Introduction

The excessive usage of petroleum-based plastics for single-use items has led to their accumulation in the environment. As these plastics do not biodegrade, they accumulate in both terrestrial and marine environments. Which causes annually the death of numerous animals, as well as the contamination of soils and waterways with the chemicals they release, and the micro and nanoparticles they accumulate (Chang et al. 2022; Issac and Kandasubramanian 2021). Furthermore, the production of petrochemical plastics not only relies on non-renewable fossil fuels, but also results in significant energy consumption and greenhouse gas emissions. In fact, abandoned plastics in nature have been found to emit methane and ethylene, two potent greenhouse gases (Singh and Sharma 2016).

To address these challenges and explore alternative materials, researchers have developed biodegradable plastics made from biopolymers such as polyhydroxyalkanoates (PHA) (Alsafadi et al. 2020; Bhargava et al. 2020; Mannina et al. 2020; Naser et al. 2021; Park et al. 2021). These biopolymers are extracted from bacteria and have similar material properties to polypropylene (McAdam et al. 2020; Meng et al. 2021). Compared to petrochemical plastics, biopolymers offer several advantages, including being renewable, biobased, and biocompatible. The existence of PHA in bacteria has been known since the 1920s when Maurice Lemoige observed the formation of polyhydroxybutyrate (PHB) inside the bacterium Bacillus megaterium (Lemoigne 1926). PHA are synthesized by microorganisms that can activate a survival mechanism when they lack certain nutrients or when they detect an excess of carbon in their nutrient medium (Jung et al. 2019; Lee et al. 2021; Tarrahi et al. 2020). The physical and chemical properties of PHAs vary significantly based on their monomer composition, leading to a broad range of characteristics. Their hydrophobicity, melting point, glass transition temperature, and degree of crystallinity are all influenced by the type of monomers they contain (Sehgal and Gupta 2020; Naser et al. 2021). Moreover, PHAs display a diverse array of mechanical characteristics, ranging from hard and crystalline to elastic. Wild and recombinant strains of Ralstonia eutropha, formerly known as Alcaligenes eutrophus or Cupriavidus necator (Biglari et al. 2018), are best known for accumulating high levels of PHA. However, recent reports have drawn attention to the significant production of PHA by various bacterial species. One such species is Aneurinibacillus sp. H1, an innovative Gram-positive bacterium exhibiting moderate thermophilicity. This microorganism is a promising route for PHA biosynthesis, as it enables the monomer composition to be adapted to suit specific needs (Sedlacek et al. 2020).

Other noteworthy strains that have been studied over the past few years include Bacillus sp. (Kanjanachumpol et al. 2013), Pseudomonas sp. (Kanavaki et al. 2021), Burkholderia sp. (Keenan et al. 2006), Azotobacter sp. (Urtuvia et al. 2022), Halomonas sp. (Thomas et al. 2020), Haloferax sp. (Singh and Singh 2018), Acetobacter sp. (Chang et al. 2021) and recombinant E. coli (Ray et al. 2012). During nutritional deficiencies, various Bacillus species were found to produce up to 90% (w/w) PHA in dry cells (Anjum et al. 2016). Due to their genetic stability, these microorganisms have emerged as model organisms in industry and science (Biedendieck et al. 2007).

Bacillus species have the capability of producing PHA copolymers utilizing simple, affordable, and structurally unbound carbon sources, in addition to having a faster growth rate than other bacteria. Moreover, the isolates have the potential to secrete various hydrolytic enzymes that can be utilized to produce PHA at a low cost, using agro-industrial and other waste materials (Israni and Shivakumar 2013). However, the high production cost and the widespread availability of low-cost petrochemical-derived plastics have hindered the development of bioplastics for packaging applications. To make the process economically attractive, several studies have demonstrated that organic-rich agricultural, agro-industrial, and domestic wastes can be employed as new raw materials for PHA production (Sehgal and Gupta 2020).

Olive mill wastewater (OMW), a by-product of the olive oil industry, has the potential to be a low-cost substrate for PHA production. OMW is rich in easily digestible carbon sources, including carbohydrates, lipids, and volatile fatty acids, which serve as direct substrates for PHA production (Alsafadi and Al-Mashaqbeh 2017). Mediterranean countries, particularly Tunisia, generate large quantities of OMW annually, with an average production of 1 million tons of olive oil mill wastewater (Marks et al. 2020). This results in serious environmental concerns such as changes in soil microbial populations, threats to surface and groundwater sources, and air pollution from phenol and sulfur dioxide emissions (Albalasmeh and Mohawesh 2023).

