Halamphora kolbei (Aleem) Álvarez-Blanco et S. Blanco 2014, A Rare Diatom from the Black Sea: Morphological Observations and Revised Description with Biochemical Composition

Simple Summary

This study addresses the isolation and taxonomic characterisation of the poorly known diatom Halamphora kolbei, integrating fundamental and applied research perspectives. Due to its small cell size and incomplete original description, H. kolbei has remained difficult to identify, and published data on the species are extremely scarce. In this work, we summarise the available literature and provide a detailed morphological description of a strain isolated from Sevastopol Bay (Black Sea) using light microscopy, scanning electron microscopy, and phylogenetic analysis. In addition, the strain was evaluated for its productivity and biochemical composition. The results demonstrate that H. kolbei is a promising source of eicosapentaenoic acid, an essential omega-3 polyunsaturated fatty acid. This study contributes to the taxonomy and phylogeny of the genus Halamphora and highlights the biotechnological potential of this species for the sustainable production of high-value nutrients.

Abstract

The species Halamphora kolbei is a rarely reported diatom, and information on its distribution and characteristics is limited. The original taxonomic description of this species is inaccurate and insufficient for identification, as the available literature contains scarce and scattered data. In the present study, we summarised the available data on this species and provided a detailed description of the strain isolated from Sevastopol Bay of the Black Sea. Findings from a phylogenetic analysis of two genes, 18S rRNA and rbcL, are presented. The process of isolation, identification and establishment of a monoculture of the H. kolbei is described in detail. In addition, the productivity and biochemical characteristics were evaluated to assess the biotechnological potential of the strain. The growth rate, biomass productivity and protein, carbohydrate, lipid and fatty acid composition of this species were determined in this study, under controlled conditions. The high content of C14–C16 acids in H. kolbei has been revealed to make it a promising source of eicosapentaenoic acid. The detailed description of the species, its photographs and phylogeny, as well as additional morphological keys presented in the paper, will enable scientists around the world to study this species as a promising biological resource.

1. Introduction

For a long time, most dorsiventral diatoms were classified as belonging to the extremely extensive genus Amphora C.G. Ehrenberg ex F. T. Kützing 1844. In fact, Amphora sensu lato is broadly defined, encompassing a heterogeneous group of diatoms that includes species from several distinct genera: Amphora Ehrenb. ex Kütz. 1844 sensu stricto, Halamphora (P. T. Cleve) Levkov [1,2,3], Tetramphora Mereschkowsky [4] and Seminavis D. G. Mann [5]. Due to the morphological similarities between species from different genera, high diversity and numerous misinterpretations, an increasing number of studies are being published to clarify their phylogeny [1,2,3,6,7,8,9,10,11,12,13,14,15].

In 2009, Z. Levkov revised the genus Amphora and separated the new genus Halamphora (P. T. Cleve) Levkov 2009, based on its morphological differences. The author’s description of the genus includes: solitary cells, moderately or strongly dorsiventral. Frustules in girdle view are broadly elliptical to rectangular with shortly protracted and broadly rounded to truncate ends, always asymmetrical about the apical plane. The two raphe systems are positioned on the same side of the valve. The valves are linear, semi-lanceolate to semi-elliptical, with a smoothly arched dorsal margin that is straight or constricted in the middle, and a ventral margin that is concave or weakly convex with inflation. The valve ends are variable: extended, shortly subprotracted, broadly rounded, or narrowly rounded and capitate. The ventral mantle is shallow. The striae are uniseriate or biseriate and consist of round, elliptical or transversely elongated areolae. The raphe lies on a raphe ledge and is elevated above the rest of the valve on the dorsal side. The proximal raphe endings are straight or bent ventrally. The distal raphe endings are dorsally bent. The distal raphe endings terminate internally with poorly developed helictoglossae and the proximal endings with fused central helictoglossae [1]. Despite its close resemblance to the genus Amphora, two main distinguishing features of the genus Halamphora have been identified: (1) the presence of a single H-shaped chloroplast appressed to the ventral girdle and strongly constricted in the middle; (2) the girdle consists of numerous bands that comprise open copulae with one or two rows of round or transversely elongated poroids. In addition, most species of the genus Halamphora are brackish and marine, while species of the genus Amphora are freshwater or brackish [1]. Furthermore, Amphora sensu lato, until recently, was a polyphyletic group that included unrelated genera that are similar in their dorsiventral form. However, molecular phylogenetic studies have confirmed that the species classified as Halamphora and Amphora form separate, albeit related, clades [3].

Currently, numerous studies describe various species of the genus Halamphora, enabling their accurate identification [16,17,18,19]. There are many small-celled species within the genus Halamphora that are rare in nature and difficult to diagnose. Such species include Halamphora kolbei (Aleem) Álvarez-Blanco et S. Blanco 2014. This species was originally described in a work devoted to littoral diatoms of the Eastern Mediterranean (Egyptian coast), where drawings of the species were presented [20]. This species was reported from the coastal waters of the Volo Bay, Greece (Aegean Sea) [21] and in the Mediterranean Sea in general without a specific location [22]. Later, after Z. Levkov distinguished the genus Halamphora [1], A. kolbei was reclassified from the genus Amphora to this genus on the basis of morphological studies conducted using light and scanning electron microscopy [23]. As H. kolbei, the species was reported in Neum Bay of the Adriatic Sea [24]. Overall, the highest number of records for this species is from the seas of the Mediterranean region. Additionally, H. kolbei has been found in the Gulf of Mexico (Atlantic Ocean) [25,26] and in the Gulf of California, a marginal sea of the Pacific Ocean [27,28].

Halamphora kolbei, a species described for the Mediterranean Sea, had not been previously documented in the Black Sea. It was recently discovered off the Turkish coast of the Black Sea [29], and later off the Bulgarian coast of the Black Sea [30]. In references describing H. kolbei, the morphological description is limited to characteristics such as length, width, and dorsal striae [20,21,22,23,29,30]. More fine details, such as girdle bands and poroids, are not provided, and the presence of ventral striae remains unclear. In total, the species is mentioned five times in the literature that includes a description of it [22,23,25,29,30]. Its rarity might result from the challenge of accurately identifying it in natural samples. At the same time, a similar species, Amphora bigibba Grunow 1875, is quite widespread in the Black Sea [31,32,33,34] and has been found in all climatic zones except the Antarctic [27,28,35,36,37,38,39,40,41,42]. H. kolbei may be confused with A. bigibba because only a few studies contain complete, detailed morphological descriptions of this species, and no phylogenetic studies have been conducted. The identification of species is a key part of many tasks, including the search for biological resources, ecology, and biodiversity. Morphological descriptions and genetic data are not available for H. kolbei. Therefore, a comprehensive study of H. kolbei, examining its morphology, genetics, and biochemical composition, is important to analyse its biotechnological potential. All of these are elements of bioprospecting, a process that is extremely important today. The term ‘bioprospecting’ is used to describe the process of searching for bioactive compounds from natural sources, such as plants, fungi, bacteria, and marine organisms. Bioprospecting from natural products involves biological material selection, collection, isolation, identification of natural organisms, bioactivity screening, safety and activity validation, mass production, and commercialisation [43]. The main stage of this process is searching for species. The concept of a species is based on genetic, biochemical and phenotypic similarities. A strain comprises the descendants of a single organism and is classified based on its unique characteristics (ecological, biochemical, functional, etc.). The term strain is commonly used to denote a pure culture [44,45,46]. Scientific works on the isolation and identification of microorganisms contribute to solving both fundamental and applied global problems. On the one hand, such works are significant in terms of evolution and taxonomy, contributing to our understanding of microbiological ecology and the preservation of biodiversity [47]. On the other hand, this process has historically played a pivotal role in the development of new biotechnologies and drugs, with many therapeutics derived from or inspired by natural products [48].

