Recent updates to microbial production and recovery of polyhydroxyalkanoates

Abstract

The increasing use of synthetic polymers and their disposal has raised concern due to their adverse effects on the environment. Thus, other sustainable alternatives to synthetic plastics have been sought, such as polyhydroxyalkanoates (PHAs), which are promising microbial polyesters, mainly due to their compostable nature, biocompatibility, thermostability, and resilience, making this biopolymer acceptable in several applications in the global market. The large-scale production of PHAs by microorganisms is still limited by the high cost of production compared to conventional plastics. This review reports some strategies mentioned in the literature aimed at production and recovery, paving the way for the bio-based economy. For this, some aspects of PHAs are addressed, such as synthesis, production systems, process control using by-products from industries, and advances and challenges in the downstream. The bioplastics properties made them a prime candidate for food, pharmaceutical, and chemical industrial applications. With this paper, it is possible to see that biodegradable polymers are promising materials, mainly for reducing the pollution produced by polymers derived from petroleum.

Introduction

Petroleum-based plastics are introduced into the market due to their low cost and high versatility. It is estimated that world plastic production in the next 20 years will reach 644 million tons per year (Gutschmann et al. 2022). Bioplastics production can contribute to achieving several Sustainable Development Goals, as well as the Paris Agreement 2016, where CO₂ reduction was the main objective.

There are several types of bioplastics, which must be differentiated into bio-based plastics, biodegradable or both, or non-biodegradable, are mainly produced from agricultural raw materials and have chemical structures identical to those of fossil origin (Rosenboom et al. 2022). Polyhydroxyalkanoates (PHAs), which are polyesters synthesized by microorganisms (Ahmady-Asbchin et al. 2020; Gutschmann et al. 2022), have different properties depending on the chain length of their incorporated monomers, their molecular weight, and post-synthetic composition (Dartiailh et al. 2020; Choi et al. 2020; Turco et al. 2021).

PHAs are mainly produced by the genera Bacillus, Pseudomonas, Azotobacter, Rhizobium, Cupravidus, Nocardia, Iodobacter, and others (Raza et al. 2018; Pakalapati et al. 2018; Mozejko-Ciesielska et al. 2019; Arul Manikandan et al. 2020; Hassemer et al. 2021; Kumar et al. 2021); they are biocompatible and non-toxic, making them excellent candidates for medical and pharmaceutical applications, such as drug delivery. Thus, replacing synthetic plastics with biodegradable polymers is a viable alternative that significantly reduces the environmental impact of excessive use and accumulation of synthetic plastics (Eesaee et al. 2022).

The production of PHAs on a large scale is still limited by the cost of production compared to conventional plastics based on fossil fuels, ranging from 1.18 to 6.12 €/kg, depending on the polymer composition, that is, its cost is about three times higher than the main petrochemical-based polymers, which cost about 1.0 €/kg (Saavedra del Oso et al. 2020). For example, the selling price of commercial P(3HB) produced by microorganisms was approximately US$ 3.50/kg, while polyethylene and polypropylene presented a cost of approximately US$ 1.20 to US$ 1.30/kg in 2018. There is an indication that the costs of substrates for the production of PHA can be 30 to 50% of the total production costs (Choi and Lee 1997; Song et al. 2022).

However, PHAs can be produced using alternative substrates based on agro-industrial residues (food residues, sugarcane molasses, rice parboiled water, wastewater from paper mills, cheese whey, and others) (Amaro et al. 2019; Mannina et al. 2019; Rebocho et al. 2019; Sen et al. 2019; Wan et al. 2019; Yadav et al. 2020; Shen et al. 2022). In this way, the appropriate selection of local raw materials can help to reduce production costs and minimize transport routes, while at the same time minimizing negative environmental impacts (Gutschmann et al. 2022; Guleria et al. 2022; Saravanan et al. 2022). To exemplify, a study carried out by De Melo (2021) in the production of P(3HB) by Bacillus megaterium using substrates from renewable sources (candy industry effluent and rice parboiling water) verified that there is a cost reduction of approximately US$ 1.0 per liter of culture medium compared to the sucrose-based synthetic medium.

PHAs produced biosynthetically are extracted and recovered through a series of downstream processing (DSP) techniques. The most commonly used DSP techniques include biomass pretreatment (mechanical and chemical), cell lysis-assisted extraction using various solvents, enzymatic extraction, ultrasonic systems, supercritical technology, and further solvent purification. Conventional extraction methods often use solvents that are toxic, expensive, and non-recyclable, making these methods environmentally unsustainable (Aramvash et al. 2018).

This review describes the challenges found in the biotechnological production of PHAs, with an emphasis on alternative low-cost substrates, potential recovery techniques for PHAs derived from microbial biomass, and some applications of biopolymers. Literature information was searched on March 5, 2023, on the following electronic platforms: ACS Publications, the Royal Society of Chemistry, Scielo, Science Direct, Springer Nature, Wiley Online Library, Taylor & Francis, and Google Scholar, using the following keywords: polyhydroxyalkanoates, polyhydroxybutyrate, production of P(3HB), P(3HB) industrial waste, and biopolymer extraction. The search was performed using isolated or associated words to retrieve the maximum number of articles. In the last fifteen years, the production of bioplastics such as P(3HB) from industrial effluents has received significant attention from the scientific community. This is evidenced by a total of 14,598 scientific articles published in recent years (2007–2022) (Fig. 1). Although there is an expressive number of publications, there are still few works using alternative substrates (e.g., agro-industrial co-products, organic residues, wastewater, and others) and mainly recovery methods with green technology (e.g., supercritical technology, ultrasonic systems, and others).

Fig. 1.

