains exhibiting the highest carbohydrate content (13.3% DW) and the lowest protein content (28.4% DW). Collectively, these results illustrate the strong influence of both ecological origin and phylogenetic lineage on the metabolic profiles of Dunaliella isolates.

Alongside primary macromolecules, additional biochemical traits, including secondary metabolites and pigments, were quantified to further characterize the metabolic profiles of the sixteen Dunaliella isolates. These parameters encompassed total phenolic content (TPC), chlorophyll a (Chl a), chlorophyll b (Chl b), and total carotenoids (TC), which collectively provide insights into the antioxidant potential and photosynthetic capacity of the strains. Considerable intra- and inter-group variation was observed, with significant differences detected between evaporation and crystallizer pond isolates (

Table 2. Total phenolic content (TPC), chlorophyll a (Chl a), chlorophyll b (Chl b), and total carotenoids (TC) in sixteen Dunaliella isolates from evaporation (SKE01–SKE08) and crystallizer ponds (SKC09–SKC16). Values are expressed as mean ± standard error (SE). Group means (bottom rows) represent average values for isolates by pond origin and lineage. Different superscript letters within each column indicate statistically significant differences among isolates or groups (Tukey’s HSD, p < 0.05). Significance levels for group comparisons: **** p < 0.0001.

Isolate	TPC (mg GAE/g DW)	Chl a (mg/g DW)	Chl b (mg/g DW)	TC (mg/g DW)
SKE01	3.97 ± 0.18 cde	4.51 ± 0.19 cd	1.48 ± 0.07 def	2.33 ± 0.11 ef
SKE02	4.57 ± 0.23 c	4.33 ± 0.14 cde	1.49 ± 0.04 def	2.40 ± 0.12 ef
SKE03	2.98 ± 0.30 e	3.34 ± 0.13 def	1.23 ± 0.05 efg	1.69 ± 0.07 f
SKE04	3.96 ± 0.04 cde	5.27 ± 0.27 bc	2.04 ± 0.11 abc	2.02 ± 0.10 ef
SKE05	4.54 ± 0.12 c	6.67 ± 0.27 a	2.34 ± 0.10 a	2.91 ± 0.12 de
SKE06	3.84 ± 0.01 cde	2.33 ± 0.06 f	1.02 ± 0.03 fg	2.30 ± 0.04 ef
SKE07	3.07 ± 0.11 e	3.17 ± 0.15 ef	0.98 ± 0.04 g	2.41 ± 0.14 ef
SKE08	4.72 ± 0.20 c	6.19 ± 0.22 ab	2.24 ± 0.08 ab	2.45 ± 0.09 def
SKC09	4.17 ± 0.10 cd	4.25 ± 0.24 cde	1.45 ± 0.09 defg	1.84 ± 0.11 f
SKC10	3.51 ± 0.15 de	4.54 ± 0.15 cd	1.55 ± 0.06 cde	2.15 ± 0.09 ef
SKC11	4.27 ± 0.10 cd	5.36 ± 0.11 bc	1.85 ± 0.05 bcd	2.23 ± 0.05 ef
SKC12	6.09 ± 0.15 b	4.50 ± 0.57 cd	1.79 ± 0.22 bcd	4.49 ± 0.46 b
SKC13	8.68 ± 0.17 a	6.00 ± 0.33 ab	2.22 ± 0.13 ab	7.34 ± 0.35 a
SKC14	6.15 ± 0.24 b	4.26 ± 0.35 cde	1.53 ± 0.12 de	3.45 ± 0.29 cd
SKC15	6.45 ± 0.47 b	4.26 ± 0.35 cde	1.63 ± 0.13 cde	4.61 ± 0.37 b
SKC16	5.93 ± 0.15 b	5.99 ± 0.16 ab	2.23 ± 0.05 ab	4.05 ± 0.14 bc
Evaporation ponds	3.96 ± 0.12	4.42 ± 0.26	1.58 ± 0.09	2.31 ± 0.07
Crystallizer ponds	5.66 ± 0.29 ****	4.92 ± 0.19	1.79 ± 0.06	3.74 ± 0.32 ****
Lineage A	5.43 ± 0.28 a	4.63 ± 0.24 a	1.73 ± 0.09 a	3.69 ± 0.29 a
Lineage B	4.10 ± 0.10 b	4.60 ± 0.11 a	1.57 ± 0.04 a	2.19 ± 0.06 b
Lineage C	3.76 ± 0.33 b	5.01 ± 0.64 a	1.79 ± 0.22 a	2.30 ± 0.24 b

