Soil samples were amassed from three different geographic marine locations adjacent to the Mediterranean Sea. These locations are from Alexandria, Baltim, and Damietta cities (Fig. 1). After removing a 2 cm from the surface of the soil with a sterilized spatula, each sample was taken with a sterilized Peterson grab sampler, preserved in a sterilized polythene bag, and labeled appropriately. The collected samples were preserved in the dark at 4 °C until further analyses.

Figure 1

The collected soil samples were serially diluted with sterilized seawater to get 10^-1 to 10^-6 CFU (Colony Forming Units) per one-gram soil on plates containing nutrient agar growth media that contains 5 g NaCl, 2 g beef extract, 1 g yeast extract, 5 g peptone, and 20 g agar dissolved in distilled water for a total volume of 1 L (Downes and Ito, 2001). After a two days incubation period at 30 °C, pure cultures of different isolates were acquired according to the colony morphological characters. Later, pure strains were saved. For storage purposes, the isolates were preserved in 50 % glycerol and 50 % nutrient broth at - 20°C (Downes and Ito, 2001; Oliveira et al., 2021). In either nutrient agar or nutrient broth, the seawater was used in place of the distilled water to maintain the proper growth of marine isolates.

Potential isolates with the competence of producing fatty acids were selected according to two assays. Initially, the potency of each strain to survive over different H₂O₂ concentrations (0.01, 0.5, 1.0 % prepared from 30 % H₂O₂ stock solution) was evaluated by H₂O₂-plate assay (Tilay and Annapure, 2012). Moreover, the ability of each H₂O₂-resistant isolate to reduce TTC into TPF was tested by spectrophotometer at 485 nm, according to Ryan et al. (2010). Only candidates that showed resistance against H₂O₂ and reduced TTC into TPF were selected for GC-MS analysis. An example of the result of these two tests is shown in Fig. 2 for the bacterial isolate #2.

Figure 2

Bacterial isolates that can produce fatty acids were grown on 50 mL seawater-based nutrient broth (SNB) cultures for 24 h, and the pellets were collected by centrifugation at 4,000 g for 15 m. Afterward, the pellets were rinsed and prepared according to Rogers and Burns (1994). Total lipids were extracted from the collected pellets according to Folch et al. (1957). Finally, samples were subjected to saponification and methylation to generate fatty acid methyl esters (FAME). Hereafter, the FAME were extracted and then 1 mL was injected into the gas chromatography device H.P. (Hewlett Packard) 6890N Gas Chromatograph connected to Agilent 5973 Mass Spectrometer at 70 eV (m/z 50 - 550); source at 230 °C and quadruple at 150 °C) in the EI mode with an HP-5ms capillary column (30 m 0.25 mm i.e., 0.25 mm film thickness; J and W Scientific, USA). The carrier gas, helium, was maintained at a flow rate of 1.0 mL/min. The inlet temperature was maintained at 300 °C and the oven was programmed for 2 min at 150 °C, then increased to 300 °C at 4 °C/min., and maintained for 20 min at 300 °C (Watanabe et al., 1996). Based on the interpretation of mass spectrometric fragmentation, confirmed by comparing our compounds’ retention times and fragmentation patterns with the Supelco 37-component FMAE standard mixture. The microbial candidate with the peak of total fatty acid production was selected for molecular identification and further experiments.

The study selected the most promising candidate based on its proficiency in augmenting fatty acid production, as revealed by FAME analysis. The chosen candidate’s growth was then assessed using various growth media that were inoculated by 0.5 McFarland of the candidate isolate. The seawater-based nutrient broth medium (SNB) emerged as the most favorable, and its components were fine-tuned to enhance and optimize the growth conditions.

First, the optimal pH was determined from 6 - 6.5 - 7 - 7.5 - 8. Second, optimal temperature was evaluated from 15 - 40 C. Finally, various carbon and nitrogen sources were evaluated to optimize the candidate’s ability to reduce TTC into TPF. In the case of the carbon source, the yeast extract of the SNB was replaced by 2 % of different carbon sources such as lactose, mannitol, glucose, fructose, sucrose, starch, and dextrose. For the nitrogen source, peptone and beef extract were replaced by four different nitrogen sources, at 2 % of either potassium nitrate, ammonium sulfate, ammonium chloride, or ammonium nitrate. Carbon or nitrogen surrogates were sterilized via a sterile Millipore filter (0.45 µm) before using them within the growth media (Wang et al., 2007). Natural wastes like sugarcane and banana peels also replaced both the carbon and nitrogen sources. All treatments were incubated for 48 h at the optimum temperature and the optimum pH that was noted from the previous results. Later, all optimal conditions were employed to collect the culture biomass of the selected candidate, and FAME profiling was estimated under the best growth conditions as previously explained.

