Fossil fuel-based plastics play a crucial role in modern lifestyle (Chae and An, 2018). Approximately 400 million tons of plastic were produced globally in 2018. More than 10 million metric tons of plastic ended up into the oceans during 2018 alone. An estimated 13 billion tons of plastic waste will reach the environment by 2050 in the absence of any improvements in current plastic waste management practices (Shams et al., 2021). Synthetic plastics have significant negative impacts on ecosystems, biota, environment, economy, and human health (Sharma and Chatterjee, 2017). As a result, there is a growing demand for bioplastics such as polyhydroxyalkanoates (PHA), polylactic acid, biopolyamide, biopolyethylene and biopolypropylene that provide new solutions in terms of life cycle and raw materials (Akinwumi et al., 2019). The conversion of biowastes to PHAs is considered as a possible eco-friendly alternative to petro-plastics and are currently gaining a lot of interest in the field of industrial applications and waste management (Kourmentza et al., 2017; Patel et al., 2021). Global bioplastic production has increased from 2.1 metric tons in 2020 to about 2.4 metric tons in 2022 (Dubey and Mishra, 2021) and the market opportunity for PHA bioplastic could reach US$100 million by 2024 (Surendran et al., 2020).

PHA are linear polyesters that consist of hydroxy fatty acids and are synthesized by a wide range of different Gram-positive and Gram-negative bacteria, as well as in some haloarchaeal species (Bugnicourt et al., 2014; Han et al., 2015; Kalia et al., 2021). Among the PHA producing microorganisms, the most-studied were Pseudomonas oleovorans, P. aeruginosa and P. putida, Bacillus megaterium, B. cereus, Ralstonia eutropha, and Haloferax mediterranei (Poli et al., 2011; Kalia et al., 2021). PHAs are usually produced when the microorganismes are cultured in the presence of excess carbon sources and limited nitrogen, phosphorus, sulfur, or oxygen (Surendran et al., 2020). They are stored as reserve foods as either inclusion bodies or within calcium (Ca) or polyphosphate complexes. In a few cases, PHA can accumulate to 90% of the total cell dry weight (CDW), which helps the microorganism survive under environmental stress conditions (Kalia et al., 2021). PHA can be classified in three main categories: short-chain-length PHA (scl-PHA) consists of 3-5 carbons atoms [3-hydroxybutyrate (3HB), and 3-hydroxyvalerate (3HV)]; medium-chain-length PHA (mcl-PHA) that consists of 6–14 carbons atoms [3-hydroxyhexanoate (3HHx), 3-hydroxyheptanoate (3HHp), and 3-hydroxytetradecanoate (3HTD)…] and also mixtures of both scl-PHA and mcl-PHA (Leong et al., 2014; Tan et al., 2014). More than 150 different monomers have been identified in PHA chains under various fermentation conditions, resulting in PHA with different characteristics (Lee et al., 2019). Due to their properties similar to those of petrochemical thermoplastics and combined properties of biodegradability and biocompatibility, PHA have gained considerable importance as promising candidates for biologically-derived plastics as well as biomaterials (Han et al., 2010, 2015, 2017; Kourmentza et al., 2017; Mitra et al., 2020).