Marine microalgae represent a valuable reservoir of bioactive molecules, offering significant potential for various biotechnological sectors, including the food, feed, pharmaceutical, cosmetic, and biofuel industries. Research into biological resources is underway, with scientists searching globally for biotechnologically valuable diatoms [49,50]. Bioprospecting has revealed that diatoms can synthesise biologically active compounds such as polyunsaturated fatty acids, carotenoids, fucoxanthin, lipids, and sterols, among others [50,51,52,53,54]. Evidence suggests that species of the genera Amphora and Halamphora exhibit high lipid content and a unique fatty acid composition [49,55,56,57,58,59,60]. Isolation and cultivation of microalgal strains of interest are essential steps to laboratory investigations and successful commercial application. This is a stepwise process with each step playing an integral role in the choice of the following [44]. Therefore, obtaining pure cultures of microalgal strains from these genera to study their morphology, taxonomic affiliation, and biochemical features addresses several urgent scientific issues in both fundamental and applied algology.

The goal of our research was to isolate an H. kolbei strain from the Black Sea in pure culture for morphological description, genetic identification, determination of culture growth kinetics, and analysis of biomass composition and specific compound content (fatty acids and fucoxanthin).

2. Materials and Methods

2.1. Isolation and Cultivation

Marine diatoms of epilithon from the Sevastopol Bay of the Black Sea were sampled in September 2021 (44°62′8229″ N, 33°52′4878″ E). Samples were collected at the water temperature of 20 °C and salinity of 18 psu. For H. kolbei isolation, a small volume of this suspension was transferred into Petri dishes and cultured in sterile F/2 medium [61] prepared with filtered seawater. The medium was sterilised by autoclaving at 1 atmosphere and a temperature of 121 °C for 15 min. After approximately one week of enrichment, single living diatom cells were isolated using micropipettes under a Bresser Bio Science Trino (Bresser Optics, Rhede, Germany) light microscope (LM). Diatom cells were transferred into sterile Petri dishes filled with culture medium, labelled and placed under controlled conditions, close to those found in nature (temperature 20 ± 1 °C, irradiation 18.5 μmol photons m⁻² s⁻¹ (4 W m⁻²), 14 h:10 h L:D cycle) in a KS-200 climate chamber (Russia). The isolated H. kolbei strain was registered in the microalgae and cyanobacteria working collection of the Department of Aquaculture and Marine Pharmacology of the A.O. Kovalevsky Institute of Biology of the Southern Seas of RAS in 2022. Strain ID: DIA-A.big.-SevBay.

2.2. Morphological Identification

Images of live cells were obtained using an Axioskop 40 C. Zeiss (Carl Zeiss AG, Oberkochen, Germany) LM at magnifications of 10 × 20, 10 × 40, and 10 × 100. Samples for scanning electron microscopy (SEM), Hitachi SU3500 (Hitachi LTD, Tokyo, Japan) were prepared according to [62]. The diatom suspension was cleaned of organic matter by treating it with KMnO₄ for 24 h, after which HCl was added, and the mixture was heated to remove insoluble salts (e.g., carbonates). The acid was then removed from the samples by rinsing them with distilled water, a process that involved repeated centrifugation. Dried diatom valve preparations were coated with gold-palladium for SEM visualisation.

The terminology of valve ultrastructure follows that of [1,31,32,63,64,65,66,67] and articles [68,69]. Systematics are provided in accordance with [70].

2.3. Molecular Study

Total DNA from the studied strain was extracted using Chelex 100 Chelating Resin, molecular biology grade (Bio-Rad Laboratories, Hercules, CA, USA), according to the manufacturer’s protocol. Partial SSU rRNA (378–382 bp, including the highly variable V4 region) and partial rbcL plastid genes were amplified using primers D512for (5′-ATTCCAGCTCCAATAGCG-3′) and D978rev (5′-GACTACGATGGTATCTAATC-3′) from [71] for SSU rDNA fragments, and rbcL404+ (5′-GCTTTACGTTTAGAAGATATG-3′) and rbcL1444- (5′-GCGAAATCAGCTGTATCTGTW G-3′) from [72] for rbcL fragments. Amplifications were performed using premade polymerase chain reaction (PCR) master mixes (ScreenMix by Evrogen, Moscow, Russia). The SSU rRNA gene was amplified under the following conditions: initial denaturation for 5 min at 95 °C, followed by 35 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 50 s extension at 72 °C, with a final extension for 10 min at 72 °C. The rbcL gene was amplified under the following conditions: initial denaturation for 4 min at 94 °C, followed by 40 cycles of 50 s denaturation at 94 °C, 50 s annealing at 53 °C, and 80 s extension at 72 °C, with a final extension for 7 min at 72 °C. PCR products were visualised by horizontal electrophoresis in 1.0% agarose gel stained with SYBR^TM Safe (Life Technologies, Carlsbad, CA, USA). The products were purified with a mixture of FastAP, 10× FastAP Buffer, Exonuclease I (Thermo Fisher Scientific, Waltham, MA, USA), and water. Sequencing was performed by Syntol LLC (Moscow, Russia).

Consensus sequences were assembled and edited from forward and reverse chromatograms using Ridom TraceEdit ver. 1.1.0 (Ridom GmbH, Münster, Germany) and Mega ver. 11 software [73].

For phylogenetic analysis, strains of Amphora Ehrenberg ex Kützing and Halamphora (Cleve) Levkov were selected from GenBank for which SSU rRNA and rbcL gene sequences were available. Two Tetramphora Mereschkowsky strains were chosen as outgroups. The reads of the H. kolbei strain were included in the alignments along with corresponding sequences of 42 diatom species downloaded from GenBank (taxa names and accession numbers are given in Table S1).

Nucleotide sequences of the SSU rRNA and rbcL genes were aligned separately with the MAFFT ver. 11 software (RIMD, Osaka, Japan) and the E-INS-i model [74]. The resulting alignments were then trimmed by removing the visually identified unpaired sites from the beginning and end of the resulting matrices. For the protein-coding sequences of the rbcL gene, it was verified that the beginning of the aligned matrix corresponds to the first position of the codon (triplet). The final alignments had lengths of 450 (SSU rDNA) and 994 (rbcL) characters. After removal of the unpaired regions, the aligned SSU rRNA gene sequences were combined with the rbcL gene sequences into a single matrix using Mega11 for phylogenetic reconstruction.