).

The TPC of the sixteen Dunaliella isolates ranged from 2.98 to 8.68 GAE/g DW. The highest TPC was recorded in strain SKC13 (8.68 mg GAE/g DW), followed by SKC15 (6.45 mg GAE/g DW), whereas the lowest was observed in strain SKE03 (2.98 mg GAE/g DW). On average, isolates from crystallizer ponds displayed significantly higher TPC values (5.66 mg GAE/g DW) than those from evaporation ponds (3.96 mg GAE/g DW). Similarly, lineage A strains exhibited the highest phenolic content (5.43 mg GAE g⁻¹ DW) relative to the other lineages. In addition, Chl a concentration ranged from 2.33 to 6.67 mg/g DW, with the highest levels detected in strains SKE05 (6.67 mg/g DW) and SKE08 (6.19 mg/g DW). In contrast, strains SKE03, SKE06, and SKE07 accumulated less than 4 mg Chl a/g DW. Moreover, among the different isolates Chl b levels varied between 0.98 and 2.34 mg/g DW, with elevated contents (>2 mg/g DW) measured in SKE05, SKE08, SKC13, and SKC16, whereas SKE06 and SKE07 showed the lowest accumulation. TC content spanned 1.69–7.34 mg/g DW, peaking in SKC13 (7.34 mg/g DW), followed by SKC15 (4.61 mg/g DW) and SKC12 (4.49 mg/g DW), while SKE03 exhibited the lowest level (1.69 mg/g DW). When grouped by pond origin, crystallizer isolates demonstrated significantly higher carotenoid accumulation (3.74 mg/g DW) relative to evaporation pond strains (2.31 mg/g DW). Additionally, when grouped by lineage, strains of lineage A showed the highest carotenoid content (3.69 mg/g DW) compared to the other lineages. In contrast, no significant differences between pond types or lineage were detected for Chl a or Chl b.

3.4. Antioxidant Capacity

The antioxidant capacity of the sixteen Dunaliella isolates was evaluated using FRAP and TEAC assays (

Table 3. Antioxidant capacity of sixteen Dunaliella isolates from evaporation (SKE01–SKE08) and crystallizer ponds (SKC09–SKC16), determined by FRAP (Ferric reducing antioxidant power) and TEAC (Trolox equivalent antioxidant capacity) assays. Values are expressed as mean ± standard error (SE). Group means (bottom rows) represent average values for isolates by pond origin and lineage. Different superscript letters within each column indicate statistically significant differences among isolates or groups (Tukey’s HSD, p < 0.05). Significance levels for group comparisons: **** p < 0.0001.

