Bioconversion of cassava starch by-product into Bacillus and related bacteria polyhydroxyalkanoates

Cristhiane L. Krueger¹ · Claudemir M. Radetski*² · Amanda G. Bendia² · Ida M. Oliveira² · Marcus A. Castro-Silva² · Carlos R. Rambo³ · Regina V. Antonio⁴ · André O.S. Lima*²

¹ Universidade Federal de Santa Catarina, Campus Universitário Trindade, Departamento de Engenharia Química e Engenharia de Alimentos, Florianópolis, SC, Brazil
² Universidade do Vale do Itajaí, Centro de Ciências Tecnológicas da Terra e do Mar, Itajaí, SC, Brazil
³ Universidade Federal de Santa Catarina, Campus Universitário Trindade, Departamento de Engenharia Elétrica, Florianópolis, SC, Brazil
⁴ Universidade Federal de Santa Catarina, Campus Universitário Trindade, Departamento de Bioquímica, Florianópolis, SC, Brazil

*Corresponding authors: radetski@univali.br; andreolima@gmail.com

Financial support: The Brazilian National Council for Scientific and Technological Development (CNPq) is acknowledged for fellowships N# 300898/2007-0 (Claudemir M. Radetski) and financial support (Regina V. Antonio). Cristhiane L. Krueger acknowledges the Brazilian Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for fellowship. Amanda G. Bendia and Ida M. Oliveira acknowledge Indústria Corn Products do Brasil for fellowships.

Keywords: Bacillus megaterium, biodegradable plastics, biopolymer, MTT assay, starch substrates.

Background: Unlike petroleum-based synthetic plastics, biodegradable biopolymer generation from industrial residue is a key strategy to reduce costs in the production process, as well as in the waste management, since efficient industrial wastewater treatment could be costly. In this context, the present work describes the prospection and use of bacterial strains capable to bioconvert cassava starch by-product into biodegradable polyhydroxyalkanoates (PHAs). Results: The first step of this study was the bacterial competence screening which was conducted with 72 strains covering 21 Bacillus and related species. The microorganism growth in a medium with a starch substrate was measured by an innovative MTT assay, while the ability of the bacteria to secrete amylase and produce PHA was evaluated by the Nile Red Dye method. Based on growth and potential for PHA production, four isolates were selected and identified as Bacillus megaterium by 16S rRNA sequencing. When cultivated in hydrolyzed cassava starch by-product, maximum production reached 4.97 g dry biomass/L with 29.7% of Poly-(3-hydroxybutyrate) (characterized by FTIR). Conclusions: MTT assay proved to be a reliable methodology for monitoring bacterial growth in insoluble media. Selected amylolytic strains could be used as an alternative industrial process for biodegradable plastics production from starchy residues, reducing costs for biodegradable biopolymer production and wastewater treatment operations.

Polymeric materials are important for society due to their characteristics of versatility, durability and resistance to degradation. However, there are many environmental problems resulting from the use of petrochemical derived polymers making necessary to find alternatives to replace these biorecalcitrant materials. One solution to this problem is the use of biodegradable polymers, which are polymers with the desirable properties of conventional plastics but also with rapid biodegradation in the environment after discarding (Sankhla et al. 2010). One such biodegradable type of polymer are polyhydroxyalkanoates (PHAs), which are polyesters containing hydroxyalkanoic acid monomers which can be biosynthesized from renewable (waste) resources (Mengmeng et al. 2009). PHAs are accumulated intracellularly by bacteria as carbon or energy reserves, mainly when there is a limitation of an essential nutrient, such as nitrogen, phosphorus, oxygen or sulphur (Verlinden et al. 2007).

Due to their chemical and physical characteristics, PHAs have a wide range of applications including conventional plastic materials, utensils, and cosmetic containers, as well as in medicine (as bone plates, surgical sutures and blood vessel replacements), in the pharmaceutical and food industries, and as biodegradable carriers for the long-term dosage of drugs, medicines, hormones, insecticides and herbicides (Reddy et al. 2003). The most studied PHA is the poly-(3-hydroxybutirate), P(3HB), which exhibits high crystallinity and higher resistance to UV than polypropylene.