The 16S rDNA sequencing was performed at the Macrogen Inc., Seoul, South Korea. In brief, DNA extraction and purification were performed via the DNeasy Mini Kit (QIAGEN Inc., Hilden, Germany), as described by the supplier’s protocol. Purified DNA was considered pure when the OD₂₆₀/OD₂₈₀ ratio was around 1.8, and the OD₂₆₀/OD₂₃₀ ratio was about 2.0. Accordingly, the polymerase chain reaction (PCR) was performed with 27F/1492R primers according to Zayed et al. (2022). The PCR thermal gradient program was a single cycle of 95 °C for 5 min, followed by 30 cycles at 95 °C for 30 s, 55 °C for 120 s, and 68 °C for 90 s, followed by a final extension at 68 °C for 10 min, and the reaction was held at 4 °C. Afterwards, the product was purified and sequenced using Sanger (dideoxy) technology via the sequencing kit BigDye (R) Terminator v3.1 Cycle Sequencing Kit with four different primers, which were 785F, 907R, 337F, and 1100R. Sequencing was performed using ABI PRISM 3730XL Analyzer (96 capillary type) at the Macrogen Inc., Seoul, South Korea. The sequences from ABI 3730 were assembled and the obtained contig was deposited in the GenBank of the National Center for Biotechnology Information (NCBI Resource Coordinators 2016) with accession number OM946589.

The sequence was used as a query in a blastn algorithm with a cutoff E-value 1e-11 to retrieve the authentic 16S rDNA homologs (Zhang et al., 2000; Zayed and Badawi, 2020). The retrieved sequences were aligned using MAFFT version 7 (Katoh et al., 2019). Maximum likelihood phylogenetic tree was inferred assuming the lowest BIC (Bayesian Information Criterion) score model which is F81+F+I (Nguyen et al., 2015; Kalyaanamoorthy et al., 2017; Hoang et al., 2018). The phylogenetic tree was visualized by using iTOL (Interactive Tree of Life) webserver (Letunic and Bork, 2016). The same phylogenetic analysis was repeated using the blast algorithm against bacterial type strains only and with the Maximum likelihood phylogenetic tree was inferred assuming the lowest BIC model which is HKY+F+I to reveal the closest type strain to MZ1.

Our target was then to identify which of the six candidates exhibits the highest efficiency in lipid production. The H₂O₂-assay plate, being a qualitative method, is insufficient to address this question. Additionally, TPF formation from TTC may lack the precision required to differentiate the lipid production capacities of efficient candidates. Therefore, gas chromatography was utilized to quantitatively assess the biosynthesized lipid quantities by each species under consistent pre-set conditions. FAME profiling was performed to reveal the bulk of intracellular and extracellular fatty acids compositions for all the six selected candidates (Figs. 4 and 5).

Figure 4

Figure 5

FAME profiles of the six bacterial candidates reveal that all of them are gram-positive due to the absence of liposaccharides and hydroxy fatty acids in their FAME profiles (Bisen et al., 2012). It has been noted that candidate II recorded the highest lipid accumulation (78.5 %). The production of caproic acid attained 46.1 %, followed by oleic acid 3.1 %, which indicates the efficiency of this candidate to manufacture caproic acid when grown on nutrient agar. Other candidates were also seen to accumulate lipids but with fewer amounts. For example, candidates XIII and XVI have a total lipid production of 25.5 % and 36.3 %, respectively, with a pronounced production of monounsaturated fatty acids (MUFAs) such as erucic acid that reached 5.6 %, and oleic acid that reached 2.9 %. Also, candidate V has a total lipid production of 25 % with efficiency in producing oleic acid that attained 2.9 %. Although candidates XI and XIV recorded the lowest total lipid production (18.6 % and 15.2 %, respectively), they efficiently produced oleic acid by 2.9 %. The creation of oleic acid was also maintained by candidate II (3.1 %) with efficiency higher than any other candidates. The results emphasize candidate II as the most potent for most fatty acids production, specifically caproic and oleic acids (Fig. 4).