Isolate	FRAP (mg Trolox/g DW)	TEAC (mg Trolox/g DW)
SKE01	40.15 ± 1.40 efg	24.16 ± 0.55 gh
SKE02	44.04 ± 1.78 def	24.90 ± 0.36 fgh
SKE03	28.13 ± 3.15 hi	18.48 ± 0.62 i
SKE04	44.07 ± 2.46 def	27.86 ± 0.30 defg
SKE05	46.24 ± 1.04 def	27.79 ± 0.30 defg
SKE06	39.40 ± 0.88 efg	26.75 ± 0.31 efg
SKE07	27.23 ± 1.39 i	21.69 ± 0.47 hi
SKE08	48.22 ± 2.28 de	32.57 ± 1.23 c
SKC09	37.68 ± 1.36 fgh	27.99 ± 0.30 def
SKC10	31.96 ± 1.24 ghi	22.49 ± 0.59 h
SKC11	44.38 ± 1.33 def	28.26 ± 0.26 de
SKC12	53.37 ± 1.42 cd	30.26 ± 0.36 cde
SKC13	76.73 ± 0.64 a	42.31 ± 1.07 a
SKC14	59.31 ± 1.04 bc	36.40 ± 0.24 b
SKC15	65.01 ± 4.37 b	43.89 ± 2.05 a
SKC16	63.68 ± 0.97 b	31.21 ± 0.22 cd
Evaporation ponds	39.68 ± 1.47	25.53 ± 0.74
Crystallizer ponds	54.02 ± 2.61 ****	32.85 ± 1.28 ****
Lineage A	53.00 ± 2.47 a	32.55 ± 1.18 a
Lineage B	39.64 ± 1.20 b	25.56 ± 0.54 b
Lineage C	37.18 ± 3.75 b	23.14 ± 1.79 b

). FRAP values ranged from 27.23 to 76.73 mg Trolox/g DW. The highest reducing power was recorded in SKC13 (76.73 mg Trolox/g DW), followed by SKC15 (65.01 mg Trolox/g DW) and SKC16 (63.68 mg Trolox/g DW). In contrast, the lowest activity was detected in SKE07 (27.23 mg Trolox/g DW) and SKE03 (28.13 mg Trolox/g DW). TEAC values spanned 18.48–43.89 mg Trolox/g DW. SKC15 exhibited the highest radical scavenging capacity (43.89 mg Trolox/g DW), followed by SKC13 (42.31 mg Trolox/g DW) and SKC14 (36.40 mg Trolox/g DW), while SKE03 showed the lowest value (18.48 mg Trolox/g DW). When grouped by pond origin, crystallizer isolates displayed statistically significantly greater antioxidant capacity in both assays, with mean FRAP and TEAC values of 54.02 and 32.85 mg Trolox/g DW, respectively, compared to 39.68 and 25.53 mg Trolox/g DW in evaporation pond isolates. Additionally, lineage A strains exhibited significantly higher antioxidant activity, with values of 53.0 mg Trolox/g DW (FRAP) and 32.2 mg Trolox/g DW (TEAC), compared to the other lineages.

To further examine the biochemical variability among the sixteen Dunaliella strains a principal component analysis (PCA) was performed. Individual variable loadings are provided in Supplementary Table S5. When studying the isolates by taxonomic lineage, plotting PC1 and PC2, representing a total contribution of 73% to the total variability, showed that the five strains of lineage A were closely grouped, SKE03 and SKE05 were also clustered together along the X-axis, while strain SKE04 was the most distant within lineage B, mostly affected by chlorophyll a and b (Supplementary Figure S6). The high proportion of variance captured by PC1 and PC2 reflects the strong covariance among several functionally related biochemical traits, which jointly contribute to the same principal components. In contrast, when grouped by pond origin, plotting PC1 and PC2 (together explaining 73% of the total variance) revealed no clear clustering (Supplementary Figure S7).

4. Discussion

Hypersaline environments such as solar saltworks are extreme yet highly productive ecosystems that sustain specialized microalgae capable of thriving under severe osmotic and light stress [56,57]. Within these habitats, the genus Dunaliella (Chlorophyta) is especially prominent due to its remarkable halotolerance, often emerging as the dominant and sometimes sole oxygenic primary producer in salt-saturated waters [58]. Saltworks ecosystems, including those of Greece, therefore represent ecologically valuable reservoirs of extremophilic biodiversity, supporting dense and metabolically versatile microalgal communities [14,29]. In this context, the isolation of sixteen Dunaliella strains from the Kalloni solar saltworks (Lesvos, Greece) underscores the rich intra-genus diversity that can be preserved even within a single hypersaline site. In this study, we report their detailed characterization, providing insights into the links between isolation sites, growth kinetics, phylogeny, and biochemical characteristics. Our findings align with previous reports of Dunaliella strains in saltworks and further highlight the biochemical profiles and biotechnological potential of these new isolates.