Scopus-indexed publication numbers in recent years (2007–2022) for biotechnological production of polyhydroxyalkanoates (Archived until March 5, 2023)

Synthesis of polyhydroxyalkanoates

Polyhydroxyalkanoates (PHAs) are widely recognized as highly promising bioplastics, with the advantages of biocompatibility and biodegradability. PHAs are water-insoluble, UV-resistant, and present good resistance to hydrolytic attack. PHAs also have chiral molecules, and their degradation depends mainly on their type and composition, environmental conditions, and the type of microorganisms (depolymerase to degrade PHAs) (Pakalapati et al. 2018).

PHAs can be divided into three types based on carbon chain length: short-chain length (scl-PHAs); medium-chain (mcl-PHAs); and long-chain (Icl-PHAs) (Raza et al. 2018). Short-chain ones (scl) have 3–5 carbon atoms and can be used in food packaging and disposable products (Johnston et al. 2018; Grigore et al. 2019). On the other hand, those with medium chain length (mcl), 6–14 carbon atoms, such as homopolymers, poly (3-hydroxyhexanoate), simplified as P(3HHx), and poly (3-hydroxyoctanoate), or P(3HO). The long chain (lcl) has 15 or more carbon atoms per monomer and is rare in nature, which makes their interest in the process of developing bioplastics low (Choi et al. 2020).

PHAs are a family of natural polyesters that are produced intracellularly by a wide range of microorganisms, mainly as an energy carbon reserve. It is estimated that more than 300 species of bacteria and more than 90 genera accumulate biopolymers naturally, such as Bacillus, Pseudomonas, Azotobacter, Rhizobium, Nocardia, and others (Raza et al. 2018; Pakalapati et al. 2018; Mozejko-Ciesielska et al. 2019; Arul Manikandan et al. 2020). Bacterial species that have been reported to be significant producers of PHAs are Alcaligenes latus (Berwig et al. 2016), Burkholderia sacchari (Oliveira-Filho et al. 2020), Cupravidos necator (Li and Wilkins 2020; Brojanigo et al. 2021), Bacillus megaterium (Hassemer et al. 2021), Ralstonia eutropha (Arul Manikandan et al. 2020), and Iodobacter sp. PCH194 (Kumar et al. 2021).

The gender Bacillus is preferred by several industries for the production of PHAs since it has advantages over other bacterial species mainly due to the absence of a lipopolysaccharide layer, which facilitates its extraction, present genetic stability, fast growth, can grow in low-cost substrates, and can produce endotoxin-free P(3HB) (PHA homopolymer) compared to Gram-negative bacteria (Mohapatra et al. 2017). The pathway of P(3HB) biosynthesis by B. megaterium, starts with the conversion of two molecules of acetyl-CoA into acetoacetyl-CoA in the presence of an enzyme acetyl-CoA acetyltransferase (phaR). Subsequently, NADPH-dependent acetoacetyl-CoA reductase (phaB) helps in the reduction of acetoacetyl-CoA to hydroxybutyryl-CoA. This compound is then used as a monomer to polymerize P(3HB) with polyhydroxyalkanoic acid (phaC) synthase.

Several strain improvement strategies have been reported for enhanced P(3HB) accumulation in different microbial systems (Park et al. 2019). Therefore, for the incorporation of monomers to occur, the substrate provided at the time of synthesis must be adequate, i.e., a carbon source that can be converted into the desired hydroxyacyl-CoA, through the metabolic pathways present in the bacterial cell, must be provided. In addition, the bacteria must present PHA synthase, an enzyme capable of incorporating the generated hydroxyacyl-CoA into the formed polyester (Gutschmann et al. 2022).

Another microbial strain that stands out for the production of P(3HB) is the Cupriavidus necator (also known as Alcaligenes eutropha, Ralstonia eutropha, and Wautersia eutropha), which can assimilate fermentable sugars from lignocellulosic biomass (glucose, xylose, and arabinose) into the biopolymer (Soto et al. 2019; Li and Wilkins 2020), where the microbial biosynthesis of P(3HB) starts through the reduction of two molecules of acetyl-CoA to form acetoacetyl-CoA, which is subsequently condensed into hydroxybutyryl-CoA and this final product can then be used as a monomer to polymerize P (3HB) (Pakalapati et al. 2018).

Bacteria belonging to the Pseudomonas species comprise two metabolic pathways to produce precursors for mcl-PHAs biopolymer synthesis. In each pathway, it is possible to use different substrates to provide 3-hydroxyacyl (3HA) precursors that are used to synthesize mcl-PHAs. Aliphatic carbon sources, for example, fatty acids, are degraded through the process of β-oxidation by Pseudomonas. This pathway is involved in the oxidation of fatty acids to enoyl-CoA, (S)-3-hydroxyacyl-CoA, and I-3-ketoacyl-CoA. They are then converted to I(R)-3-hydroxyacyl-CoA. Pseudomonas species are also capable of using the fatty acid synthesis pathway again to synthesize mcl-PHA monomers from other carbon sources such as ethanol, glucose, or gluconate. Therefore, substrates are oxidized to acetyl-CoA and then transformed by various reactions to malonyl-CoA and activated by transacylation to (R)-3HA-acyl transporter protein (ACP). The resulting acyl-ACP and malonyl-ACP intermediates are converted to (R)-3-hydroxyacyl-ACP and are further elongated by successive two-carbon units. In most Pseudomonas, the PhaG-specific transacylase transforms IL-ACP intermediates into (R)-3-hydroxyacyl-CoA (Mozejko-Ciesielska et al. 2019; Rebocho et al. 2020). Various strains of Pseudomonas putida were recognized as efficient cell producers for mcl-PHA production, which was fundamentally studied and optimized using conventional carbon sources (Mozejko-Ciesielska et al. 2019), and the strain Pseudomonas citronellolis was used in the production of mcl-PHA using alternative substrates, a by-product of the fruit industry (e.g., apple pulp) (Rebocho et al. 2020).