The manufacture of PHAs from OMW currently involves multi-stage processes, which include pretreatment to remove polyphenols, acidogenic fermentation to generate volatile fatty acids, and the accumulation of PHAs using pure or mixed culture cell methods. This approach has a high cost and requires significant resources (Alsafadi and Al-Mashaqbeh 2017). This work represents the first attempt to produce PHAs from untreated OMW using the genus Bacillus. We developed a new culture method for the production of PHA from OMW in order to reduce its production cost, while helping to reduce pollution.

Materials and methods

Materials

The strains Bacillus thioparans OM39, Bacillus thuringiensis OM55, Bacillus cereus OM75, Bacillus amyloliquefaciens OM81, and Klebsiella oxytoca OM91 were retrieved from the Laboratory Bioresources, Environment and Biotechnologies (BeB, LR22ES04) at the Higher Institute of Applied Biological Sciences of Tunis (ISSBAT). In a previous study of Arous et al. (2018), these strains were isolated and identified from a sequential batch reactor (SBR) that utilized OMW as a substrate. Fresh OMW was collected from a three-phase decanter olive mill located in the Nabeul region of Tunisia and stored in clean 1.5-l bottles at 4 °C until use. Prior to use, the OMW was decanted and filtered. The OMW was then subjected to various analyses (Table 1), including assessment of pH, electrical conductivity (EC), total solids (TS), volatile solids (VS), chemical oxygen demand (COD), total phenols, ammoniacal nitrogen (NH4-N), total sugars, and total phosphorus (Fleyfel et al. 2022).

Growth conditions

PHAs are produced by bacteria only under nitrogen-limited conditions (Naser et al. 2021). The PHA production process involved two steps: (i) the bacteria were grown in 250-mL Erlenmeyer flasks containing 50 mL of Brain Heart Infusion (BHI) medium. The flasks were inoculated with 1% (v/v) of the inoculum and then incubated at 37 °C with constant shaking at 120 rpm for 24 h. (ii) Once an optimal cell concentration was reached, the cell mass was collected by centrifugation at 6000 rpm for 20 min, washed with sterile distilled water, and inoculated into nitrogen-limited medium (NLM). The culture was incubated at 37 °C on a rotatory shaker (120 rpm) for 72 h. After selecting the best PHA-producing strain, the same fermentation protocol was repeated, except that glucose was replaced by OMW at different concentrations (25%, 50%, 75%, 100% (v/v) diluted in pure water).

Screening for PHAs producing strains

The PHA-producing bacteria were identified using a nitrogen-limiting medium (NLM) agar and incubated for 2 days at 37 °C. The composition of the medium was according to Chaijamrus and Udpuay (2008) and contained (in g/L): 20 glucose (as the sole carbon source), 0.5 NH4Cl, 2.3 KH2PO4, 2.3 Na2HPO4, 0.5 MgSO4·7H2O, 0.01 CaCl2, and 20 agar, supplemented with 5 mL/L trace element solution containing (in g/L): 0.2 ZnSO4·7H2O, 1.8 MnCl2·4H2O, 2.8 H3BO4, and 0.1 CuSO4. The pH was adjusted to 7 and the medium was autoclaved for 15 min at 121 °C. This medium was used for all subsequent experiments in this study.

Sudan Black B plate assay

PHA production potential of all isolates was evaluated qualitatively using Sudan Black B dye (Liu et al. 1998). An ethanolic solution of Sudan Black B (0.05%, w/v) was added to the colonies, and the plates were incubated for 30 min. Colonies that turned dark blue in color after rinsing with 96% ethanol were considered positive for PHA production (Phanse et al. 2011).

Sudan Black B staining

The bacterial smears were first air-dried and then fixed on glass slides. They were then stained with Sudan Black B (0.3% in 70% ethanol) for 15 min, followed by washing with ethanol, rinsing, and staining with safranin for 5 min (Wei et al. 2011). Finally, the stained smears were observed under an optical microscope with a 100× oil-immersion objective.