4.1. Phylogenetic Clustering Reveals the Biodiversity of Dunaliella Isolates Across Three Lineages

Previous studies have reported the successful isolation of Dunaliella strains from highly saline natural environments. For instance, Gharajeh et al. (2020) [59] described three strains, each isolated from a distinct site in the Persian Gulf, a salt marsh, and a salt lake. Another study reported 15 strains assigned to nine species, isolated from nine saltworks [31], suggesting that local hypersaline environments act as hotspots for fine-scale diversification rather than being dominated by a single species. In line with this literature, our study presents 16 new Dunaliella isolates from two ponds, further highlighting the substantial biodiversity potential of such extreme environments.

Based on multi-gene phylogenetic analysis of the isolated strains and across all four gene-markers, within-lineage sequence identities were frequently high (99.6–100%), whereas between-lineage identities were much lower, with ITS sequences diverging below ~93%. This pattern indicates limited gene flow among clades and reveals strong genetic similarity within lineages. These results support the established understanding that the Dunaliella genus harbors hidden diversity that is not adequately represented by morphological characteristics alone [60]. Previous surveys in Greece [31] have identified multiple coexisting Dunaliella lineages in adjacent basins and lagoons, suggesting that local hypersaline environments act as hotspots for fine-scale diversification rather than being dominated by a single species. Our three-clade pattern is consistent with recent literature and the consistent divergence of SKE03 and SKE05 across nuclear and plastid markers, supports the hypothesis of distinct species or at least ecotypes. However, we found no evidence of clear lineage partitioning between the two isolation sites (crystallizer vs. evaporation ponds). These findings underscore that microalgae display substantial biodiversity even within extreme habitats such as solar saltworks, a diversity that remains largely uncharacterized and underexploited for biotechnological applications despite its considerable potential.

4.2. Coupled Genetic and Environmental Influences on Biokinetics and Biochemical Traits of Isolated Dunaliella Strains

Since all sixteen isolates were cultivated under the same conditions prior to kinetic and biochemical assays, the observed differences in performance most likely reflect inherent genetic variation rather than temporary phenotypic plasticity. This interpretation is consistent with earlier studies showing that Dunaliella lineages persistent trait variations are linked to their genetic background, even under standardized cultivation [61]. In this study, SKE05 from the divergent lineage C displayed the maximum specific growth rate and the highest biomass productivity alongside a moderate allocation of proteins and carbohydrates and pigments, indicating a strategy focused on rapid primary production (Supplementary Table S4). In contrast, cluster A includes strains that exhibit lower growth capacity (reduced μ_max values), but enhanced pigment and antioxidant content in the final biomass. Cluster B occupied an intermediate position, with balanced protein levels and moderate to high productivity. This contrast points to divergent evolutionary strategies, with one strategy emphasizing rapid cell division supported by high chlorophyll content and moderate protein/carbohydrate allocation, while the other favors slower growth accompanied by the accumulation of pigments, phenolics, and storage compounds, leading to larger cell sizes and high final biomass. Comparable lineage–trait coupling has been reported in Mediterranean saltworks [22,62], where D. salina strains differed markedly in growth and pigment profiles despite identical cultivation conditions. Other studies also indicate that performance is shaped by heritable lineage-specific traits [28,63].