PHAs that are produced on an industrial scale include poly(3-hydroxybutyrate) P(3HB), poly(3-hydroxybutyrate-co-4-hydroxybutyrate) (P(3HB-co-4HB)), poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV), and poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBHHx) (Tan et al. 2017). P(3HB) is one of the most prominent and important thermoplastic polyesters, with characteristics similar to synthetic plastics, such as crystallinity, a high melting point, optical purity, and desirable water and gas barrier properties. In with has several applications, including the production of packaging materials, automotive components, and home appliances (Amaro et al. 2019). It also gains space in studies for applications in the medical field, such as suture threads, surgical meshes, dressings, and fabrics for bone and nerve regeneration, for example (Hossian et al. 2022). Another characteristic is the degradation rate, which is faster compared to other bioplastics (Dietrich et al. 2020). Due to this characteristic of biodegradability, it does not release waste that is harmful to the environment when compared to conventional plastics (Alarfaj et al. 2015). One of the difficulties arising from P(3HB), which often hinders its introduction on an industrial scale, is its high fragility, low ductility, and high crystallinity, which can cause physical aging during secondary crystallization and result in the formation of irregular pores on the surface, limiting the flexibility of the material (Tănase et al. 2015).

Another factor that may limit industrial production is the high cost of P(3HB) which is in the range of US$ 2.25–2.75/lb, which is three- to four-times higher than conventional plastics (US$ 0.60–0.87/lb), and this restricts its use in the industrial and commercial sectors. However, the total production cost of P(3HB) depends on the microorganism (yield and productivity), substrates (carbon and nitrogen sources), type and conditions of fermentation (temperature, aeration, and pH), and recovery processes and/or purification (Sedlacek et al. 2019). Since the carbon source can represent 25–45% of the total production costs (Nath et al. 2008; Yeo et al. 2018; Amaro et al. 2019; Carlozzi et al. 2019; Singh et al. 2019), searching for alternative substrates from cheaper carbon sources becomes relevant for the industrial sector.

Strategies of production and process control of polyhydroxyalkanoates using by-products from industries

By-products generated in various agro industries are gaining importance as substrates for the obtention of PHAs (Al-Battashi et al. 2019; De Donno et al. 2021; Sirohi et al. 2020a). In recent years, there are numerous studies on the conversion of alternative carbon sources into PHAs (Table 1), including frying oil (Tufail et al. 2017), soybean oil (Schmidt et al. 2016), residual animal fat (Gutschmann et al. 2022), fruit pomace (Rebocho et al. 2020), vinasse and sugarcane molasses (Zanfonato et al. 2018; Dalsasso et al. 2019; Sen et al. 2019), whey (Amaro et al. 2019), rice bran, oil cake, wheat bran, rice straw (Krishnan et al. 2017), and cassava peel (Hierro-Iglesias et al. 2022), among others.

Table 1.

Overview of some carbon sources for microbial PHA production

Carbon sources	Microorganism	Fermentation system	Operational conditions	PHA type	Cell mass (g/L)	Yield (g PHA/g)	PHA (g/L)	References
Glucose and urea	Ralstonia eutropha	–	–	P(3HB)	164	0.48	–	Tsang et al. (2019)
	Cupriavidus necator NSDG-GG	Feed-batch	30 °C, pH 6.8, 300 rpm, 1 vvm, 72 h	P(3HB)	111	0.50	–	Biglari et al. (2020)
Molasses	C. necator	–	–	P(3HB)	2.86	0.78	–	Tsang et al (2019)
	Methylobacterium sp. ISTM1	Shake flash	pH 7.0, 72 h	PHA	1.7	–	1.4	Tyagi et al. (2022)
Molasses and Vinasse	C. necator DSM 545	Batch	35 °C, pH 7.0, 450 rpm, 0.125 vvm, 48 h	P(3HB)	12.4	–	12.6	Dalsasso et al. (2019)
Parboiled water and Candy industry effluent	Bacillus megaterium ATCC 14581	Batch	30 °C, pH 7.0, 4 vvm, 500 rpm, 32 h	P(3HB)	7.55	2.85	3.78	Hassemer et al. (2021)
Cheese whey	B. megaterium CCM 3027	Shake flash	30 °C, pH 7.0, 150 rpm, 48 h	P(3HB)	2.9	1.5	–	Obruca et al. (2011)
	Thermus thermophilus HB8	Shake flash	35 °C, pH 6.5, 200 rpm, 48 h	P(3HB)	1.6	–	0.5	Pantazaki et al. (2009)
Sugar maple (wood hydrolysate)	Burkholderia cepacia	Batch	30 °C, pH 7.0, 150 rpm, 1 vvm, 96 h	P(3HB)	16.97	0.19	8.72	Pan et al. (2012)
Apple pulp waste	Cupriavidus necator DSM 428 and Pseudomonas citronellolis NRRL B-2504	Batch	30 °C, pH 7.0, 30% of the air saturation (300 to 800 rpm), 48 h	mcl-PHA	4	1.2	–	Rebocho et al. (2020)
Mango peels	B. megaterium MC B. thuringienis IAM 12.077	–	–	P(3HB)	–	–	3.28 4.03	Gowda and Shivakumar (2014)
Glycerol	C. necator DSM 545	Batch	30 °C, pH 6.8, 200 rpm, 40 h	P(3HB)	6.76	–	4.84	Gahlawat and Soni (2017)
	Pandorea sp. MA O3	Shake flash	30 °C, pH 7.0, 150 rpm, 72 h	P(3HB	2.0	–	2.9	De Paula et al. (2017)
Vinasse and glycerol	C. necator DSM 545	Batch	35 °C, pH 7.0, 450–950 rpm, 0.1–1.0 vvm, 16 h	P(3HB)	5.0	–	26% of total biomass	Zanfonato et al. (2018)
Xylose (Waste office paper)	Burkholderia sacchari DSM 17165	Shake flash	30 °C, pH 7.2, 160 rpm, 72 h	P(3HB)	3.63	0.15	1.6	Al-battashi et al. (2019)
Coffee grounds oil	C. necator DSM 428	Batch	30 °C, pH ~ 7.0, 200 rpm, 48 h	PHA	16.7	0.77	–	Cruz et al. (2014)
Oleic acid	Pseudomonas oleovorans NRRL B-14683	Batch	–	PHA	1.3	0.5	–	Raza et al. (2018, 2019)
Ryegrass	Pseudomona putida KT2440	Shake flash	30 °C, pH ~ 7.0, 200 rpm, 48 h	mcl-PHAs	0.93	0.32	0.2	Davis et al. (2013)
Corn straw	P. putida KT2440	Feed-batch	30 °C, pH 7.0	mcl-PHAs	–	–	1.38	Arreola-Vargas et al. (2021)
Rice straw	P. putida KT2440	Shake flash	30 °C, 72 h	mcl-PHAs	–	0.47	0.6	Hossain et al. (2022)
Waste potato starch	R. eutropha NCIMB 11599	Fed-batch	30 °C, pH 6.8, 150 rpm initial increased up to 1000 rpm, DO 20%	P(3HB)	–	0.52	–	Haas et al. (2008)
Bamboo	Halomonas alcalina M2	Shake flash	30 °C, pH 10, 200 rpm, 48 h	P(3HB)	0.71	–	0.35	Luo et al. (2022)
Chicory Roots	C. necator DSM 545	Batch	30 °C, pH 7.0, 150 rpm, 48 h	P(3HB)	14	0.38	–	Haas et al. (2015)
Wheat bran	Halomonas boliviensis LC1	Batch	35 °C, pH 7.7, air rate/agitation (1.0 L/min/700 rpm and 4.0 L/min/900 rpm after 2 h), 48 h	P(3HB)	3.19	1.08		Van-Thuoc et al. (2008)