Poly (3-hydroxybutyrate-co-3-hydroxyvalerate), PHBV, is a biopolyester with good mechanical properties and biodegradability, with broad applications in a wide variety of sectors (food and medical packaging, drug release and transport systems, hygiene products, etc.) due to its excellent properties such as absorption capacity, low cytotoxicity, piezoelectricity, and thermoplasticity compared to brittle PHB (Tan et al., 2014; Rivera-Briso and Serrano, 2018; Chen et al., 2019). As reported by Ibrahim et al. (2020), PHBV is a very promising copolymer that has great potential to replace conventional non-degradable polymers and has a great sustainability potential in circular economy development strategy. For most industrial uses of PHBV, the 3HV fraction should be at least in the range of 10–20% and usually chemical precursors such as propionate and valerate were added to obtain such 3HV fraction (Hermann-Krauss et al., 2013). Increasing 3HV content enhances the biodegradability and reduces the crystallinity and melting point of the bio-copolymer (Policastro et al., 2021). However, even though it owns environmental advantages and more suitable properties compared to other bioplastics, the main problem facing commercial production is the high production costs and low productivity rate. Therefore, the current challenge for researchers is the implementation of efficient and low cost PHBV production processes (Albuquerque and Malafaia, 2018; Patel et al., 2021 a and b; Policastro et al., 2021). Therefore, many approaches have been developed to minimize the production cost such as 1) the use of agricultural wastes and industrial by-products (e.g., fruit peels, bagasse, waste plant oils, crude glycerol, cheese whey, etc.) as inexpensive carbon sources with widespread availability and renewability (Kulkarni et al., 2015; Patel et al., 2015; de Paula et al., 2017; Ray et al., 2018:; Surendran et al., 2020), 2) the application of experimental designs and response surface methodology as statistical tools for the optimization of microbial biomass and PHA production using different cultivation modes (batch, fed-batch, and continuous fermentations) (El-malek et al., 2020; Singh et al., 2021), 3) the use of extremophilic bacteria as efficient PHA-producing strains (Cai et al., 2011; Yin et al., 2015; Kourmentza et al., 2017; Chen et al., 2018; Patel S. K. S. et al., 2021; Dubey and Mishra 2021). As an example, PHA production can be more environmentally friendly low-cost feed like using greenhouse gases (biogas) by thermophilic methanotrophs such as under non-sterile conditions (Patel S. K. S. et al., 2021).

Microbial halophiles are considered to be a promising cell factory for high-value biomolecules synthesis (extremolytes, extremozymes, biosurfactants, bioplastics, etc.) due to their unique characteristics of thriving under extremes of temperature, salinity, pH, and solvent conditions (Mitra et al., 2020). Extremely halotolerant species (up to 20% salinity) of the genus Halomonas, mostly associated with saline or hypersaline environments such as ocean waters and hypersaline lakes have shown PHA production performance under open and non-sterile conditions allowing the development of continuous bioprocesses without contamination (Kourmentza et al., 2017; Neifar et al., 2019; Athmika et al., 2021). The Chinese company Bluepha produces both PHB and PHBV from a halophilic Halomonas strain (Mitra et al., 2020). The genomes of PHA-producing extremophilic strains were studied via bioinformatics and genetic engineering tools for the expression of the PHA synthesis genes in non-producing strains exposing less restrictive growth and production conditions (Cai et al., 2011; Thomas et al., 2020). Policastro et al. (2021) reviewed the most relevant studies on PHBV production by mesophilic and extremophilic strains and the strategies used for costs reduction. Halophilic and halotolerant Halomonas strains such as Halomonas sp. TD-0, Halomonas sp. SF2003, Halomonas i4786, H. pacifica ASL10 and H. salifodiane ASL11 have been reported as promising candidates able to produce short chain length PHA up to 80 wt% of cell dry weight (CDW), using conventional carbon sources as well as carbonaceous by-products from food wastes (Elain et al., 2016; El-Malek et al., 2020; Thomas et al., 2020; Dubey and Mishra, 2021).

Thus, the present study investigates the ability of the halophilic bacterium Halomonas desertis G11, isolated from the largest salt-lake in Tunisian Sahara (Neifar et al., 2019), to directly produce PHBV from a low-cost and non-toxic substrate and without addition of PHV precursors. The study focused on the use of glycerol as a carbon substrate taking into account that it is one of the promising substrates for large-scale production of PHA due to the growing production of biodiesel as an alternative renewable energy source (Kumar et al., 2015; Ray et al., 2018; Kondaveeti et al., 2019). The culture conditions were optimized using experimental design and response surface methodologies. This work will also contribute to a better understanding of PHA metabolism and PHA synthase functionality of G11 strain by genomic and structural approaches.

Halophilic microorganisms especially Halomonas spp., have been attracted a lot of interest, recently due to their potential to produce exoenzymes, exopolysaccharides, osmolytes, biosurfactants, and bioplastics (Neifar et al., 2019). In this work, we report the capacity of the hydrocarbonoclastic, plant growth-promoting halophilic bacterium H. desertis G11 (Naili et al., 2018; Neifar et al., 2019; Riahi et al., 2020) to produce PHA. The ability of this strain to grow at NaCl concentrations up to 20% with an optimum at 5%–10% allowing an open and continuous fermentation process without contamination and to produce many compatible organic solutes such as osmoregulated periplasmic glucans, choline-glycine betaine, and ectoine hydroxyectoine make it a well-equipped bacterium to cope with the osmotic stress (Neifar et al., 2019).