Determination of cell weight

The culture medium for PHA production was centrifuged at 4 °C for 20 min at 6000 rpm. The resulting pellets were washed with distilled water and transferred into pre-weighed sterile glass flasks. The flasks were then oven-dried at 55 °C for 24 h, cooled down in a desiccator, and weighed to determine the PHA yield (Wang and Lee 1997). The experiment was performed in triplicate. The supernatant was collected to measure the biochemical parameters such as glucose abatement rate (which confirms its consumption by bacteria), phenolic compounds, and soluble COD.

Extraction of the accumulated PHA

The dried biomass was collected, and the cellular granules were weighed and crushed using a mortar. Then, 10 mL of chloroform were added to the flasks, which were sealed and kept at room temperature for 24 h to facilitate the release of intracellular PHAs. The mixture was filtered through Whatman No. 1 paper and collected in a pre-weighed flask. The solvent was evaporated using a rotavapor, and the PHA obtained was weighed. The biopolymer content was calculated as a percentage of cellular dry weight using Eq. 1. The experiment was performed in triplicate:

$$\mathrm{PHA \%}=[\mathrm{weight \,of\, PHA }(\mathrm{g}/\mathrm{L})/\mathrm{weight \,of \,dry \,cells }(\mathrm{g}/\mathrm{L})]\times 100.$$
(1)

A film of PHA was produced and characterized.

PHA characterization

UV–Vis spectroscopy

The extracted PHA was dissolved in chloroform and its spectra were scanned in the range of 200–320 nm using a Spectrum Instruments SP-UV 300RB. The spectra were analyzed for a net peak at 240 nm, and chloroform blank was used for comparison (Selvakumar et al. 2011).

FTIR spectroscopy

The PHA film was characterized via Fourier Transform Infrared (FTIR) spectrometry in the range of 4000 to 400 cm−1 according to Beji et al. (2023) using the PerkinElmer Frontier IR/NIR systems (USA).

Differential scanning calorimetry (DSC)

The thermal properties of the PHA film were studied using differential scanning calorimetry with a DSC Q2000 and an RCS90 cooling system (TA Instruments). Samples of PHA weighing 5–10 mg were placed in aluminum pans and heated from − 40 °C to 400 °C at a rate of 10 °C per minute in a nitrogen atmosphere (García-Quiles et al. 2019).

Thermogravimetric analysis (TGA)

The thermogravimetric analysis (TGA) of the PHA sample was conducted using a Mettler Toledo TGA/DSC 3 + Equipment. The sample was heated from 25 to 800 °C at a rate of 10 °C/min under a N2 flux (30 mL/min) (Beji et al. 2023).

Statistical analyses

The mean values of three analyses were calculated and expressed as mean ± standard deviation using Origin Pro 2018 software. Statistical analysis was performed using a Student’s t-test to determine the differences among means at a 95% confidence level using Statgraphics software.

Results and discussion

There is a growing interest in finding sustainable alternatives to petroleum-based plastics which pose a significant environmental threat. This study explores the potential of using OMW as an economical carbon source for producing bacterial PHAs, utilizing B. thioparans OM39, B. thuringiensis OM55, B. cereus OM75, B. amyloliquefaciens OM81, and K. oxytoca OM91 previously isolated from OMW (Arous et al. 2018).

Confirmation of PHA production

The five strains were screened for their ability to accumulate PHA granules using the Sudan Black B (SB) plate test, which is a rapid and viable colony staining method. The test distinguishes between PHA-accumulating bacteria, which appeared blue-black, and non-PHA producers, which appeared white. Upon staining with SB, all five isolates appeared blue-black, indicating the presence of PHA granules (Supplementary Data 1). The SB-stained bacterial cells were examined under a microscope, revealing the presence of vesicles like granules (look like holes) within pink-colored cells, as shown in Supplementary Data 2. This staining method indicated the presence of PHA granules in the cells (Wei et al. 2011). The staining also enabled a comparison of the PHA production ability among the bacterial strains, revealing that Cupriavidus taiwanensis 184 had a significantly higher amount of PHA granules (black region) compared to the other strains.