The fruit juice industry is responsible for generating by-products that represent 20–60% (w/w) of processed fruit (Evcan and Tari 2015). It is estimated that approximately 25% of the weight of the grape used in the production of wines, musts, and juices results in bagasse (bark, bunch, and seeds) (Dwyer et al. 2014). For the manufacture of apple juice, 65–75% (w/w) of the original mass of fruit is used (Evcan and Tari 2015). Grape, apple, or mango pomace are solid by-products composed mainly of polysaccharides (30%) and are possible carbon sources for the production of PHAs (Rebocho et al. 2019; Kourilova et al. 2021; Sirohi et al. 2020b).

Among the food processing industries, the pickle industry deserves specific attention, as it generates brine wastewater with a salt concentration of about 2–10%, which results in challenges for biological treatment (Wan et al. 2019). Due to their high content of organic acids, effluents from the pickle industry can be used as one of the alternatives for the production of PHAs.

Other by-products that are generated by the food industries are rice parboiling water and wastewater from the candy sector. In the parboiling of rice, about 4 L/kg of effluent are generated, with effluent being mainly a source of phosphorus, nitrogen, calcium, and magnesium, with low viscosity and an acidic pH (Hassemer et al. 2021). The candy sector (candy and confectionery) has a wide range of sugar-based products and processes (caramels, candies, chewing gum), chocolate-based and confectionery (cakes, chocolate-covered biscuits, and others) (Miah et al. 2018), which generate wastewater, mainly when cleaning equipment, which is rich in carbon sources (sucrose, glucose, and others). Studies regarding the use of these co-products in the production of PHAs are still scarce (Hassemer et al. 2021), but both have potential applications as substrates for the cultivation of microorganisms and the production of biocompounds (Mukherjee et al. 2016).

Whey, another by-product that is interesting for acidogenic fermentation, is mainly composed of lactose, proteins, mineral salts and lipids, and can also be used as nutrients for microbial cultures (Gouveia et al. 2017). The use of lactose from whey for the production of PHAs will drastically reduce the production costs of the biopolymer (Amaro et al. 2019), with around 85 million tons/year generated worldwide and only 60% used in food applications (Gouveia et al. 2017).

In the industry of sugar and ethanol from sugarcane, three by-products are obtained: sugarcane bagasse, molasses, and vinasse. Sugarcane bagasse is used to generate electricity and to produce second-generation ethanol on a commercial scale (Calil Neto et al. 2018). Also, in a fed-batch system from Burkholderia sacchari, 105 g/L of P(3HB) was obtained using sugarcane bagasse hydrolyzate rich in glucose, xylose, and arabinose (Cesário et al. 2014).

Molasses is one industrial residue from the crystallization of sugar produced in large quantities by the sugar and alcohol industries, where 1 ton of cane generates about 40 to 60 kg of molasses. This by-product has a high concentration of sugars and micronutrients (calcium, phosphorus, biotin, niacin, and riboflavin), but is deficient in essential minerals such as cobalt and selenium, which increase the activity of the PHB synthase enzyme (Tripathi et al. 2013). Another residue produced is vinasse, a liquid residue generated by the distillation of alcoholic must in the ethanol production process, which has a pH between 3.5 and 5, a dark brown color due to the presence of melanoidins, and minerals such as nitrogen, phosphorus, and potassium (Hoarau et al. 2018). Some studies reported the use of these by-products in the synthesis of PHAs (Tyagi et al. 2022).

Vegetable oil for PHAs biosynthesis is efficient due to its high carbon content and for contributing to higher PHAs yields during fermentation (Daly et al. 2018). Furthermore, the oxidation of fatty acids via the β-oxidation pathway facilitates the incorporation of CH₂ groups, generates PHAs heteropolymers, and increases the number of carbons in the side chain of the molecule. The incorporation of CH₂ groups in a PHAs chain modifies the physicochemical characteristics of the polymers, such as their lower oxygen permeability, lower melting point, and greater elasticity, making PHAs suitable for specific packaging applications (Dietrich et al. 2020).

Although there are reports of PHAs production using vegetable oils (Daly et al. 2018), one should take into account the fact that many of these oils are edible, so they should be focused on human food and not on the production of new materials. An alternative to being used in this process is residual frying oils (Tufail et al. 2017).