In this study, the potential of halophile H. desertis G11 to produce PHA was firstly revealed using Nile bleu A staining plate assay (Ostle and Holt, 1982) under saline condition (5% NaCl). The bright pink fluorescence displayed by the bacterial colonies under UV light indicates intracellular accumulation of PHA by G11. Although the mechanism of selective PHA staining by Nile blue is not yet fully understood, it has been suggested that the dye diffuses through the bacterial membrane and labels cytoplasmic PHA granules by binding to lipid molecules in the surrounding layer. Thus, this method has been adapted for many years, for its affinity to bind to PHA granules and has therefore been accepted as a presumptive test for PHA storage in bacteria. This method has been applied to other strains such as Bacillus subtilis MANA18 (Al-Hamdani and Jaber, 2017) and Vibrio sp. BTTC26 (Raghul. 2012). The Halomonas PHA granules have a diameter range of 0.2–0.5 µM as reported by Muhammadi et al. (2015) and Obruca et al. (2020). The use of lipophilic dye staining and TEM and SEM microscopy techniques have been applied to detect PHA granules in other Halomonas species including Halomonas sp. SF2003, H. boliviensis LC1, Halomonas sp. TD01, H. pacifica ASL10 and H. salifodiane ASL11 and H. hydrothermalis (Quillaguaman et al., 2006; Fu et al., 2014; Thomas et al., 2019; El-malek et al., 2020; Obruca et al., 2020).

PHA production by H. desertis G11 was assayed in submerged cultures using glycerol as a substrate. Statistical analysis and model fitting of the optimization of PHA production was performed using BBD and RSM. The optimum PHA production (1.5 g/L; 68% of DCW) was obtained at 4 days while longer incubation time showed decreasing trend. Incubation period for PHA production by different organisms reported so far was longer than 24 h (Kulkarni et al., 2010). H. boliviensis required 33 h incubation period for highest PHA accumulation (Quillaguaman et al., 2006). Other species of Halomonas such as H. profoundus, H. elongate and Halomonas marina required 3 days for PHA production (Mothes et al., 2008; Simon-Colin et al., 2008; Biswas et al., 2009). In contrast, a longer incubation time of 120 h was required by the halophilic strain H. elongata to produce similar PHA content (50–60 wt%) (Mothes et al., 2008; Biswas et al., 2009).

Halophiles are of particular interest since they eliminate the need to maintain aseptic conditions and enable continuous production in saline (even sea) water and they serve as feasible genetic sources for industrially important compounds like biopolymers, osmolytes and extremozymes (Chen et al., 2018; Rodriguez-Perez et al., 2018). Several halophilic microorganisms demonstrated their ability to produce PHA. Most of them belong to the family Halomonadaceae (Haba et al., 2011). The genus Halomonas is known to accumulate scl-PHA (Table 2). Most Halomonas sp. has a NaCl requirement of 3%–15% for optimal growth and this concentration is sufficient enough to prevent microbial contamination. Halomonas species including H. marina, H. boliviensis, H. hydrothermalis, Halomonas sp. TD01, H. elongata DSM 2581, and H. campaniensis, isolated from different saline habitats in the word, have been reported as high polymer accumulators from different sugars, sodium acetate and butyric acid, and agricultural residues (Mothes et al., 2007; Van-Thuoc et al., 2008; Biswas et al., 2009; Quillaguamán et al., 2010; Tao et al., 2017; Cristea et al., 2018).