The commercial use of PHAs is, however, limited mainly due to its relatively high production cost. Therefore, the use of low cost substrates is an important factor, along with the need to obtain strains for efficient substrate to polymer conversion (Khardenavis et al. 2007). Low cost substrates include starchy substrates, which are important compounds in food industry by-products. Also, agro-based industrial wastewaters, activated sludge, paper mill wastewater, crude glycerol waste stream, and waste cooking oil can be a suitable alternative as low cost substrate and biomass respectively for PHA production (Beun et al. 2000; Bengtsson et al. 2008; Costa et al. 2009; Chakravarty et al. 2010; Dobroth et al. 2011). Cassava is widely cultivated in the world, which implies that a considerable amount of by-products is produced, and an appropriate destination for these by-products must be found to minimize environmental impacts. According to Boonnop et al. (2009), typical cassava starch composition is 68.9% of dry matter (essentially carbohydrates) which can be detailed as follow: 95.9% of organic matter, 2.8% of protein, and 2.3% of fat matter. In relation to the limitations of cassava starch by-product as a substrate for PHA producing microorganisms, the most important parameter is the water solubility. Thus, PHA yield can be limited by amylase hydrolytic activity from microorganisms. Alternatively, chemical hydrolysis (alkaline or acid), high pressure or temperature could increase carbohydrate release from cellulosic material and increase PHA production.

One group of environmental microorganisms with great biodegradation potential is the Bacillus which is genetically and biochemically well understood and offers several advantages, such as, short generation time, attractive host for gene expression, it can easily grow to very high cell density using inexpensive carbon and nitrogen sources, and capability of secreting large amounts of enzymes (Law et al. 2003). With respect to the bacterial growth determination in turbid medium, most available methodologies are laborious and time-consuming. In this respect, available methodologies for bacterial growth determination considered as laborious and time-consuming for insoluble substrates are the microscopic counts or culture and colony counts that represent only the viable cells. In relation to soluble substrates, these methodologies can be resumed as assessing cell density by spectrophotometry or determination of cellular mass. In the first case a calibration curve is necessary to establish the correspondence between the cellular density and the number of cells, while in the second case, the invasive character of the sampling method must be considered. In other words, acceptable quantification implies a reasonable mass sampled, which may hinder the removal of aliquots from a single vial. An alternative method to measure bacterial growth is the use of the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, which is widely used for in vitro measurement of metabolic viability of cell cultures subjected to different culture conditions (Sieuwerts et al. 1995). This assay is based on the ability of viable cells to produce formazan, which can be measured by visible spectrophotometry. Thus, the objectives of this study were to select a bacterial strain capable of growing and producing PHA from a cassava starch by-product as the carbon source, to identify selected microorganisms by molecular methods, and to assess the applicability of the MTT assay for the bacterial growth determination.

Bacterial strains and culture media

The collection of 72 strains evaluated was obtained in the Applied Microbiology Laboratory (UNIVALI, SC, Brazil) (Table 1). The strains were isolated from environmental samples and classified by morphological, growth and biochemical characteristics, according to a published methodology (Logan and De Vos, 2009). For the culture maintenance, the microorganisms were grown in M9 media, which contained (per liter of distilled water): 6.78 g Na₂HPO₄·7H₂O; 3.0 g KH₂PO₄; 0.5 g NaCl; 1.0 g NH₄Cl, enriched with pure starch or cassava starch by-product (5 g/L), to evaluate the growth of the strains. In the assaying step, two pre-cultures were used: one in Luria-Bertani (LB) medium (per liter of distilled water: 10 g tryptone; 5 g yeast extract; 10 g NaCl), and the other in the medium described by Ramsay et al. (1990) without nutrient limitation (per liter of distilled water: 6.7 g Na₂HPO₄·7H₂O; 1.5 g KH₂PO₄; 1.0 g (NH₄)₂SO₄; 0.2 g MgSO₄·7H₂O; 0.01 g CaCl₂; 0.06 g ferrous ammonium citrate; 1 mL trace-element solution. Each liter of trace-element solution in distilled water contained 0.3 g H₃BO₃; 0.2 g CoCl₂·6 H₂O; 0.1 g ZnSO₄·7H₂O; 0.03 g MnCl₂·4H₂O; 0.03 g NaMoO₄·2H₂O; 0.02 g NiCl₂·6H₂O; 0.01 g CuSO₄·5H₂O), enriched with cassava starch by-product (20 g/L). For PHA production assays, the medium described by Ramsay et al. (1990) with nutrient limitation (the same as that described above, except with 2.5 g/L Na₂HPO₄·7H₂O and 0.83 g/L KH₂PO₄) and cassava starch by-product (20 g/L) were used.

Selection of amylolytic microorganisms

The different measured parameters for the selection of microorganisms for the PHA production were based in their ability to grow in medium containing the residue as a source of carbon (diameter of the colony and MTT absorbance assays) and their ability to produce amylase to degrade the substrate (diameter of the halo and enzymatic index). Thus, the 72 strains were inoculated in solid M9 media with 5 g/L of commercial starch and were cultivated for 48 hrs (37ºC). The strain growth was evaluated by visual comparison and classified into: i) without, ii) poor or iii) vigorous growth. The strains with vigorous growth were selected and inoculated in solid M9 media with 5 g/L of cassava starch by-product at four points on each plate, and cultivated for 24 hrs (37ºC). These plates were stained with iodine solution (1% in ethanol), and the colony diameter and starch degradation halo were measured. The enzymatic index was calculated from the ratio between the average halo and colony diameters, to evaluate the starch degradation capacity through sugar consumption. The differences among treatments were analyzed using ANOVA (P ≤ 0.05) and the Tukey’s test (P ≤ 0.05).

