*2.3. Isolation of PHA from Bacterial Biomass*

Right after cultivation, the isolation of PHA from wet bacterial biomass is performed. Predominantly for halophilic strain *H. halophila*, but also for *S. thermodepolymerans*, a very important strategy is the hypotonic environment, which induces disruption of cells leading to release of the intracellular cell content including PHA granules. The main idea of isolation is based on the use of hypotonic conditions together with SDS (TCI, Zwijndrecht, Belgium) applied in low but still sufficient concentration along with the elevated temperature. These factors should lead to cell disruption and the release of water-insoluble PHA granules. For both microorganisms, we focused at the beginning on the optimization of SDS concentration (in range 1 to 10 g/L), temperature treatment (higher and lower than cultivation temperature), and the ratio of microbial biomass vs. volume of SDS solution.

The first optimized parameter was the most efficient concentration of SDS. The experiment was performed in this way: 10 mL of the culture after cultivation was centrifuged in a plastic test tube with a screw cap (3460× *g*, 10 min), the supernatant was discarded and replaced by prepared aqueous solution of SDS (1; 2.5; 5 and 10 g/L), in replicates for each SDS concentration. Pellets were resuspended and suspensions were incubated in a water bath at 70 ◦C for 120 min. After incubation, test tubes with suspensions were centrifuged again (3460× *g*, 5 min), supernatants were discarded and preserved for further use, the pellets containing predominantly isolated PHA granules were washed with distilled water and then dried to constant weight at 70 ◦C. The amount of the product was determined gravimetrically and the purity of obtained PHA was determined by GC-FID as the same as described above.

The second, optimized parameter was the temperature of isolation or combination of different temperatures, i.e., temperature treatment. For both strains, the same treatments were tested: 50 ◦C (120 min), 70 ◦C (120 min), 90 ◦C (120 min) and a combination of coolingdown and heating-up; 4 ◦C (120 min) followed by 70 ◦C (120 min). SDS was applied at a concentration of 5 g/L, otherwise, the experiment was performed in the same way as described above for optimization of SDS concentration.

To reduce the amount of detergent (SDS) using a higher biomass amount, the third series of optimization experiments was based on a comparison of the effectivity of isolation with different ratios of initial biomass concentration and 10 mL of aqueous SDS (5 g/L) solution in the most to verify the purity of obtained material and compare product yield. Also, the purpose of this experiment was to verify the robustness of the isolation process concerning different initial biomass concentrations—using a higher ratio of biomass to SDS solution can significantly reduce the cost of the isolation process (amount of SDS per biomass) and also significantly reduce environmental burden caused by SDS in wastewater. In practice, different amounts of biomass were obtained by centrifugation of a larger volume of cell culture when the volume of SDS solution was still 10 mL. However, the approach was the same as previous ones, using optimal SDS solution concentration of 5 g/L for both strains and optimal "heat treatment": 70 ◦C for 2 h for *H. halophila* and keeping at temperature 90 ◦C for 2 h for *S. thermodepolymerans*. All isolation experiments based on long-term exposition to elevated temperatures were performed in a water bath (BL 4/150, WSL, Czestochowa, Poland).

#### *2.4. Characterization of Isolated PHA Polymer*

In this work, we produced a homopolymer of 3-hydroxybutyrate-poly(3-hydroxybutyrate) (PHB), the most common representative of PHA. Quantification of the polymer in bacterial biomass and determination of purity of the isolated samples were performed by GC-FID as described above.

Further, we used Fourier transform infrared spectroscopy (FT-IR, Nicolet iS50, Thermo Scientific, Waltham, MA, USA) with the built-in single-reflection diamond attenuated total reflectance (ATR) crystal to collect infrared spectra of the original biomass, polymer isolates, and the reference material, commercially available P(3HB) (Biomer, Schwalbach am Taunus, Germany). Each spectrum was collected in the range of 4000–400 cm−<sup>1</sup> as an average of 16 scans with a resolution of 4 cm−<sup>1</sup> (data spacing 0.5 cm<sup>−</sup>1).

For determination of molecular weight and its distribution (polydispersity index PDI) of the polymer, we used size-exclusion chromatography with multi-angle light scattering detector (SEC-MALS) (SEC chromatography, column PLgel mixed-C 5 μm, 300 × 7.5 mm, Agilent Technologies, Wilmington, DE, USA; detectors: MALS–DAWN HELEOS II, differential refractometer OPTILAB T-REX, Wyatt Technology, Dernbach, Germany). Using this method, we compared polymers isolated by different approaches, which were the most promising based on results from GC connected with the highest material purity. Approximately 200 mg of sample (biomass, isolated material) was weighed into glass pyrex test tubes, 10 mL of chloroform was added, then test tubes were tightly screwed with caps and during 12 h incubated at 70 ◦C in a thermoblock. After 12 h, the mixture was filtered through a paper filter on Petri dishes, and chloroform was freely evaporated. Then 1.5 mg of foil was weighed to the vial (volume 4 mL) and 1.5 mL of chloroform (LachNer, Neratovice, CZE) was added. The mixture was incubated at 50 ◦C in the thermoblock till all foil was dissolved, then the solution was filtered using nylon filters with 0.45 μm pore size into small vials, and the samples were analyzed.

Selected samples were also analyzed by electron microscopy. Bacterial cultures and isolated PHA samples were fixed using the high-pressure freezing method (EM ICE, Leica Microsystems, Wetzlar, Germany). Samples were centrifuged for 3 min at 4000 rpm and the generated pellet was pipetted on the 0.2 mm side of the 6 mm Al carrier type A and closed with the flat side of carrier type B (without using any cryo-protectant) for further processing for observation in the cryogenic scanning electron microscope (cryo-SEM). For freeze-substitution procedure followed by observation in the transmission electron microscope (TEM), samples were frozen in 0.2 mm side of the 3 mm Au carrier type A and closed with the flat side of carrier B. Both of the carriers were pre-treated with 1%

aqueous solution of lecithin (Sigma-Aldrich, Darmstadt, Germany) in chloroform. For cryo-SEM, high-pressure-frozen samples were transferred into a cryo vacuum preparation chamber (ACE 600, Leica Microsystems, Wetzlar, Germany), where they underwent freezefracture followed by sublimation for 7 min at −95 ◦C. Samples were then transferred into a scanning electron microscope equipped with a cryo stage (Magellan 400/L, FEI, Hillsboro, OR, USA) and observed using 1 keV electron beam at −120 ◦C. For TEM, high-pressurefrozen samples were transferred into a freeze substitution unit (AFS2, Leica Microsystems, Wetzlar, Germany) containing a freeze-substitution solution of 1.5% OsO4 in acetone. The freeze-substitution protocol was set as previously described in Kourilova et al. [28]. Thus prepared freeze-substitution samples were washed in pure acetone (3 × 15 min) and gradually infiltrated with epoxy resin (Epoxy Embedding Medium, Sigma-Aldrich, Darmstadt, Germany), mixtures with acetone 1:2, 1:1, 2:1 and pure resin for 1 h each. After the final exchange of pure resin, samples were left overnight under vacuum, then embedded in fresh resin and cured at 62 ◦C heat for 48 h. Cured blocks of samples were cut to ultrathin sections using a diamond knife with cutting angle 45◦ (Diatome, Nidau, Switzerland) and ultramicrotome (UTC 7, Leica Microsystems, Wetzlar, Germany), and stained using solutions of uranyl acetate and lead citrate. Contrasted ultrathin sections were imaged using a transmission electron microscope (Talos F200C, Thermo Fisher Scientific, Waltham, MA, USA) using a 200 keV beam.