Table 1 OMW physicochemical characteristics
Full size table

Bacteria growth and accumulation of PHA

All five bacterial strains showed a significant ability to produce PHAs, ranging from 13.6% of dry cell weight for B. thuringiensis to 30.2% for B. amyloliquefaciens (Table 2). These findings are consistent with those reported by Chen et al. (1991) when cultivating various Bacillus strains using glucose as the carbon source. The reduction in residual sugars (from 20.0 g/L to values between 18.5 and 7.8 g/L) and the increase in glucose abatement rate (reaching 61.2%) indicate that these strains can use glucose as a carbon source and convert it into PHA in the form of reserve granules in their cells. The obtained bioplastics are displayed in Supplementary Data 3. B. amyloliquefaciens OM81 exhibited the highest biomass and PHA yields when grown on nitrogen-limiting medium supplemented with glucose as a carbon source, among the five strains tested. The residual glucose concentration was reduced to 7.8 g/L, with a glucose abatement rate of 61.2%, and the PHA yield was 0.14 g/L of dry weight, corresponding to a production rate of 30.2%. Due to its excellent ability to convert glucose into PHA-rich biomass, this strain was selected for further investigation. The extracted PHA (Supplementary Data 3A) was subjected to extensive physicochemical (UV–visible), structural (FTIR), and thermal (DSC and TGA) characterizations.

Table 2 Growth and PHA production of selected strains in glucose-based medium
Full size table

The biomass and PHA production of B. amyloliquefaciens OM81 were determined in media with varying concentrations of untreated OMW (25%, 50%, 75%, and 100% (v/v) diluted in water). The results presented in Table 3 show that growth was observed in all media, including 100% OMW, indicating that the strain was not inhibited by the raw OMW. The best biomass production was obtained in the medium with 100% OMW (2.81 g/L), which could be attributed to the strain’s prior acclimation to high OMW concentrations in a batch reactor (SBR) and the presence of low levels of toxic compounds, such as tannins and simple phenolic compounds in the raw OMW. The highest yield of PHAs (11.2%) was obtained in the presence of 50% OMW concentration, which can be attributed to the dilution of OMW, leading to a decrease in both COD and phenolic compounds of the raw OMW, thus promoting cell proliferation and PHA production. However, these yields are relatively low compared to those observed in synthetic medium (glucose). Alsafadi and Al-Mashaqbeh (2017) reported that the optimal cell growth of Haloferax mediterranei was achieved at 5% of OMW, with growth being inhibited by increasing concentrations of OMW from 50 to 75%.

Table 3 Effect of OMW concentration on growth and PHA production of Bacillus amyloliquefaciens strain OM81
Full size table

UV–visible spectrophotometric analysis

PHAs produced by the various bacterial strains were analyzed using UV–visible spectrophotometry. The bacterial cultures were scanned, and a peak was observed at 243 nm with four strains, as shown in Fig. 1. This peak was particularly prominent in OM39, OM81, and OM91, suggesting that the PHA produced by these bacteria is likely to be the polyhydroxybutyrate biopolyester (Selvakumar et al. 2011). This analysis was conducted on extracts obtained from the culture of OM81, which yielded the highest amount of PHA, grown in media with increasing OMW concentrations. Spectra obtained from 25%, 50%, 75% and 100% OMW concentrations also showed a peak at 243 nm, corresponding to the PHB peak, as depicted in Fig. 2. Selvakumar et al (2011) and Getachew and Woldesenbet (2016) obtained similar results in their studies on Haloarcula marismortui MTCC 1 and Bacillus spp., respectively.

Fig. 1
figure 1

UV–Vis spectrophotometer scanning spectrum of PHA compounds extracted from bacterial strains grown in glucose-based medium of the five bacterial strains: (◆) B. thioparans OM39, (◊) B. thuringiensis OM55, (△) B. cereus OM75, (▲) B. amyloliquefaciens OM81, and (□) K. oxytoca OM91

Full size image
Fig. 2
figure 2

UV–Vis spectrophotometer scanning spectrum of PHA compounds extracted from the strain Bacillus amyloliquefaciens OM81 grown in media where glucose was replaced by OMW at different concentrations: (◆) 25%, (◊) 50%, (△) 75%, and (▲) 100% (v/v)

Full size image

Structural and thermal characterization of PHA produced by B. amyloliquefaciens strain OM81