Crude glycerol, a waste from biodiesel production (renewable fuel) obtained mainly from oils and fats, has gained popularity as a low-cost substrate for the production of PHAs (De Paula et al. 2017; Gahlawat and Soni 2017; Kawaguchi et al. 2017; Chol et al. 2018; Yadav et al. 2020). Among these strategies is the use of combined substrates, such as using vinasse simultaneously with a carbon source with a high concentration, such as sugarcane molasses or glycerol. In a study carried out by Zanfonato et al. (2018), when using the mineral medium with vinasse, there was a high growth rate (0.41/h) of the recombinant bacterium Ralstonia eutropha, but there was a need of combined substrates to promote the accumulation of P(3HB), and supplementing vinasse with 30 g/L of glycerol produced approximately 5 g/L of P(3HB).

The production of PHAs from renewable lignocellulosic substrates offers economic, social, and environmental benefits. Among the agro-industrial residues that have already been reported as substrates for the obtention of biopolymers are rice straw (Hossain et al. 2022), corn straw (Li and Wilkins 2020; Arreola-Vargas et al. 2021), ryegrass (Davis et al. 2013), bamboo (Luo et al. 2022), and others, which are potential renewable lignocellulose raw materials. These materials are currently considered agricultural waste or are only used in low-value applications such as garden mulch, lightweight cardboard, and acoustic ceilings. Thus, there is a large amount of these residues that could be reused in the production of biopolymers (Hossain et al. 2022; Sohn et al. 2022).

To use lignocellulose, its sugars must be released efficiently for further conversion into bioproducts. Due to the chemical structure of cellulose, a pretreatment step is necessary using acids, hot water, sulphite, and ionic liquid, they allow a high release of sugar from enzymatic hydrolysis, but the pretreatments are conducted at high temperatures (80–220 °C) (Jimenez-Gutierrez et al. 2021; Sheng et al. 2021; Zan et al. 2021; Zhang et al. 2021; Hossain et al. 2022; Sohn et al. 2022). Sugars released from lignocellulose can be naturally metabolized by bacterial fermentation and produce medium-chain polyhydroxyalkanoates (mcl-PHAs), widely used in the biomedical area (Hossain et al. 2022).

The production of PHAs on a large scale is still a major challenge. In addition to microbial species and alternative substrates, which mainly aim to increase yield and reduce production costs, it is necessary to optimize process parameters (e.g., temperature, pH, aeration, agitation, dissolved oxygen, and others) and the farming system. In Table 1, there are studies that evaluated some process parameters and culture systems for cell productivity and/or the formation of PHAs.

The extracellular environment's pH level has a significant impact on the enzymatic activity of bacteria. More PHAs are produced in neutral pH fermentation media than in acidic or basic conditions. When it comes to pure culture media, any change in neutral pH slows down the fermentation process by lowering the cellular activities of the enzymes, such as Pha C and Pha Z, which in turn affects the growth rate and, eventually, the cell biomass. Ultimately, the medium’s composition completely determined how pH affected the formation of PHAs. For neutral fermentation media, a broad trend of pH value changes is typically seen. The pH of a fermentation medium rises in the early stages of incubation after substrate loading before resetting (Xu et al. 2018). Continuous and fed-batch fermentations have both seen a similar impact from pH fluctuation (Liu et al. 2018).

Controlling the dissolved oxygen (DO) rate in a system is essential to maximizing the productivity of fermentation systems, including the production of PHAs, since O₂ is closely linked to the maintenance of the cell membrane and contributes to the activation of different metabolic pathways. Although there are still a few studies that work with the optimization of the process in a bioreactor and consider the effect of oxygenation on the productivity of PHAs, most of the studies refer to the control of agitation (Liu et al. 2017). Accurate control of the aeration conditions of fermentation systems is usually difficult when using conventional O₂ electrodes due to the sensor’s detection limit. In this way, monitoring the redox potential, also known as redox potential, can be seen as a more accurate alternative due to its fast response and high sensitivity. In an aqueous system, the redox potential refers to the capacity of different molecules to receive or donate electrons through the action of oxidizing or reducing agents. The redox potential (ROP) is also directly linked to the metabolic routes of microorganisms, allowing that, through the control of the ROP, it is possible to redistribute the metabolic flow of different cells for the production of specific molecules (Verhagen et al. 2020).

In this aspect, to exemplify, Kulpreecha et al. (2009) cultivated B. megaterium and noticed that when increasing the concentration of dissolved O₂ from 60 to 80%, no significant difference in the concentration of P(3HB) was found, although cell growth was reduced. In addition, the authors pointed out that by reducing the amount of dissolved O₂ from 60 to 40%, the cultivation time was prolonged and productivity was drastically reduced.

In large-scale applications, it is convenient to carry out the production in bioreactors, considering that they adequately represent industrial reality. Therefore, several systems can be used, such as discontinuous (batch), fed-batch, semi-continuous, or continuous (Manowattana et al. 2018). Simple batch bioreactor systems present low risks of contamination, high flexibility of operation, allow the realization of successive phases in the same bioreactor, and require strict control of the conditions to guarantee the genetic stability of the microorganism (Mata-Gómez et al. 2014). The use of fed-batch cultures is driven by the search for high cell density and/or maximum formation of PHAs, and the concentrations of fed nutrients can be controlled by changes in the feeding rate while avoiding substrate inhibitions and/or limitations (Biglari et al. 2020). However, the fed-batch strategies require the development of a control system for analyzing medium residues and fresh nutrients addition, including strategies for oxygen demand (DO)-stat, pH-stat, and carbon source-stat (glucose-stat) (Divakar et al. 2017; Gahlawat and Srivastava 2017; Biglari et al. 2020).