Bayesian inference (BI) was used to analyse the dataset with BEAST ver. 1.10.1 software (BEAST Developers, Auckland, New Zealand) [75]. The most appropriate partition-specific substitution models, shape parameter α, and proportion of invariable sites (pinvar) were selected using the Bayesian information criterion (BIC) in jModel-Test ver. 2.1.10 software (Vigo, Spain) [76]. This BIC-based model selection procedure selected the following models, shape parameter α, and proportion of invariable sites (pinvar): NKY + I + G, α = 0.4170, and pinvar = 0.6920 for the first codon position of the rbcL gene; JC + I + G, α = 0.4320, and pinvar = 0.8160 for the second codon position of the rbcL gene; GTR + G, α = 0.5530 for the third codon position of the rbcL gene, NKY + I + G, α = 0.2820, and pinvar = 0.5320 for 18S rDNA. A Yule process tree prior was used as the speciation model. Five MCMC analyses were run for seven million generations (burn-in: 1000 million generations). Convergence diagnostics were performed in the Tracer ver. 1.7.1 software (MCMC Trace Analysis Tool, Edinburgh, UK) [75]. The initial 10% trees were discarded, and the remaining trees were used to construct a final chronogram with 90% posterior probabilities. FigTree ver. 1.4.4 (University of Edinburgh, Edinburgh, UK) and Adobe Photoshop CC ver. 19.0 software (Adobe, San Jose, CA, USA) were used for viewing and editing the trees.

2.4. Intensive Cultivation

2.4.1. Cultivation Conditions

The cells of the obtained strain were used as an inoculum at a concentration of 10% (v/v) for batch cultivation in one-litre flasks. An intensive culture of H. kolbei was grown in 5F nutrient medium under constant temperature conditions of 20 ± 1 °C. The photobioreactor was provided with round-the-clock, one-sided illumination by white PILA LF36W/54-765-2450 lm lamps (Philips, Warszawa, Poland). The average irradiation of the photobioreactor’s working surface was 170 μmol photons m⁻² s⁻¹ (33 W m⁻²). During the growth process, the culture was bubbled with air at a rate of 1.1 litres per litre of culture per minute using a compressor unit. To increase the solubility of atmospheric CO₂, a dispersant attachment was used in the culture medium. 5F nutrient medium was prepared with sterile seawater supplemented with the following components: 0.75 g·L⁻¹ NaNO₃, 0.05 g·L⁻¹ NaH₂PO₄ × H₂O, 0.3 g·L⁻¹ Na₂SiO₃ × 9H₂O, 1∙10⁻⁴ g·L⁻¹ CuSO₄ × 5H₂O, 2.2∙10⁻⁴ g·L⁻¹ ZnSO₄ × 7H₂O, 1∙10⁻⁴ g·L⁻¹ CoCl₂ × 6H₂O, 1.8∙10⁻⁴ g·L⁻¹ MnCl₂ × 4H₂O, 0.63∙10⁻⁴ g·L⁻¹ Na₂MoO₄ × 2H₂O, 0.0437 g·L⁻¹ Na₂EDTA, 0.0315 g·L⁻¹ FeSO₄ × 7H₂O [77]. The nutrient medium was supplemented with 0.1 g/L of Polysorb MP sorbent (particle size up to 90 μm) (Polisorb, Kopeisk, Russia), as a microcarrier. The addition of the microcarrier prevented the culture from adhering to the flask walls. The density of the culture at the start of the batch cultivation was 0.2 g of dry weight (DW) per 1 L.

2.4.2. Growth Kinetics and Calculations

Microalgae growth was monitored daily by measuring the wet biomass weight on a Scout SPX222 (OHAUS Ltd., Parsippany, NJ, USA) analytical balance after centrifugation at 1600× g for 5 min at 20 °C on an OPN-3 centrifuge (TNK Dastan, Bishkek, Kyrgyzstan). The microalgae biomass was rinsed with distilled water via centrifugation at 6000× g for 15 min at the end of the cultivation. The supernatant was then removed, and the biomass was weighed on an analytical balance with an accuracy of 0.0001 g. Subsequently, the biomass was dried at 105 °C in a RA-50/500VZ drying oven (Micron, Shatura, Russia) to constant weight. The water fraction in the wet biomass was determined using the method described in [78]. A coefficient (k = 0.14), representing the relationship between dry weight and wet weight and determined experimentally, was used to convert the data to dry weight.

The specific growth rate was determined by measuring changes in culture density during the exponential growth phase using the equation:

$$\mu ={\displaystyle \frac{lnB-ln{B}_0}{t-t_0}}$$

where μ is the specific growth rate, day⁻¹; B—is the current culture density at time t, g·L⁻¹; B₀ is the culture density at the previous time t₀, g·L⁻¹.

The biomass productivity during the linear part of the exponential growth phase was calculated using the equation:

$$P={\displaystyle \frac{B-B_0}{t-t_0}}$$

where P is the biomass productivity, g·L⁻¹·day⁻¹; B is the current culture density at time t, g·L⁻¹; B₀ is the culture density at the previous time t₀, g·L⁻¹.

2.4.3. Analysis of Biomass Composition

Lipid Content and Fatty Acid Composition

Total lipid mass fraction was quantified via the phosphovanillin method [79]. An aliquot (10 mL) of an algal suspension with a known cell concentration was centrifuged at 1600× g for 5 min. The resulting algal pellet was then used for extraction after the removal of the supernatant. This was homogenised with quartz sand using a mechanical homogeniser, then extracted with a Folch mixture (first 5 mL at a 1:1 v/v ratio, then 5 mL at a 2:1 v/v ratio). The extraction process was repeated twice to ensure complete lipid recovery. The combined lipid extracts were washed 3–4 times with water to remove non-lipid contaminants. Subsequently, an aliquot of the purified lipid extract was evaporated and digested in sulfuric acid by heating in a water bath at 100 °C for 20 min. After cooling, the phosphovaniline reagent was added to develop colour. The optical density of the resulting solution was measured at a wavelength of 530 nm using an SF-2000 spectrophotometer (OKB Spektr, Saint Petersburg, Russia). A calibration curve was prepared using standard solutions of olive oil in chloroform as previously described [79].

The lipid productivity over the entire cultivation period was calculated using the following equation:

$$P_L={\displaystyle \frac{C_L·(B_{st}-B_0)}{t}}$$

where $P_L$ is the lipid productivity (mg·L⁻¹ day); $C_L$ is the lipid content in the biomass (mg·g⁻¹); $B_{st}$ is the culture density in the stationary growth phase (g·L⁻¹); $B_0$ is the culture density at the initial time point (g·L⁻¹); t is the cultivation time (days).

Fatty Acid (FA) Determination

Fatty acid analysis was performed by extracting lipids from the raw material using a Folch mixture. The extracted lipids were washed with distilled water to remove the non-lipid impurities.

A lipid extract aliquot with 3–5 mg of total lipids was evaporated to dryness using a rotary evaporator RV 10 (IKA, Staufen, Germany) at 25 °C. The residue was subjected to saponification by adding 5 mL of an alkaline methanol solution (10 mL of 3 N NaOH with 90 mL of 90% methanol) and heating at 64–65 °C for two hours. After cooling, a few drops of a 1% phenolphthalein solution in ethanol were added. The mixture was extracted three times with 5 mL of hexane to remove unsaponifiable fractions. The aqueous-alcohol phase was acidified by adding 0.3 mL of 6 N HCl. Then, fatty acids were extracted three to four times. The combined hexane extracts were evaporated to dryness using a rotary evaporator at 30 °C. To the residue, 5 mL of a 3% (v/v) hydrochloric acid in methanol was added for esterification of the FA, and the mixture was refluxed for two hours. After cooling, the resulting fatty acid methyl esters (FAMEs) were extracted three times with 5 mL of hexane. The hexane fraction was filtered through an ash-free filter paper and stored at −20 °C for no longer than 24 h prior to analysis.