Although this study used methods that look for PUFAs-producing bacterial species, it highlights a bacterial species that is most efficient in producing SFAs, mainly n-caproic acids. Substantially, the results of screening the fatty acids contents of six marine microbes showed salinity as an adaptive evolutionary regulator, which stabilizes the evolutionary trait of the MUFAs and SFAs accumulation in comparison with PUFAs, at least for these screened microbes (Figs. 4 and 5). Both MUFAs and SFAs are industrially valuable in enhancing fuel properties. Thus, finding a microbial species with an evolved trait of SFAs production will minimize the cost of biodiesel production, and will help improve its features.

The partial 16S rDNA revealed that candidate II is Bacillus subtilis and its 16S rDNA sequence was deposited on NCBI (accession number OM946589.1). The Blastn algorithm showed that its closest subspecies is Bacillus subtilis spizizenii TU-B-10, with no gaps and a 100 % identity, which covered the query sequence that includes all 16S rDNA hypervariable regions from V1 to V9 (Fig. 6). However, the phylogenetic relationship of MZ1 with bacterial type strains revealed its uniqueness and its close affinity with numerous halophilic Bacillus species such as B. swezeyi, B. haynessi and B. mojavensis, where they all clustered in a distinct clade (Fig. 7). The B. subtilis MZ1 is phylogenetically closely related to many other halotolerant/marine Bacilli species (Figs. 6 and 7). For examples, B. cabrialesii and B. tequilensis can tolerate up to 2 % and 5 % NaCl, respectively. Additionally, the closest subspecies to MZ1 strain, which is B. spizizenii can tolerate 7 % NaCl (de los Santos Villalobos et al., 2019). Also, B. swezeyi, B. haynessi and B. mojavensis that are clustered in a distinct clade with MZ1 (Fig. 7) can tolerate 12 %, 12 % and 10 % NaCl (w/v), respectively (Reimer et al., 2022).

Figure 6

Figure 7

Bacillus subtilis MZ1 had the highest levels of total lipid production (78.5 %) and the highest levels of production for each individual fatty acid when compared to the other candidates studied. Bacillus subtilis MZ1 was screened for its ability to grow on a variety of different media, and it was found to prefer nutrient broth, and seawater-based nutrient broth over the other growth media tested, particularly those based on natural wastes like sugarcane and banana peels (Fig. 8). The seawater-based nutrient broth was then chosen to adjust its components to maximise lipid production by B. subtilis MZ1 on it. First, the ability of lipid biosynthesis to reduce TTC into TPF was adjusted utilising diverse carbon and nitrogen sources, as well as variable pH and temperature degrees. The optimal conditions for B. subtilis MZ1 to reduce TTC into TPF were a pH of 7 and a temperature of 37 °C (Fig. 9). However, giving diverse nitrogen sources had no obvious influence on TTC reduction, as demonstrated in Fig. 9. Surprisingly, glucose, rather than the other sources of carbon, shown a significant impact on enhancing TTC to TPF conversion (Fig. 9).

Figure 8

Figure 9

To test the effect of modifying the growth medium composition on lipid profile, the nutritional growth medium was employed with glucose as an extra carbon source and ammonium sulphate as another nitrogen source at pH 7 and 37 °C. Lipid production of B. subtilis MZ1 is mostly rich with caproic acid, when grown on SNB or caprylic acid, when glucose was used as an additional carbon source in the SNB growth medium. From the applied perspectives, n-caproic acids are vital in the biodiesel industry, as flavor compounds, and as antimicrobial agents (Zhu et al., 2017). Additionally, the MZ1 strain produces SFAs such as myristic, pentadecanoic, stearic, besides other MUFAs such as oleic, erucic; and other PUFAs such as cis-5,8,11,14,17-eicosapentaenoic, and cis-13,16-docosadienoic in considerable amounts. The results demonstrated that the modified growth medium efficiently changed the B. subtilis MZ1 metabolic pathways to create octanoic acid (C8:0, caprylic) instead of hexanoic acid (C6:0, caproic). This metabolic shift suggests that by managing the given feedstock, it is possible to improve carbon chain elongation and produce more valuable longer-chain carboxylic acids (Figs. 4 and 8), and as is shown by the gas chromatograms. We suppose that elongation is the most likely to happen in our case throughout the process of the reverse β-oxidation (Magdalena et al., 2020).