Advances and challenges in the downstream of polyhydroxyalkanoates

The complete polyhydroxyalkanoate extraction process consists of several steps, the first of which may or may not be performed: the separation of cellular biomass with the intention of increasing polymer recovery by destabilizing and/or disrupting the microbial cell wall (Haque et al. 2022). The extraction system without previous treatment of the biomass begins with the separation of the solid material, composed of cells with intracellular biopolymers, from the culture broth, which is usually carried out by centrifugation. For treated biomass, this is the second step of extraction (De Donno et al. 2021). The next step is the recovery of the biopolymer using chemical, physical, enzymatic, and/or combined methods (Aramvash et al. 2018). In addition, it can be used to recover the biopolymer sedimentation (centrifugation) or precipitation process (De Donno et al. 2021). After the biosynthesis of polyester and of separation of cellular biomass, the process necessary for the recovery of PHAs constitutes another non-negligible cost factor, especially in large-scale production. The choice of the appropriate separation method of PHAs from residual biomass is dependent on several factors, such as the producing strain, the required purity of the product, the availability of isolation agents, and the impact on the molar mass that is acceptable (Policastro et al. 2021).

The main recovery techniques are solvent extraction, chemical, mechanical, and enzymatic disruption of cells, supercritical fluid extraction, ultrasound technology, genetically modified cell fragility, air classification, dissolved air flotation, or spontaneous release of PHA granules (Haque et al. 2022). Regarding costs, purity, and characteristics, all these methods have specific advantages and disadvantages that must be considered at the time of application.

However, downstream techniques are still challenging, mainly with the pretreatment of biomass and the use of solvents, to make the processes economically viable and environmentally sustainable. In this regard, the use of more environmentally friendly extraction solvents in the recovery stages should be further explored for the efficient replacement of synthetic plastic production (Haque et al. 2022). This study presents one review of the many methods of production and recovery of PHAs. Table 2 shows some PHA recovery strategies.

Table 2.

PHA extraction strategies using a solvent

Solvent type and extraction conditions	Microorganism/PHA (wt%)	Recovery (%)	Purity (%)	References
MeOH (50 °C, NaClO pretreatment)	PHB (98)	81	99	Aramvash et al. (2018)
PrOH (100 °C, NaClO pretreatment)	PHB (98)	28	97	Aramvash et al. (2018)
Butyl acetate (103 °C)	PHB (7)	96	98	Aramvash et al. (2015)
Ethylene carbonate (100 °C, NaClO pretreatment)	PHB (98)	90	99	Aramvash et al. (2018)
Dimethyl carbonate (90 °C)	PHB (74)	94	93	Samorì et al. (2015)
Acetone (120 °C)	P(4HB-HV) (88)	92	98	Koller et al. (2013)
Chloroform (ethanol pretreatment)	Raslthonia eutropha ATCC 17697	60	–	Vega-Castro et al. (2016)
Chloroform	Burkholderia sacchari DSM 17165	96	–	(Rosengart et al. (2015)
1,3-dioxolane	Cupriavidus necator A-04	90	–	Yabueng and Napathorn (2018)
Cyclohexanone dimethyl carbonate (ethanol pretreatment)	PHAs*	30	–	Abasi et al. (2022)
Dimethyl carbonate, chloroform, dichloromethane (1-butane pretreatment)	PHAs*	91–98	–	Reis et al. (2020)
Cyclohexanone (120 °C)	PHB (82)	99	99	Jiang et al. (2018)
Ionic liquid (1-Ethyl-3-Methylimidazolium Diethyl Phosphate)	Halomonas hydrothermalis	60	–	Dubey et al. (2018)
DMSO (150 °C)	PHB (98)	61	95	Aramvash et al. (2018)

*Unspecified o microorganisms

Solvent extraction techniques

The extraction technique of PHAs with organic solvents due to its simplicity is the most used, obtaining polymers purified and of low molar mass (Haque et al. 2022). The most commonly used halogenated solvents are dichloromethane, chloroform, 1,2-dichloroethane, 1,1,2,2-tetrachloroethane, and 1,1,2-trichloroethane (Table 2) (Koller et al. 2013; Pagliano et al. 2021).

Some processes have been used to improve extraction by halogenated solvents; the pretreatment method with chemical compounds and/or physical treatment (heating) is used to enhance cell breakdown, promote solvent percolation, and increase accessibility to the PHA granules. Normally, a high proportion of solvent is used for biomass, and this leads to an increase in costs. Pretreatments for solvent recovery are also required (Pagliano et al. 2021). López-Abelairas et al. (2015) used NaClO from microbial biomass and dichloromethane to recover 90% of PHA. The association of NaClO and alcohols (methanol or ethanol) on the precipitation of PHA provided extraction yields of up to 94% (up to 99% purity).

The main disadvantages of this method are the environmental impact and the possible toxicity of the polymer because of the presence of solvent residues. This also makes the use of solvent-extracted PHA unfeasible in the food and biomedical industries (Kosseva and Rusbandi 2018; Pagliano et al. 2021).

Some strategies have been conducted to replace conventional halogenated/chlorinated solvents with halogen-free solvents such as alkanes (e.g., pentane or hexane), alcohols (e.g., butanol or ethanol), cyclic carbonates and esters (e.g., carbonate dimethyl, ethers, ketones, amides, and organosulfur compounds. Some solvents are non-toxic to humans and have been designated “Generally Recognized as Safe” (GRAS), mainly because they are from renewable resources (e.g., butanol and ethanol) (Pagliano et al. 2021).

The alkanes (e.g., hexane) promote PHAs with high purity (> 89%) but low molecular weight (0.2 MDa) and yields (50%). Probably due to the low solubility of the polymer in these solvents; they are nonetheless beneficial in the reduction of endotoxins from Gram-negative bacteria, providing useful polymers for applications in the medical field (Pérez-Rivero et al. 2019). However, alkanes (e.g., hexane and pentane) are not considered “green” solvents, being “hazardous” compounds, with restrictions on their use in scale-up (Prat et al. 2016).

The esters (e.g., butyl and ethyl acetate) are solvents considered beneficial to health and the environment. However, extractions are generally made at temperatures near 100 °C (Prat et al. 2016). Normally, ketones are less efficient than esters and alcohols and usually also require high temperatures (e.g., treatment with cyclohexanone at 120 °C/3 h recovery of 99% of PHA) (Jiang et al. 2018).