FAMEs were analysed with a Khromatek Crystal 5000.2 (CJS Special Design Bureau “Khromatek”, Yoshkar-Ola, Russia) gas chromatograph equipped with a flame ionisation detector (FID) and a mass spectrometric detector (MSD). The analysis was performed at the Spectrometry and Chromatography core facility in A. O. Kovalevsky Institute of Biology of the Southern Seas of RAS. Separation was achieved using a 60 m BPX5 capillary column (Trajan Scientific Europe Ltd./SGE Analytical Science, Milton Keynes, UK) with an internal diameter of 0.25 cm and a stationary phase thickness of 0.25 μm. The temperature program was as follows: (1) hold at 40 °C for 1 min, (2) increase to 220 °C at 10 °C/min and hold for 6 min, then (3) increase to 280 °C at 10 °C/min and hold for 10 min.

The composition of FA was determined using PID. A mixture of acids, “Supelco 37 component FAME mix” (Sigma-Aldrich, St. Louis, MO, USA) was used as an external standard. FAs not present in the commercial standard mixture were identified by analysing the chromatogram obtained in total ion current mode. Their mass spectra were compared to reference spectra in the NIST 14 library. The mass scanning range was 50–450 m/z. Margaric acid was employed as an internal standard [80].

The FA productivity ($P_{FA}$) over the entire cultivation period was calculated using an equation analogous to that for lipid productivity, substituting the total lipid content with the total FA content in the biomass.

Determination of Total Carbohydrates

The carbohydrate content and mass fraction in the biomass were determined following the Dubois method [81] using crystalline glucose as the standard. A 1 mL sample was taken from a suspension with a predetermined cell concentration and centrifuged for five min at 1600× g. After discarding the supernatant, the resulting algal pellet was rinsed twice with 5 mL of distilled water. Subsequently, 2.5 mL of a phenolic reagent was added to the pellet. After 10 min, 2.5 mL of sulfuric acid was added, and the mixture was placed in a water bath at 100 °C for five min. Following colour development, the optical density of the samples was quantified at a wavelength of 490 nm using an SF-2000 spectrophotometer (OKB Spectr, Saint-Petersburg, Russia). A standard curve was prepared using glucose (0.1 µg·µL⁻¹).

Protein Determination

The protein content was determined by the Lowry method [82], using crystalline human serum albumin as the standard for the calibration curve. A 1 mL aliquot of a suspension with a predetermined cell concentration was centrifuged for five min at 3000× g. The supernatant was removed, and the algal pellet was washed with 5 mL of distilled water. Protein hydrolysis was then carried out in a water bath at 100 °C with 1 N alkali for 10 min. After hydrolysis, the samples were processed according to the Lowry protocol, and their optical density was measured using an SF-2000 spectrophotometer (Russia).

Extraction of Fucoxanthin

Fucoxanthin (Fx) content in the biomass was determined using thin-layer chromatography (TLC) [83]. Fx was extracted from the raw biomass with 96% ethanol using a mechanical homogeniser with quartz sand until the biomass was completely discoloured. An aliquot (0.1 mL) of the extract was applied to chromatographic glass plates coated with a 0.5 mm layer of silica gel. Carotenoid separation was performed using an acetone-hexane (3:7, v/v) solvent system. The silica gel zone corresponding to Fx was extracted with ethanol three times until complete decolouration. The combined extracts were centrifuged 3000× g. The Fx concentration was determined by measuring the optical density of the solution at a wavelength of 448 nm and converting it to the dry weight of the algae using the equation:

$$Fx={\displaystyle \frac{10\cdot V_1\cdot V_2\cdot D_{448}}{E\cdot V_3\cdot W}}$$

where Fx is the fucoxanthin concentration in the biomass, mg·g⁻¹; V₁ is the volume of the combined eluate from the TLC plate, mL; V₂ is the total volume of the initial extract, mL; V₃ is the volume of the aliquot applied to the TLC plate, mL; D₄₄₈ is the optical density measured at 448 nm; W is the dry weight of algal sample, g; and E = 1280 mL/(mg·cm) is the specific extinction coefficient for 1% solution of Fx in ethanol [84].

For identification, crystalline fucoxanthin, previously characterised by HPLC, mass spectrometry and NMR spectroscopy, was used as a standard [85].

The Fx productivity ($P_{Fx}$) over the entire cultivation period was calculated using an equation analogous to that for lipid or FA productivity, substituting the total lipid or FA content with the total Fx content in the biomass.

Determination of the Ash Residue

The mass fraction of the ash residue was determined, accounting for the biomass water content. It was calculated from the weight difference before and after sample combustion in a muffle furnace at 550 °C for 4 h.

2.5. Statistical Analysis

For each morphological characteristic (length, width, number of striae, and poroids), 30 measurements were taken. Statistical analyses were performed using Past 4.17 [86]. Data distribution was assessed using the Shapiro–Wilk normality test (W) for each dataset. Since the morphometric results had a non-normal distribution, they are presented as median (med) with interquartile range (IQR: 25th–75th percentile). All analyses of the culture characteristics (dry weight, culture density, productivity, and content of compounds in biomass) were performed in triplicate. The results are presented as mean ± confidence intervals (CI). CI for the mean were calculated using the Student t-distribution at the 95% confidence level using the built-in function in Microsoft Excel. Data plotting was performed using Grapher 3D ver. 3 (Golden Software LLC, Golden, CO, USA).

3. Results

In this research, we present H. kolbei, a little-known taxon discovered in the Black Sea epilithon, as a potential candidate for biotechnological applications. Our analysis is based on morphological and genetic data collected from cultured strains. We also explore the similarities and distinctions with morphologically similar species.

3.1. Systematics and Morphology

Class Bacillariophyceae Haeckel 1878 emend. D. G. Mann in Round et al. 1990

Order Naviculales Bessey 1907

Family Catenulaceae Mereschkowsky 1902

Genus Halamphora (P.T. Cleve) Levkov 2009

Halamphora kolbei (Aleem) Álvarez-Blanco et S. Blanco 2014

Description (Figure 1 and Figure 2). In natural populations, cells occur singly. In culture, the species forms a film. Chloroplast is H-shaped. Frustules are elliptical and constricted in the middle. The valves are asymmetric with respect to the apical axis but symmetric with respect to the transapical axis. The valve mantle is deeper on the dorsal margin than on the ventral margin. The raphe is located on the ventral margin. Apices are broadly rostrate. The cells are small, 8.6–15 µm in length (W = 0.894, p = 0.006; med = 13.5, IQR: 12.5–14.0), 3.5–8.5 µm in width at the widest part of the valve (W = 0.0771, p = 0.00002; med = 4.5, IQR: 4.0–5.1), 2.4–8.1 µm at the constriction (W = 0.748, p = 0.000008; med = 3.5, IQR: 3.0–4.2) and up to 12 µm in girdle view. Dorsal striae slightly radiate, 24–32 in 10 µm (W = 0.900, p = 0.008; med = 26, IQR: 26–28), ventral striae 34–38 in 10 µm (W = 0.880, p = 0.003; med = 36, IQR: 34–38). Girdle bands are numerous (near 10–16), with 36–42 poroids in 10 µm (W = 0.875, p = 0.002; med = 39, IQR: 37–42). Along the dorsal side of the sternum, there is a row of areolae oriented along the apical axis (indicated by arrows on Figure 2g,h), contrasting with the areolae on the rest of the dorsal surface.