Additionally, biochemical allocation patterns were significantly influenced by the site of origin, with clear distinctions between evaporation and crystallizer pond isolates. Evaporation ponds represent intermediate stages of solar saltworks, where seawater gradually concentrates, whereas crystallizer ponds are terminal basins with near-saturation salinity and active halite precipitation. These habitats impose the most extreme osmotic and oxidative stress, forcing Dunaliella to undergo extensive metabolic alterations to survive [21,64]. Persistent selective pressure in these extreme environments has likely driven the evolutionary fixation of traits that favor the accumulation of stress-protective metabolites and a larger cell volume. In this study, crystallizer isolates (SKC09–SKC16) exhibited elevated levels of carbohydrates, carotenoids, phenolics, and antioxidant capacity, whereas evaporation pond strains retained comparatively protein-rich or more balanced biochemical profiles, even after standardized cultivation. Functionally, carotenoids act as photoprotectants and antioxidants [65,66], phenolics provide a potent antioxidant defense [67], and carbohydrates contribute both to osmotic adjustment and to extracellular surface shielding in hypersaline ecosystems [68,69]. While residual maternal or isolation-related carry-over effects cannot be entirely excluded, the application of multiple successive acclimation and cultivation cycles under identical laboratory conditions prior to all physiological and biochemical measurements minimizes such effects and supports a predominantly genetic basis for the observed trait differences. Conclusively, these findings emphasize that metabolic allocation patterns are strongly shaped by the environmental origin of the strains, suggesting that long-term ecological selection in different pond habitats has reinforced distinct genetic adaptation.

The coexistence of genetically distinct Dunaliella lineages across successive saltwork ponds underscores the dynamic and complex nature of these hypersaline ecosystems, where microbial populations adapt to gradients of salinity and ionic composition [21]. As water progresses from the initial evaporation stages to crystallizer ponds, physicochemical transitions, including increasing salinity, ionic imbalances, and altered light penetration that progressively influence community structure, favoring highly halotolerant, carotenoid-rich Dunaliella strains [70]. This spatial succession is often reversible or cyclical, driven by seasonal fluctuations and water management practices, enabling strains adapted to intermediate salinities to recolonize both more saline and less saline habitats intermittently [71]. These dynamics reflect the intricate balance between physiological adaptation and ecological interactions shaping community assembly. Future incorporation of culture-independent molecular approaches, including quantitative PCR and metagenomic analyses tracking the taxonomic and functional shifts along this evaporation–crystallization continuum, would provide valuable insights into the ecological dynamics, dominance patterns, and adaptive mechanisms that drive Dunaliella diversification within solar saltworks [72].

4.3. Biokinetic and Biochemical Diversity of Dunaliella Strains Highlights Multifaceted Biotechnological Potential

From a biotechnological perspective, the Kalloni saltworks isolates constitute a diversified set of candidate strains rather than a single “optimal” strain, with its quantified metrics aligning with applications previously established for Dunaliella. In terms of biokinetics, the isolates demonstrated rapid growth rates and high biomass productivity, comparable to well-studied Dunaliella reference strains, including the model D. salina CCAP 19/18, which has been extensively characterized for its growth physiology and carotenoid production [73,74,75]. The combination of fast growth and enhanced biomass accumulation, particularly in isolate SKE05, underscores the versatility of these strains as cultivation candidates, with the potential to be tailored for different process objectives. Rapidly growing isolates could potentially be favored for bulk biomass accumulation in repeated-harvest cultivation strategies, whereas strains exhibiting comparatively higher productivity of biomass and associated metabolites under controlled laboratory conditions may represent promising candidates for future investigations aimed at maximizing yields of high-value compounds, such as carotenoids, glycerol, and compatible solutes, subject to validation under scalable cultivation systems [76,77].