Carbonates (e.g., dimethyl and ethylene carbonate) are non-toxic, biodegradable, and considered efficient in extraction, obtaining polymers with high purity (Aramvash et al. 2018). Although they present limitations on industrial scale; they are, therefore, indicated as potential substitutes for other solvents (ketones, esters, chlorinated, and others), for example. Reis et al. (2020) using dimethyl carbonate obtained results compatible with conventional chlorinated solvents, of 39, 37.5, and 31.7 g of PHA/100 g of the cell when using dichloromethane, chloroform, and dimethyl carbonate, respectively, and when associating 1-butanol. the purification went from 91.2 to 98%.

Other solvents are used for the extraction of PHAs, such as anisole, dimethyl sulfoxide (DMSO), and dimethyl formamide (DMF), but to increase the recovery, it is necessary to use an antisolvent, such as ethanol. Ionic liquids were also used (e.g., 1-ethyl-3-methylimidazolium diethyl or dimethyl phosphate), although with low purity 30–86% (Dubey et al. 2018), considering the ionic liquids more with the function of disaggregating cellular biomass and promoting the solubilization of PHAs (Pagliano et al. 2021).

From a scale-up perspective, green solvents (carbonates, ketones, or esters) can be viable and sustainable alternatives and be recommended for recovery polymers (Vogli et al. 2020; Pagliano et al. 2021).

Cell lysis methods

For cell lysis, some alternatives for PHA recovery can be chemicals with oxidizing, alkaline, or acidic functions, surfactants, and/or enzyme methods. Also recommended are mechanical methods, ultrasonic systems, supercritical fluids, genetically induced cell lysis, and gamma irradiation (Table 3).

Table 3.

PHA recovery strategies using chemical compounds, enzymes, physical treatment, and/or combined cell lysis systems

Methods (conditions)	Microorganism/PHA (wt%)	Recovery (%)	Purity (%)	References
Chemical compound
NH4OH	P(HB-HV)*	73	–	Maninna et al. (2019)
NaClO (37 °C, 1.9 M)	PHB (65)	> 80	99	López-Abelairas et al. (2015)
NaClO	Ralstonia eutropha H16	87	–	Heinrich et al. (2012)
NH4OH (90 °C, 0.1 M)	P(HB-HV)	75	70	Samorì et al. (2015)
NaOH (37 °C, 0.5 M)	PHB (65)	< 80	> 90	López-Abelairas et al. (2015)
H2SO4 (80 °C)	PHB (65)	80	99	López-Abelairas et al. (2015)
Acetic acid (100 °C, NaClO pretreatment)	PHB (98)	37	97	Aramvash et al. (2018)
Lysol and NaOH	P(3HB-4HB)	98	97	Irdahayu et al. (2017)
Sodium dodecyl sulfate	PHA (70)	81	90	Yang et al. (2011)
Polyethylene glycol 8000 and sodium sulfate	R. eutropha H16	65	–	Leong et al. (2015)
Ethylene oxide and propylene oxide	Cupriavidus necator H16	94.8	–	Leong et al. (2017)
Enzymes
Protease from A. oryzae (48 °C, heat pretreatment at 85 °C)	P(HB-HV) (79)b	–	99	Kachrimanidou et al. (2016)
Trypsin (50 °C, heat pretreatment at 85 °C)	PHB (75)a	88	53	Kapritchkoff et al. (2006)
Bromelain (50 °C, heat pretreatment at 85 °C)	PHB (75)a	59	89	Kapritchkoff et al. (2006)
Trypsin and Bromelain (heat pretreatment at 85 °C)	PHB (75)a	91	83	Kapritchkoff et al. (2006)
Pancreatin (50 °C, heat pretreatment at 85 °C)	PHB (75)a	90	62	Kapritchkoff et al. (2006)
Phospholipase- Lecitase 100S (40 °C, heat pretreatment at 100 °C)	PHB (52)b	65	–	Holmes and Lim (1990)
Glycosidase: Celumax (60 °C, heat pretreatment at 120 °C)	P(HB-HV) (75)a	86	–	Neves and Müller (2012)
Physical and/or combined treatment
Ultrasonication and propylene carbonate	Cupriavidus necator DSM 545	92	–	Quines et al. (2017)
Ultrasonication	C. necator	90	–	Deshmukh et al. (2020)
Gamma irradiation	Bacillus flexus	54	–	Khattab et al. (2021)
Maceration with spheres	C. necator Haloferax mediterranei	95	–	Aramvash et al. (2018) and Gutt et al. (2016)
Genetically induced	Alcanivorax borkumensis SK2 Pseudomonas putida CA-3 Bacillus megaterium	90	–	Zhou et al. (2020) and Kurian and Das (2021)
Supercritical technology with CO2	B. megaterium ATCC 14581	73–75	–	Daly et al. (2018) and Reato et al. (2021)
Pretreatment for cell fragility	H. mediterranei Azotobacter vinelandii	90	–	Koller (2020) and Umesh et al. (2021)

*Unspecified o microorganisms

^aPHA recovered by solvent extraction

^bPHA recovered by centrifugation

The oxidant compound sodium hypochlorite (NaClO) is capable of promoting cell lysis and solubilizing non-PHA material, obtaining high purity (99%), and yielding greater than 80% of PHA (López-Abelairas et al. 2015). The disadvantage of these compounds is that they are considered toxic, limiting their application, and when used in high concentrations, they may change the molecular weight of the polymer (Koller et al. 2013).

Acids (e.g., hydrochloric or sulfuric) are also used to perform the digestion of cellular material potential. But they also have disadvantages: high acid concentrations may degrade the PHAs and change their mechanical properties, such as a reduction in their molecular weight and slight changes in their tensile strength rating. Another relevant point is that this method requires a post-treatment to remove the acid residue, which may remain present after polymer recovery (Kosseva and Rusbandi 2018). For example, H₂SO₄ or HCl promotes high recoveries (> 90%) and purity (99%), but reduces the molecular weight (0.06 MDa) (Yu and Chen 2006).