Figure 1. Halamphora kolbei, strain DIA-A.big.-SevBay under LM: living cells (a,b); cleaned frustules in valve view (c). Scale bar: 10 μm.

Figure 2. Halamphora kolbei, strain DIA-A.big.-SevBay frustules, SEM: in ventral girdle view (a,b); external valve view (c–e); in dorsal girdle view demonstrating details of girdle bands (f); internal valve view (g,h). Scale bar: (a–g) 5 μm, (h) 2 μm. Blue arrows indicate a row of areolae that oriented along the apical axis.

3.2. Phylogenetic Analysis

Phylogenetic analysis based on the 18S rRNA and rbcL genes showed that Amphora and Halamphora strains form separate clades (Figure 3). The Halamphora clade has maximum statistical support (PP = 1), whereas support for the Amphora clade is lower (PP = 0.89).

Figure 3. Phylogenetic position of the Halamphora kolbei, strain DIA-A.big.-SevBay (indicated in bold) based on Bayesian inference for the partial rbcL and 18S rRNA genes. The total length of the alignment is 1444 characters. Posterior probabilities (PP) from BI are presented on the nodes in order. Only PP above 50 are shown. Strain numbers (if available) are indicated for all sequences.

Halamphora strains form two highly supported subclades (PP = 1). The first includes H. hyalina and H. pseudohyalina, the second includes all other strains of our dataset. In general, the branches within the clade are well supported (PP = 0.73 and higher). The relationships within the Amphora clade are well supported (PP 0.74–1) with the exception of branches of some species with weakly statistical support, namely A. thrumensis/A. commutata (PP 0.57), A. cf. immarginata/A. lativentralis (PP 0.56), A. helenensis (PP 0.63). Halamphora strains assigned to a single species (H. gibba, H. subacutiuscula, H. americana, H. subtropica) form maximally supported branches (PP = 1). The same is observed among strains of Amphora species (e.g., A. commutata, A. abludens, A. aliformis, A. lativentralis, A. copulata, etc.), indicating the absence of cryptic forms.

The new strain, H. kolbei DIA-A.big.-SevBay, is placed within the Halamphora clade and forms a separate branch together with the strain Amphora bigibba PMFTBO124 with maximum statistical support (PP = 1). The latter strain was isolated from the surface of the sea turtle (Caretta caretta) shells from the Adriatic Sea (Crotaria). Probably, this strain is misidentified and belongs to the genus Halamphora. Only SEM images of the internal valve side of the strain are publicly available [69], which does not allow for reliable identification. The resulting Maximum Likelihood tree presented in the cited work shows that strain PMFTBO124 (as in our analysis) is placed in a clade containing representatives of Halamphora, confirming its affiliation with this genus.

3.3. Biotechnological Characteristics

3.3.1. Microalgae Growth

The biotechnological characteristics of the H. kolbei strain from the Black Sea were assessed through batch cultivation. An intensive culture with a maximum density of 1.44 ± 0.02 g DW·L⁻¹ was used to study biochemical characteristics. The culture reached this maximum density on day 10 (Figure 4).

Figure 4. Growth kinetics of Halamphora kolbei strain in an intensive batch culture. Numbers indicate growth phases: 1—lag phase, 2—exponential phase, 3—deceleration phase, 4—stationary phase.

The growth curve of H. kolbei in 5F medium shows an exponential phase with a linear stage of growth. On the first day of cultivation, the lag phase (1) transitioned into the exponential growth phase, during which the culture increased, followed by a straight line, indicating linear growth (2). The highest levels of biomass productivity (0.20 g·L⁻¹·day⁻¹) were seen during the first six days of the experiment. On the seventh day, the culture growth rate decreased (3) from 0.48 day⁻¹ to 0.06 day⁻¹, and the culture entered the stationary phase. The stationary phase was observed from the 10th day of the experiment (4). The specific growth rate is a critical parameter for selecting the production strain, as a high growth rate enables high productivity, reduces contamination risk, and shortens the cultivation time.

3.3.2. Biochemical Composition of the Biomass

A biochemical analysis of H. kolbei biomass at different growth phases under batch cultivation was carried out to assess the biotechnological potential (Table 1).

Table 1. Biochemical composition of H. kolbei biomass during batch cultivation.

Growth Phases	Lipids, %	Proteins, %	Carbohydrates, %	Ash, %	PUFA, mg·g−1	Fx, mg·g−1
Exponential Culture density B = 1.2 g∙L−1	12 ± 0.1	40 ± 0.4	17.0 ± 0.2	33 ± 0.3	—	7 ± 0.4
Stationary Culture density B = 1.44 g∙L−1	22 ± 0.1	35± 0.4	8 ± 0.2	37 ± 0.3	42.26 ± 0.5	17 ± 0.4

In H. kolbei biomass, the mass fraction of total proteins reached a maximum (40%) during the exponential growth phase. In the stationary phase, under conditions of nutrient depletion, protein synthesis slowed down and its content decreased to 35%. Protein productivity during exponential growth was 6.7 mg·L⁻¹·day⁻¹, and decreased to 2.7 mg·L⁻¹·day⁻¹ in the stationary phase. A similar trend was observed for the carbohydrate content, with a maximum mass fraction (17%) recorded during active growth and a decrease to 8% during the transition to the stationary phase. Carbohydrate productivity during exponential growth was 2.8 mg·L⁻¹·day⁻¹, and decreased to 0.6 mg·L⁻¹ day⁻¹ in the stationary phase. The lipid content in H. kolbei biomass doubled to 22% when the culture transitioned to the stationary phase, compared to 12% in the exponential phase. Lipid productivity over the entire cultivation period was 22 mg·L⁻¹·day⁻¹. The ash content in the H. kolbei culture averaged 33–37% of dry weight across different growth phases. This indicates that diatom biomass has a relatively low calorific value of approximately 3 kcal·g⁻¹ dry weight, as organic matter accounts for 62–67% of the dry weight.

3.3.3. Fatty Acid Composition

Fatty acids with chain lengths ranging from 14 to 22 carbon atoms were identified in the biomass of the studied H. kolbei strain (Table 2).

Table 2. The fatty acid composition and productivity of the H. kolbei strain during batch mode cultivation.