FTIR spectrophotometric analysis

The FTIR spectrum of PHA produced by strain OM81 using glucose as a carbon substrate (Fig. 3) revealed the presence of various absorption peaks. The peak at 2917–2849 cm−1 was assigned to the presence of an alkyl-CH3 group, while the absorption bands at 1730 and 1600 cm−1 indicated the elongation of the carbonyl groups C=O (a distinctive PHB peak) and −C=C, respectively (Guo et al. 2019; Miranda et al. 2023; Mohapatra et al. 2017). Peaks at 1492 and 1451 cm−1 were assigned to the asymmetrical deformation of the C–H bond in the group CH2, and the peak at 1366 cm−1 indicated the methyl group CH3 (Valdez-Calderón et al. 2022). The vibrations between 1250 and 1027 cm−1 corresponded to the elongation of the C–O–C bond of the ester function (Muneer et al. 2022). These FTIR spectrum results were consistent with those found in the literature for PHB (Mohapatra et al. 2017).

Fig. 3
figure 3

FTIR spectrum of PHA extracted from the strain Bacillus amyloliquefaciens OM81 grown in glucose-based medium

Full size image

Differential scanning calorimetry and thermogravimetric analyses

The PHA extracted from OM81 bacterial strain, produced in the presence of excess glucose in synthetic medium, was subjected to thermal analyses using differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). The resulting DSC thermogram exhibited two peaks: the first peak at 69 °C was attributed to the boiling point of residual chloroform, while the second peak at 147.3 °C corresponded to the melting of the produced PHA (Alsafadi and Al-Mashaqbeh 2017; Kim et al. 2020). Mohapatra et al. (2017) found similar results for PHB homopolymer, showing a dual melting point at temperatures of 120.6 °C and 145.9 °C attributed to the re-crystallization followed by cross-linking isomerization reactions. In our case, only the second peak was observed. These findings indicate the potential of B. amyloliquefaciens strain OM81 to produce PHA as a suitable material for bioplastic manufacture, especially for food packaging applications as confirmed by Tripathi et al. (2019). A lower melting temperature (112.8 °C) was observed by Choi et al. (2021) on short-chain-length PHA copolymer produced by Pseudomonas sp. B14-6 b.

Thermal stability analysis of the produced PHA was conducted using TGA. The TGA and DTG curves are shown in Fig. 4. The first weight loss observed in the 25–75 °C temperature range was attributed to the evaporation of residual chloroform. The second weight loss was observed in the temperature range of 210–355 °C, and the third weight loss was observed in the temperature range of 360–460 °C. The observed phenomenon could be attributed to the cross-linking isomerization process that takes place before the degradation as stated before by Mohapatra et al. (2017). However, it is challenging to determine if the cross-linking process follows a distinct mechanism for the various groups in PHB. These results are consistent with previous studies, which suggested that cross-linking isomerization can improve the polymer’s thermal stability (Mohapatra et al. 2017). Further research has indicated that introducing a (4HB) monomer into the polymer composition substantially reduced the melting temperature and concurrently resulted in a gradual improvement in thermal decomposition resistance (Sedlacek et al. 2020). The first derivative TGA profile revealed a degradation temperature (Td) of 420 °C (Fig. 4), lower than the one observed by Mohapatra et al. 2017 (466.8 °C), but higher than the degradation temperature (288 °C) observed by Miranda et al. (2023).

Fig. 4
figure 4

TGA (A) and DTG (B) curves of PHA extracted from the strain Bacillus amyloliquefaciens OM81 grown in glucose-based medium

Full size image

Conclusions

The present study revealed that all five bacterial strains tested were capable of synthesizing PHA in a nitrogen-limited environment. Among them, B. amyloliquefaciens strain OM81 exhibited the highest biomass and PHA yield of 0.14 g/L of dry weight, corresponding to a production rate of 30.2%, a residual glucose concentration reduced to 7.8 g/L, and a glucose abatement rate of 61.2%. The synthesized PHA was subjected to structural and thermal characterization. FTIR spectrum revealed an absorption band at 1730 cm−1 assigned to the elongation of the PHB carbonyl groups. Additionally, we investigated the suitability of using OMW (11.8%) as a carbon source for PHA synthesis by this strain. The study underscores the potential of utilizing OMW to produce biodegradable polymers, while also reducing production costs. However, further optimization of the biopolyester production process is needed. Future research should also investigate a low-cost pretreatment of OMW and the use of other waste streams as potential carbon sources for PHA synthesis.