Alkalis, (e.g., NaOH and KOH), act on the lipid layer of the microorganism cells, causing saponification, increasing the permeability, and providing the release of non-PHA fractions (Kosseva and Rusbandi 2018; Mannina et al. 2020; Kurian and Das 2021). Table 3 shows some studies using alkali for cellular lysing.

Surfactants are initially incorporated into the lipid layer, causing the rupture of the membrane, forming phospholipids, solubilizing proteins and molecules of non-polymeric cell matter, and finally releasing PHA (Pérez-Rivero et al. 2019; Mannina et al. 2020).

The most commonly used surfactants, such as alkylbenzene sulfonates or sodium dodecyl sulfate, when associated with other compounds or alone (e.g., NaClO and NaOH), increase the purity and yield of polymer recovery (Pérez-Rivero et al. 2019). Cationic surfactants such as benzalkonium chloride, hexadecyltrimethylammonium bromide, or palmitoyl carnitine are compounds less used than the anionic ones. Also can be used as non-ionic surfactant (e.g., Tween-20 and Triton X-100) (Pagliano et al. 2021). Cell lysis methods using alkaline or surfactant compounds present lower costs (1.02–5.23 €/kg PHA) than solvent (1.95–6.61 €/kg PHA) (Pagliano et al. 2021).

Cell disruption can be performed with enzymes, for example, proteases, nucleases, lysozymes, and lipases, which have hydrolytic effects on cellular membranes. The most commonly used protease is the alcalase (subtilisin A), with high PHA recovery (> 90%) in the optimum enzyme temperature range (45–65 °C). Other proteases such as bromelain and trypsin provide good polymer recoveries (90%). It is recommended the use of cocktails of enzymes (proteases, lysozymes, nucleases, phospholipases, and others), together with surfactants and chelating agents to improve the recovery of PHA (Pagliano et al. 2021).

The use of proteolytic enzymes for biomass dissolution is advantageous, as it significantly reduces the degradation of PHAs because of their selectivity, low energy cost, and mild operating conditions. Enzymes can thus initiate cell lysis for the subsequent recovery of polymers. After releasing the PHA cells and the primary purification step, often additional steps must be taken to obtain products of sufficient purity, for example, treatments with hydrogen peroxide associated with enzymatic breaking processes or chelating agents, treatment with ozone, or mixing with other polymers, followed by additional solvent extraction (Suzuki et al. 2008). Other chemicals, such as ethylenediamine tetraacetic acid (EDTA) and sodium dodecyl sulfate (SDS), along with enzymes, can eliminate the use of pre-heat treatment in the process (Israni et al. 2018). Centrifugation and membrane filtration (ultrafiltration and diafiltration) are also used (Kathiraser et al. 2007). The enzymes have a high recovery rate and a wide purity range (88.8–97%), so the process relies heavily on purified enzymes. Therefore, the use of enzymes in scale-up has to take the cost–benefit ratio into account (Kurian and Das 2021).

The most mechanical methods used for cell rupture are high-pressure homogenization, ball milling, and/or associated with liquid nitrogen. In general, mechanical rupture is favored because of the little damage it causes to products and the environment and because it does not involve the use of any chemical product. The disadvantages of this method include a lengthy processing time, a high capital investment cost, and difficulty scaling (Gutt et al. 2016; Aramvash et al. 2018).

The ultrasound bath is one of the alternatives for the recovery of PHA. This tool helps to accelerate the extraction and contributes to the recovery and purity of the biopolymer (Pradhan et al. 2017, 2018). This technology is usually employed in conjunction with solvent extraction. It allows for increasing the mass transfer rate in the system, making the solvent more easily permeate the cell membrane (Kosseva and Rusbandi 2018). Another important point is the fact that ultrasound technology causes negligible changes in the mechanical and structural characteristics of PHAs. However, the technique can also be applied to remove toxins from the medium, something essential for the use of these PHAs by the pharmaceutical and food industries (Kosseva and Rusbandi 2018).

The application of supercritical technology has become attractive to the industry because of its non-toxicity and recovery potential. The main fluid used is carbon dioxide (CO₂), which is efficient in extracting lipids and other hydrophobic compounds (Reato et al. 2021). Supercritical CO₂ (scCO₂) is considered a convenient and safe solvent because of its non-toxicity, non-flammability, and low reactivity. In addition, after extraction, it evaporates completely without leaving residues, thus avoiding subsequent drying (Darani and Mozafari 2009). Since numerous hydrophobic compounds are highly soluble in CO₂, it is used in the extraction of PHA (Koller et al. 2013; Daly et al. 2018; Reato et al. 2021).

Conclusions

The production of bioplastics is necessary to solve the environmental problems that arise from conventional plastics of petrochemical origin. PHAs can play an important role because they have properties (biodegradability, biocompatibility, thermostability, and resilience) desired by the market and can be produced from different alternative and low-cost substrates (for example, agro-industrial co-products, organic waste, and effluents, among others).

However, the production and recovery of PHAs on a large scale is still a major challenge due to their high production costs and difficulty controlling their physical properties. Therefore, this study may motivate and guide future studies focusing on economically viable and environmentally sustainable production and recovery of PHAs.

Author contributions

RNdM and GdSH wrote the main text of the manuscript. JS, AJ, and EV prepared the tables and edited and revised the manuscript. All authors read and approved the final manuscript.

Funding

This study was financed in part by the National Council for Scientific and Technological Development—Brazil (CNPq)—project number 431493/2018-9, the Coordination for the Improvement of Higher Education Personnel—Brazil (CAPES)—Finance Code 001, and the Research Support Foundation of the State of Rio Grande of Sul—Brazil (FAPERGS).

Data availability

Data supporting the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.