Fatty Acid	The Percentage Share of the Total FA, %	Concentration, mg·g−1 Dry Weight	Productivity, mg·L−1·day−1
Saturated FA
Myristic (C14:0)	5.21 ± 0.06	3.99 ± 0.04	0.38
Pentadecanoic (C15:0)	0.54 ± 0.01	0.41 ± 0.05	0.04
Palmitic (C16:0)	12.02 ± 0.56	9.21 ± 0.32	0.88
Stearic (C18:0)	0.99 ± 0.01	0.76 ± 0.03	0.07
Total SFA	18.76	14.37	1.37
Monounsaturated FA
Palmitoleic (C16:1ω7)	19.78 ± 0.55	15.60 ± 0.53	1.49
Elaidic (E) C18:1 (ω9)	0.77 ± 0.01	0.59 ± 0.01	0.06
Oleic (C18:1ω9)	4.38 ± 0.04	3.36 ± 0.06	0.32
Total MUFA	24.93	19.55	1.87
Polyunsaturated FA
Hexadecatetraenoic (C16:4ω1)	0.36 ± 0.01	0.28 ± 0.01	0.03
Hexadecatrienoic (C16:3ω4)	3.33 ± 0.03	2.55 ± 0.21	0.24
γ-linolenic (C18:3ω6)	3.71 ± 0.04	2.84 ± 0.05	0.27
Linoleic (C18:2ω6)	1.74 ± 0.05	1.33 ± 0.03	0.13
Arachidonic (C20:4ω6)	1.76 ± 0.1	1.35 ± 0.02	0.13
Eicosotrienoic (C20:3ω6)	0.28 ± 0.01	0.21 ± 0.01	0.02
Eicosopentaenoic (C20:5ω3)	37.77 ± 0.42	28.95 ± 0.25	2.77
Docosahexaenoic (C22:6ω3)	7.33 ± 0.07	5.62 ± 0.05	0.54
Total PUFA	56.28	43.13	4.13
MUFA + PUFA	81.21	62.68	6.00
SFA + MUFA + PUFA	99.97	77.05	7.37
ω6/ω3	0.170	0.17

Notes: SFA—saturated fatty acids, MUFA—monounsaturated fatty acids, PUFA—polyunsaturated fatty acids.

The H. kolbei strain showed an interesting content of essential long-chain PUFAs, with ω-3 FAs being predominant, followed by ω-6 and ω-9 FAs. Among monoenoic fatty acids, palmitoleic acid (C16:1ω7) accounted for the highest proportion at 19.78% of total fatty acids. Among polyenoic fatty acids, eicosapentaenoic acid (C20:5ω3) was the most abundant at 37.77%, followed by docosahexaenoic acid (C22:6ω3) at 7.33%. This suggests that this species could be a valuable source of the rare and valuable eicosapentaenoic acid (EPA). The predominant saturated fatty acid was palmitic acid (C16:0, 12.02%). Overall, polyunsaturated fatty acids (PUFAs) predominated in the studied strain (56.28%), with an ω6/ω3 ratio of 0.17. This ratio indicates that our strain is rich in omega-3 fatty acids. The productivities over the entire cultivation period were 1.37 mg·L⁻¹ day⁻¹ for SFA; 1.87 mg·L⁻¹ day⁻¹ for MUFA; and 4.13 mg·L⁻¹ day⁻¹ for PUFA.

3.3.4. Fucoxanthin

Active accumulation of Fx was observed during the deceleration and stationary growth phases. At the beginning of the stationary phase of growth (day 10), the Fx content was 10.0 ± 0.4 mg·g⁻¹ dry weight (culture density 14.4 mg·L⁻¹). By day fourteenth, the Fx content reached 17.0 ± 0.4 mg·g⁻¹ dry weight (culture density 24.99 mg·L⁻¹). Thus, the Fx productivity was 2.64 mg·L⁻¹·day⁻¹ during the stationary phase and 1.78 mg·L⁻¹day⁻¹ over the entire batch cultivation period.

4. Discussion

4.1. Comparative Analysis of Morphological, Phylogenetic and Literature Data on H. kolbei and Closely Related Species

Morphological and phylogenetic analysis showed that our strain DIA-A.big.-SevBay from the Black Sea belongs to the genus Halamphora. The genus Halamphora is known for the following morphological features: a single, H-shaped chloroplast appressed to the ventral girdle and constricted in the middle; a girdle consisting of numerous bands comprising open copulae with one or two rows of round or transversely elongated poroids [1]. Images obtained with LM (Figure 1) and SEM (Figure 2) confirm the results of molecular-genetic analysis (Figure 3), allowing us to identify the studied strain as belonging to the genus Halamphora.

Since our strain exhibits a strong morphological affinity with H. kolbei and species identified as A. bigibba [34,39], we compared their morphological features and reviewed the literature on similar species (Table 3). Despite their bigibboid frustule shape being a common feature of these two species, A. bigibba has larger cells than H. kolbei, and the latter has more dorsal striae per 10 µm.

Table 3. Comparison of morphological features of H. kolbei and closely related species.

Taxon	Frustule Shape	Valve Apices	Length, µm	Width, µm	Dorsal Striae in 10 µm	Distribution	References
Halamphora kolbei strain (DIA-A.big.-SevBay)	elliptical, constricted in the middle	broadly rostrate	12.5–15	3.5–8.5 in valve view/till 12 in gridle view	24–31	The Black Sea	This study
Amphora bigibba from natural populations (most likely H. kolbei)	elliptical, constricted in the middle	broadly rostrate	8.6–12	2.8 (in valve view)	32	The Crimean coast of the Black Sea	[34]
A. bigibba from natural populations (most likely H. kolbei)	elliptical, constricted in the middle	broadly rostrate	11.5–12	3 (in valve view) 11.5 (in gridle view)	26–32	The Israeli sector of the Mediterranean Sea	[39]
A. bigibba (most likely H. kolbei)	elliptical, constricted in the middle	rostrate	13	3.5	30	Culture from the Adriatic Sea. The Croatian National Diatom Collection	[69]
A. bigibba	elliptical, constricted in the middle	broadly rostrate	14–36	10–15	15–20	Sevastopol and Novorossiysk bays of the Black Sea	[31]
	elliptical, constricted in the middle	broadly rostrate	15–25	4–6 (in valve view)	18–22	The Black Sea coasts of the Crimea and the Caucasus	[32]
	elliptical, constricted in the middle	rostrate	18–22	4–5	19–21	The Gulf of California	[37]
	lanceolate with strong constriction in the middle of the dorsal side	rostrate	12–20	4–5	19–20	Micronesia	[68]
	lanceolate with strong constriction in the middle of the dorsal side	broadly rostrate	28–33	valve width—5, frustule width—13–14	22–23	Kuwait	[87]
	lanceolate with strong constriction in the middle	protracted, rostrated	52.5	10 in valve view	17	Volo Bay of the Aegean Sea	[21]
Halamphora kolbei (Aleem) Álvarez-Blanco et Blanco 2014	elliptical, constricted in the middle	broadly rostrate	15	7.5 in gridle view	–
	semi-lanceolate	rostrate to slightly capitate	12–18	2–4	21–24	The Mediterranean Sea	[22]
	elliptical, constricted in the middle	protracted	11.7–13.3	2.7–3.0	26	The Eastern Mediterranean	[23]
	elliptical, constricted in the middle	protracted, rostrate	13.8–17	3.16–3.40	22	The Turkish coast of the Black Sea	[29]
	elliptical, constricted in the middle	rostrate	15–18	3.2–3.9	23–25	Florida Bay	[25]
	elliptical, constricted in the middle	rostrate	10.5–15.5	2.3–2.4	24–27	The Bulgarian coast of the Black Sea	[30]
Halamphora capitata (R.Hagelstein) Álvarez-Blanco et Blanco 2014	lanceolate with strong constriction in the middle of the dorsal side	protracted	20	4.5	20–22	The Mediterranean Sea	[23]
	elliptical, constricted in the middle	protracted	17	4	24	The Gulf of California	[37]
	lanceolate with strong constriction in the middle of the dorsal side	protracted, rostrate	22.7	4.48	21	The Turkish coast of the Black Sea	[29]
Halamphora angularis (Gregory) Levkov 2009	semi-lanceolate slightly constricted in the middle	subprotracted, subcapitate and ventrally bent	40–65	10–13 (in widest part of the valve)	11–12	The North Sea, The Baltic Sea	[1]