The use of glycerol as a substrate allows higher PHBV content with reduced cost, it does not consume phosphoenolpyruvate of the PTS transporter system compared with other sugars, more oxaloacetate and energy can be saved for PHBV production (Jincy et al., 2013; Tan et al., 2014). As crude glycerol contains high salt concentration, ranging from 3 to 7% w/v, the possibility of using halophilic bacteria could potentially provide an attractive production system for PHA (Van-Thuoc et al., 2015). Glycerol has previously been reported as an ideal carbon source for PHA production in some vibrios (Chien et al., 2007; Wei et al., 2011; Raghul. 2012), Ralstonia eutropha (Taidi et al., 1994). In addition, results with Bacillus megaterium demonstrated the valorization of glycerol for PHB production (Naranjo et al., 2013). They were able to produce 4.8 g/L of PHB using 2% purified glycerol under controlled conditions. Jincy et al. (2013) performed statistical optimization for PHB production (0.60 g/L) using 2% crude glycerol by Bacillus firmus NII 0830. Studies also revealed the ability of several Halomonas strains to accumulate PHA from glycerol among them, H. campisalis ATCC 700597T synthesizes PHB with a yield of 24.4%, H. meridiana NBRC 15608T which gives about 25% by weight of PHB. Similarly, Halomonas sp. KM-1 accumulates about 34% of PHB and 63.6% at 3% and 10% glycerol respectively (Kawata and Aiba, 2010). More recently, Dubey and Mishra (2021) reported a PHA content of 0.20 ± 0.02 g/100 ml in H. daqingensis and 0.21 ± 0.01 g/100 ml in H. ventosae when cultivated in 3% algal biodiesel waste residue, 5% NaCl supplementation at 35°C within 48 h of incubation. These result points to the future use of a cheap industrial by-product such as glycerol from biodiesel waste in the industrial production of PHA by the halophilic Halomonas species.

The chemical structures of the purified polymer were analyzed by FT-IR and ¹H and ¹³C NMR. The assigned signals in the corresponding spectra agree with earlier studies (Kemavongse et al., 2008; Shamala et al., 2009; Kulkarni et al., 2011; Getachew and Woldesenbet, 2016; El-malek et al., 2020), confirming the chemical structure of PHBV. Based on the peak areas of the CH₃ of HB unit and the CH₃ of HV unit, the mole fraction of 3HV was determined as 52.04 mol %. It has been reported that Halomonas species can accumulate PHB, PHBV and poly (3-hydroxybutyrate-co-4-hydroxybutyrate) (P3HB4HB) (Ren et al., 2018; Chen et al., 2019; Ye et al., 2020). The copolymer PHBV is of high commercial interest due to its favorable properties compared to PHB (Tian et al., 2001; Chen. 2009; Ferre-Guell and Winterburn, 2018). Most PHBV producers require co-substrates such as propionic acid or valeric acid. H. profundus produces PHBV with different molar fractions of 3HV in the presence of valeric acid and propionic acid (Simon-Colin et al., 2008). Similarly, a prerequisite to the formation of 3HV-CoA in E. coli is the intracellular presence of propionyl-CoA as a precursor (Srirangan et al., 2016). The supplementation of precursors due to the lack of intracellular propionyl-CoA as a precursor in the biotransformation systems is toxic which not only inhibit cell growth but also increase production cost but also makes PHBV production difficult to control and thus increases production cost (Steinbüchel and Lutke-Eversloh, 2003). CRISPR/Cas9 method was used to engineer the TCA cycle in H. bluephagenesis on its chromosome for production of PHBV from glucose as a sole carbon source (Yu et al., 2020). The recombinant Halomonas TD08 is able to produce PHBV consisting of 4–6 mol% 3HV, from various carbohydrates as the sole carbon source (Tan et al., 2014). The copolymer formation without addition of precursors or genetic engineering tools has been reported in a few microorganisms such as Bacillus cereus SPV, Rhodobacter rubber, Nocardia corallina, Cupriavidus necator SH-69, Agrobacterium sp. strain SH1 and GW-014, Rhodospirillum Rubum, and H. campisalis MCM B-1027 (Valappil et al., 2008; Kulkarni et al., 2010; Policastro et al., 2021). To our knowledge, the moderately haloalkalitolerant strain, H. campisalis isolated from alkaline and saline Lonar Lake (Buldhana India) was the only Halomonas strain reported to accumulate PHBV (10.4 mol% HV) using a simple carbon source without addition of any precursors (Kulkarni et al., 2010).

The metabolic pathways relevant to glycerol assimilation and PHBV biosynthesis was predicted by genome analysis of H. desertis G11. In the predicted pathways, glycerol can be metabolized to two essential precursors acetyl-CoA through glycolysis pathway and converted to the 3HB monomer and to propionyl-CoA via threonine biosynthetic pathway and converted to 3HV monomer. These precursors lead to PHBV formation by three crucial enzymes. The genes encoding these three key enzymes of PHBV biosynthesis (PhaA, PhaB, and PhaC) are frequently organized in a single operon, as in C. necator or Pseudomonas sp. (Babel et al., 2001; Luengo et al., 2003). This organization is absent in other bacteria such as Halomonas sp. SF 2003. Idem, the genes encoding the three key enzymes of PHA biosynthesis pathway are not clustered in an operon in G11 strain genome but are scattered all over the genome. It has been reported that the organization of the biosynthetic genes depends on the bacterial species and also on the class of PHA synthases.