Moreover, the Kalloni protein-rich isolates surpassed previously reported values for Dunaliella strains exploited in nutraceutical and pharmaceutical contexts, including the food-candidate D. viridis TAV01 (35.7% DW) highlighting their potential for high-value applications [22,59,78,79]. Specifically, isolate SKE05 exhibited both maximum biomass productivity and high protein content, similar to Dunaliella sp. ABRIINW-G2/1 which combined biomass yields of ~2.3 g/L with protein contents of ~39–41% DW and were highlighted for applications in nutraceuticals and aquafeeds [59]. These observations further suggest that certain evaporation pond isolates may possess traits of interest for agri-food and feed-related research applications, potentially reducing the need for stress-induction protocols. Lipid contents (5–15% DW) also fall within the reported range for Dunaliella species exploited in food, nutraceutical, and bioenergy sectors, supporting their valorization through extraction of polyunsaturated fatty acids, bioactive lipids, and biodiesel precursors [77,80,81]. Carbohydrate-rich isolates (e.g., SKC13, SKC14) may be of interest for biorefinery models where polysaccharides, fermentable sugars, and extracellular polymers are exploited as co-products [80,82,83]. Notably, both lipid and carbohydrate yields may be strategically enhanced under stress conditions including nutrient limitation (e.g., nitrogen starvation), salinity stress or high light intensity [84,85,86], providing an additional lever for targeted valorization strategies [87,88].

Beyond their primary metabolite profiles, the isolated strains exhibited substantial accumulation of secondary metabolites, with carotenoids and phenolics emerging as prominent bioactive targets. Carotenoid-rich strains especially from the crystallizer ponds (SKC13–SKC15) fall within the range reported for Dunaliella isolates [31,45,89] and align with commercial D. salina model for β-carotene and xanthophyll production like 19/18 CCAP strain [75]. Strains SKC13 and SKC14, despite their low growth rate and cell productivity, accumulated the highest carotenoid levels due to their large cell size and pigment storage capacity. This pattern is consistent with reports on Dunaliella strain CONC-007, which exhibited low growth kinetics but the highest β-carotene yields per culture volume [90], making it notable in the context of carotenoid accumulation dynamics. Nonetheless, D. salina production systems can achieve percent-level β-carotene concentrations under two-phase or stress-induction protocols [91]. Moreover, the isolates also exhibited elevated phenolic content and strong antioxidant activity. Reports in the recent literature indicate a broad range of phenolic accumulation and antioxidant capacity across both wild and commercial Dunaliella strains. Notably, in this study, isolates obtained from the crystallizer ponds displayed particularly high phenolic content and antioxidant activity, ranking among the highest values reported to date for this genus [92,93,94,95]. These findings highlight the concurrent enrichment of carotenoids and phenolics in the Kalloni isolates, together with their elevated antioxidant capacity, underscoring their potential as high-value bioresources. The combined bioactive profile and elevated antioxidant activity further underscore the potential relevance of these strains for future application-oriented research in functional food, nutraceutical, and pharmaceutical contexts [96].

5. Conclusions

Our findings indicate that the sixteen Dunaliella isolates from Kalloni solar saltworks fall into three genetic lineages whose trait differences remain stable under standard cultivation conditions. The increased allocation of carotenoids, phenolics, and carbohydrates in strains from crystallizer ponds, alongside the protein-centric allocation in evaporation ponds strains, can be attributed to genetic differences that have been shaped by prolonged selection at their respective source ponds, rather than by short-term acclimatization. Additionally, SKE05 strain exhibits the highest maximum growth rate and biomass productivity, suggesting potential suitability for high-throughput biomass production; SKC13–SKC15, characterized by elevated levels of pigments and carbohydrates, with potential relevance for carotenoid and antioxidant products along with carbohydrate co-products, while SKE08/SKC11 are identified as protein-rich candidates for further investigation in food and feed research. Consequently, the Kalloni saltworks harbor a diverse array of strains representing a valuable biological resource for sustainable bioproduction research, with relevance to agri-food, nutraceutical, cosmeceutical, and biorefinery sectors. The translation of these laboratory-scale observations will require further evaluation of cultivation scalability, process optimization, cost-effectiveness, and downstream processing constraints. Future directions involve genome-scale delineation, particularly for the SKE03/SKE05 lineage, pilot-scale cultivation, and process optimization to convert these baseline traits into industrial-scale yields. Additionally, in the future detailed metabolome analysis in sixteen isolates will be performed and the molecular and biochemical mechanisms involved in adaptation in high salinity environments between the different Dunaliella genotypes will be explored.