As shown in Table 3, the strain in this study and the species previously found in natural populations in the Black Sea range in length from 8.6 to 12 µm. However, according to [31,32], A. bigibba in the Black Sea is significantly longer, ranging from 14 to 36 µm. In [31], the illustration (Table VIII, Figure 19) representing A. bigibba shows numerous smooth girdle bands without poroids. The image of the species A. bigibba Grunow 1875 presented by the author matches the description in [63] (pl. 25: Figures 74 and 75). Therefore, we suggest that this species was previously misidentified due to a lack of opportunities for detailed morphological and genetic identification of specimens from natural populations [34]. The same is true for our work in the Israeli sector of the Mediterranean Sea [39], where we most likely observed the species H. kolbei. It cannot be ruled out that the same error occurred for the strain A. bigibba PMFTBO124 from the Adriatic Sea [69]. The phylogenetic tree we constructed (Figure 3) shows that this strain is similar to ours and belongs to the Halamphora clade. According to the image in [69], the strain dimensions are 13 × 3.5 µm, which is characteristic of H. kolbei.

A clear example for comparing H. kolbei and A. bigibba in terms of size and striae is the work of N. Foged [21], in which both species were found in the same water area of the Volo Bay (Aegean Sea) (Figure 5). The figure clearly shows that the difference in size between the species is more than twofold. A. bigibba has large striae that are clearly visible under a light microscope, while those of H. kolbei are practically indistinguishable.

Figure 5. Images of diatoms Amphora bigibba (12) and Halamphora kolbei (13). Scale bar—10 µm. Source: [21] (p. 63 Plate XI). Reproduced with permission from the copyright holder, Schweizerbart Science Publishers.

It should be noted that A. bigibba is characterised by the presence of two rows of areolae oriented along the apical axis along the dorsal side of the sternum, contrasting with the areolae on the rest of the dorsal surface, as noted by [68]. Such areolae are present in a single row for H. kolbei (indicated by arrows in Figure 2g,h).

We have now expanded the description of H. kolbei with information on the number of ventral striae in 10 μm; however, their presence was not previously indicated for this species. For example, I. Alvarez-Blanco and S. Blanco, [23], who transferred Amphora kolbei to the genus Halamphora, did not note ventral striae in H. kolbei under LM [23] (pl. 34, Figures 7–9). However, in [30] it was suggested that the species Amphora sp. 4 [23] (pl. 81, Figure 2), which looks identical to the one described by the authors as H. kolbei [30] and to our strain. In [30], H. kolbei was reported on the Bulgarian coast of the Black Sea and noted that ventral striae are present in the Black Sea population. The authors state that ventral striae may be weakly visible under LM, but depending on the valve position, they may not be visible at all. Additionally, similarities for Amphora sp. 4 [23] are visible on the dorsal side of the sternum, where there is a row of areolae oriented along the apical axis (Figure 6), just as in our strain.

Figure 6. Image of diatom Amphora sp. 4. Scale bar—2 µm. Source: [23] (Pl. 81, Figure 2). Reproduced with permission from the copyright holder, Schweizerbart Science Publishers.

In general, there are relatively few diatoms of the genera Amphora and Halamphora that have such a unique bigibboid shape; most of them are listed in Table 3. In addition to the above-mentioned A. bigibba, which is most similar to H. kolbei, we also list the species H. capitata and H. angularis. H. capitata is described as being more constricted in the middle, with ends that are more retracted and capitate than those of H. kolbei [23,29,37]. Previously, this species was considered a variety of A. bigibba and was identified as a separate species by the same authors as H. kolbei [23]. Accordingly, like A. bigibba, H. capitata is larger and has fewer striae per 10 μm (Table 3) than H. kolbei. H. angularis is the largest of all the species compared and also has very broad striae, the number of which per 10 μm is approximately half that of H. kolbei [1].

Thus, based on phylogenetic analysis of the rbcL and V4 18S rDNA regions and morphological characteristics, our study strain belongs to H. kolbei.

4.2. H. kolbei as a Promising Species for Biotechnology

In our study, the maximum productivity of the Black Sea strain H. kolbei was 0.20 g DW·L⁻¹·day⁻¹ (R² = 0.98). It is important to note that data in the literature show that similar species of the genera Amphora and Halamphora have significantly lower productivity (DW): 0.043 g·L⁻¹ day⁻¹ for Amphora copulata, 0.069–0.1 g·L⁻¹·day⁻¹ for A. subtropica, and 0.071 g·L⁻¹·day⁻¹ for H. coffeiformis [55,88,89]. Thus, H. kolbei cultures are characterised by high productivity. An important biotechnological characteristic of H. kolbei is its ability to settle quickly to the bottom of a photobioreactor, which can significantly reduce biomass production costs.

The genus Halamphora is known as a promising source of protein. Under optimal conditions with high nitrogen concentrations in the culture medium, the protein content in the biomass of diatoms from this genus can reach 30–45% of dry weight, and in some cases even higher [90,91]. In the exponential growth phase, the protein content in the biomass of H. kolbei DIA-A.big-SevBay reached its maximum (40%), as during the period of active growth, cells intensively synthesise structural and enzymatic proteins.

During batch cultivation of H. kolbei, the total carbohydrate content in the biomass showed wide variation depending on the growth phase, ranging from 8 to 17%. This is consistent with literature data indicating that the total carbohydrate content in Halamphora biomass can vary considerably, though under standard conditions it typically ranges from 10 to 28% of dry weight [90,91]. For example, at the onset of the stationary growth phase, H. coffeiformis was reported to contain 28 g of carbohydrates per 100 g of dry weight, equivalent to 28% [91]. The observed decrease in carbohydrate content in H. kolbei biomass during the stationary phase may be associated with a reduction in the rate of the dark reactions of photosynthesis. This slowdown likely results from limited ATP biosynthesis due to nutrient deprivation in the medium. For instance, phosphorus deficiency slows down ATP biosynthesis, leading to a decreased rate of carbohydrate synthesis in the Calvin cycle [92].

It is interesting to note that the ash content of H. kolbei reaches 33–37% of dry weight, which is consistent with the literature data obtained for H. coffeiformis, in which the ash content reached 38% DW [60]. In most planktonic diatoms, the ash content is 10–20% DW, while in benthic diatoms, it reaches 30–60% DW [90,93,94,95]. The higher ash content in benthic diatoms may be related to their denser frustules with a higher silica content.