Based on phylogenetic analyses performed in this study, PhaC1 present in the genome of H. desertis G11 belongs to class I. Similar results were found with Halomonas sp. TD01 (Cai et al., 2011), Halomonas sp. SF2003 (Thomas et al., 2020) and H. sp. O-1 (Ilham et al., 2014). Multiple alignments of PHA synthases of several Halomonas species allowed identifying different PhaC box consensus sequences (Thomas et al., 2020). Analysis of CI-PhaC amino acids sequences of different halophiles PHA-producing Halomonas strains showed similar lipase box-like sequence (Ser-X-Cys-X-Gly) (figX). From the database analysis the most common described pattern is Glycine-X-Cysteine-X-Glycine (G-X-C-X-G) which can be modified to (GS)-X-C-X-(GA)-(GA) (Nambu et al., 2019). The PhaC_Hd has a lipase box-like sequence, S-X-C-X-G located between the residue 371 and the residue 375.

Structural analysis by different web servers for protein modeling (Phyre2, iterative TASSER, Swiss-Model), generated only two reliable three-dimensional protein structure models for Pha C1 of H. desertis G11 (Figure 7). The abilities of both strains Chromobacterium sp. USM2 and C. necator to produce PHBV from different carbon sources, have been previously reported (Obruca et al., 2010; Aramvash et al., 2016; Chek et al., 2017). Obruca et al. (2010) investigated the production of PHBV by Cupriavidus necator H16 in fed-batch mode from waste rapeseed oil using propanol as a precursor of 3-hydroxyvalerate. Aramvash et al. (2016) studied the effects of using a combination of substrates (fructose, propanol, citric acid, acetic acid, propionic acid and beef extract) on the PHBV production. The results of batch PHBV optimization by RSM, indicated that fructose in combination with propanol and beef extract, showed better PHBV yield, as compared to that of propionic acid. Chek et al. (2017) published a high-resolution crystal structure of the catalytic domain of homodimer Class I PHA synthase from Chromobacterium sp. USM2. This enzyme produces poly (3-hydroxybutyrate-co-3-hydroxyhexanoate) copolymer and poly (3-hydroxybutyrate-co-3-hydroxyvalerate-co-3-hydroxyhexanoate) terpolymer, which are good materials for industry due to their softness and flexibility (Jia et al., 2000; Amara and Rehm 2003; Tian et al., 2005).

The production of PHA by the halophilic strain H. desertis G11 on glycerol-based medium under saline condition was successfully optimized through statistical approach (BBD and RSM) indicating that G11 is an efficient PHA producer (1.5 g/L; 68% of DCW) with a high potential for industry applications. The NMR and FTIR spectra showed the presence of required functional groups in biosynthesized PHBV based on chemical shifts and bands observed. Genomic analysis revealed the presence of the key genes involved in the biosynthesis and biodegradation of PHA including phaA, phaB, phaC, phaZ, phaP and phaR. Through multiple sequences alignments with its homologs, H. desertis PhaC1 can be classified as a class I PHA synthase. Threonine pathway seems to have a key role in PHV synthesis.

The promising potential to use crude glycerol as low-cost substrate, in addition to its capacity to grow in front of stressful conditions and to produce PHA copolymers, make H. desertis G11 a versatile strain with a high potential for biotechnological use. The future research will be focused on 1) scaling up of the optimized process in bioreactor (fed batch and continuous processes) using biodiesel waste as carbon substrate; 2) studying the thermomechanical and visco-elastic properties of produced PHA; 3) valorizing low-cost substrates for the production of PHA under non-sterile high-salting conditions, 4) studying the role of PHA synthase in the synthesis of PHBV by H. desertis G11 using X-ray crystallography and exploiting the phaC gene for genetic and enzyme engineering, and 5) developing suitable genetic tools for manipulating the H. desertis G11 as a platform for synthesizing polymers.