The Fx content in biomass of the studied strain (17 mg∙g⁻¹ dry weight) is in the average range characterising the genera Halamphora and Amphora. For example, in Halamphora coffeiformis biomass, Fx content reaches 38 mg∙g⁻¹ dry weight [96]. Halamphora subturgida is a highly productive strain in terms of Fx production, reaching a peak concentration of 25.70 ± 0.03 µg·mL⁻¹ under nitrogen-replete conditions (1.764 µM NO₃⁻) [54]. A less productive strain is Amphora subtropica, which yields 15 mg of Fx per g of dry biomass [97].

The growing demand for energy, the decline in fuel reserves and the increase in CO₂ and greenhouse gas emissions have all accelerated the transition to renewable and unconventional energy sources [98]. A critical aspect of biodiesel production is the yield of biodiesel per unit area, which reflects the efficiency and viability of different feedstocks. Microalgae emerge as a frontrunner in terms of biodiesel yield, offering a substantially higher output in comparison to conventional oil crops. The biodiesel yield from microalgae is reported to be approximately 20,020 gallons per acre per year [99]. This remarkable yield underscores the exceptional potential of microalgae as a feedstock for bio-diesel production, reflecting their rapid growth rates, high lipid content, and ability to thrive in diverse environments [100]. The significance of strain selection arises from the need for high biomass productivity and energetic compound content, and the essential process characteristics. Researchers meticulously select strains with rapid growth rates, efficient substrate utilisation, and high biofuel productivity, possessing metabolic pathways conducive to desired bioenergy compound synthesis, ideal for large-scale biofuel production [101]. Efficient feedstock utilisation is another crucial benefit of strain selection. Moreover, certain microbial strains possess the capability to capture and assimilate carbon dioxide during their growth, thereby functioning as carbon sinks [102]. Microalgal biofuels solve the problem of the global energy crisis, noting their capacity to sequester 1.3 kg of CO₂ to produce 1 kg of biomass [103]. Diatoms are a well-known source of lipids. They have a high species richness, with many lipid-rich taxa. This variety leads to a diverse fatty acid composition, making them useful for biodiesel production and as food supplements, among other applications [98]. Diatoms are relatively easy to culture, and some strains can be subjected to environmental stress to increase lipid production [49,104,105,106]. Lipids are also the main component in diatom species relevant for aquafeeds [107]. The increase in lipid concentration to 22% during the stationary growth phase is likely due to a decrease in irradiance per unit biomass and growth limitation by phosphorus and silicon in the nutrient medium [108,109]. It is known that the lipid content of diatoms can reach 50%: Nitzschia sp.—47%, Cylindrotheca sp.—37% [98]. The basal lipid content in biomass of representatives of Halamphora or Amphora corresponds with the literature [110]. However, under nitrogen starvation conditions, the lipid content of certain species can reach up to 45% of dry weight [55,110]. Under our cultivation conditions, the nitrogen content in the nutrient medium was sufficient even in the stationary growth phase, so the lipid content did not exceed 22% of the dry weight. In the study [103], the cost breakdown of microalgal biofuel production was evaluated: 77% of expenses were for cultivation, 12% to harvesting and 7.9% to lipid extraction, while emphasising the potential for simultaneous biomass production in wastewater to reduce economic and environmental costs. Therefore, the development of algal biofuel is expected to be regionally focused [111]. This means it will be essential to cultivate strains that are well-adapted to diverse environmental conditions and inhabit local waters.

The fatty acid composition of the H. kolbei strain was determined, revealing certain characteristics. Our findings on the FA composition under the specified culture conditions are consistent with earlier studies on the genus Halamphora [59,60,112]. In total, 15 different fatty acids were detected. There are studies where 46 different fatty acids were detected for Halamphora cf. salinicola [49]. For our H. kolbei strain, the highest content among the monoenoic fatty acids was found for palmitoleic acid (C16:1ω7, accounting for 19.78% of total fatty acids), and among the polyenoic acids for eicosapentaenoic acid (C20:5ω3, accounting for 37.77%) and docosahexaenoic acid (C22:6ω3, accounting for 7.33%). For example, the fatty acid profile of thermophilic Halamphora spp. strains showed a significant content of two essential fatty acids: eicosapentaenoic acid (EPA, 20:5 ω3) and arachidonic acid (ARA, 20:4ω6). These fatty acids reached approximately 15% and 21%, respectively, of the total fatty acids at 20 °C and 30 °C [59]. The eicosapentaenoic acid content of most diatoms is typically 15–25% [113], which is considerably lower than in our strain. However, the arachidonic acid content in H. kolbei is low, whereas in some diatoms it can reach 10–15% [88]. H. kolbei has a higher palmitic acid content than many other diatoms, which may indicate a high level of protection against photooxidative stress [88].

Thus, the H. kolbei strain we studied has a superior fatty acid composition compared to similar species. This fact is of great practical importance since eicosapentaenoic acid helps prevent cardiovascular disease [114]. Docosahexaenoic acid (DHA) plays an important role in brain function, vision, and the cardiovascular and immune systems, and reduces the risk of neurodegenerative diseases (Alzheimer’s, Parkinson’s) [115]. Docosahexaenoic acid (DHA) is a major component of the photoreceptors in the retina of the eye, supporting visual function. It suppresses chronic inflammation by inhibiting pro-inflammatory cytokines (IL-6 and TNF-α) and stimulating the production of anti-inflammatory mediators. It also modulates the immune response by enhancing the activity of T-regulatory cells [115]. Consuming 3–4 g per day of long-chain ω-3 PUFAs has been shown to reduce hypertriglyceridaemia, thrombogenesis, and inflammatory and immune processes [116]. It should also be emphasised that ω-3 from diatoms is absorbed 20–30% better than ω-3 from plant sources [117,118]. As a result, H. kolbei biomass, along with products that have a low ω6/ω3 PUFA ratio (ω6/ω3 ≤ 1), can be used to prevent diseases such as cardiac arrhythmia, stroke, and myocardial infarction, as well as atherosclerotic changes in blood vessels [116]. It is also a promising source of eicosapentaenoic acid (EPA) for pharmaceutical and dietary supplement production because its bioavailability is higher than that of fish oil. Taking into account essential FAs, the level of ARA and EPA found in H. kolbei was even higher than that reported for the fish oils used for aquafeed [107,119]. Therefore, H. kolbei biomass can be used as a feed supplement in aquaculture, as its high C14-C16 acid content makes it ideal for fish larvae. Numerous studies on species of the genus Halamphora confirm these results [59,60,91,96].

5. Conclusions

A comparative analysis of morphological and phylogenetic data allowed us to identify the isolated Black Sea strain DIA-A.big.-SevBay as Halamphora kolbei. The description of the species is supplemented with new data on such morphological characteristics as a description of the girdle part of the frustula, including the number of poroids in 10 μm, and ventral striae. The biochemical composition of this strain’s biomass was found to have a high content of eicosapentaenoic acid (37.77%) and an increased content of palmitoleic acid (19.78%). Thus, studied strain is a valuable biological resource, important for the creation of pharmaceuticals and biologically active supplements. It can also be used as a feed supplement in aquaculture and as a raw material for biodiesel due to its high content of C14–C16 fatty acids.

The resulting phylogenetic sequence will be added to the GenBank, and additional morphological keys will allow scientists around the world to study this species as a promising biological resource. This scientific research is extremely important in terms of evolution and taxonomy. It contributes to our understanding of microbial ecology and the conservation of biodiversity.