Determination of bacterial growth by MTT assay

The microorganisms were cultivated in two pre-cultures: 5 mL of LB medium for 24 hrs (37ºC, 150 rpm), followed by transfer to 50 mL of the medium described by Ramsay et al. (1990) without limitation, and 20 g/L of cassava starch by-product for 24 hrs. A 10% solution of pre-culture was transferred to 300 mL of limited medium with 20 g/L of cassava starch by-product and cultivated for 48 hrs. Bacterial growth was determined by MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) (Freshney, 2005). The culture was diluted (1:6) in medium without limitation of starch by-product, with six replicates, and 10 µl of MTT (5 mg/mL diluted in PBS buffer) were added followed by incubation for 1 hr (37ºC). Subsequently, 100 µL of 10% SDS in 0.01 M HCl were added and the mixture was incubated for 18 hrs at 25ºC. Spectrophotometry measurements were carried out at 540 nm. To compare the viable bacteria cells with the MTT assay response, the microorganisms from the medium containing cassava by-product were inoculated at different times in a Petri dish (LB medium). Aliquots were sampled after different time periods for unity forming colony (UFC) count and absorbance measurement.

PHA production and characterization

For qualitative detection of PHA production, cells were cultivated with Nile Red Dye (0.5 µg/mL diluted in DMSO), which stains PHA granules, as described for Jendrossek et al. (2007), and observed under an epifluorescence microscope (Olympus BX40). For PHA production and quantification strains were cultivated (48 hrs, 150 rpm at 37ºC) in hydrolyzed starch cassava by-product. Although the selected microorganisms showed ability to grow in the medium of raw starch cassava by-product, the final assays were performed with hydrolyzed products for two reasons: i) it would be difficult to determine the bacteria ability to produce PHA using crude insoluble substrate, since the unmetabolized substrate fraction can interfere with the polymer quantification, and ii) in previous tests was observed under microscope that two of isolates showed no precipitate of Nile Red Dye inside the cells (i.e., no PHA accumulation) under the tested conditions. For this purpose appropriated amount of by-product were hydrolyzed at 121ºC for 30 min with 1% HCl solution (Law et al. 2003; Dalcanton et al. 2010). Hydrolysis of cassava starch by-product using this methodology released about 5% (m/m) in sugar (glucose-quantified by standard DNS procedure). After cells growth, they were recovered by centrifugation (1,800 g; 10 min; 20ºC; Eppendorf 5810R centrifuge ), freezed (-20ºC) and lyophilized for 5 hrs in a Jouan LP3 lyophilizer. Samples were then stored in a freezer (-20ºC) for further analysis. The accumulated PHA was determined by gas chromatography, as described by Chia et al. (2010). PHA was extracted from lyophilized cells by the method described by Shishatskaya and Volova (2004). The characterization of PHA was carried out by Fourier transform infrared spectroscopy (FTIR) (Perkin Elmer).

The 16S ribosomal RNA gene amplification and analysis

Genomic DNA from Bacillus isolates was obtained (DNeasy Blood & Tissue Kit, Qiagen - MD) from overnight culture cells grown in LB medium (37ºC, 150 rpm) and used as a DNA template for the PCR reactions (Sambrook and Russell, 2001). The 16S rRNA gene fragment (1.5 Kb) was amplified by PCR using primers (F-5’-AGAGTTTGATCCTGGCTCAG-3’; R-5’-AAGGAGGTGATCCAGCCGCA-3’) as described by Hiraishi (1992). Amplification reaction mixtures (20 μL) consisted of 0.2 mM of each dNTP, 0.5 μM of each primer, 1 U of Platinum Taq DNA polymerase (Invitrogen, Brazil), 1X Taq polymerase buffer, 5 mM of MgCl2, 5% DMSO and 50 ng of DNA. The thermal cycling conditions consisted of one cycle at 94ºC for 5 min followed by 35 cycles of 30 sec at 94ºC, 30 sec at 58.5ºC, and 1.75 min at 72ºC, and a final extension step of 4 min at 72ºC. 16S rRNA amplicons were sequenced at Macrogen (Seoul, Korea) using standard protocols with dye terminator and capillary sequencer (ABI3730XL, Applied Biosystems, CA). For this purpose, each PCR product was cloned in pTZ57R/T using the InsTAclone™ PCR Cloning Kit (Fermentas, MD) and the recombinant plasmid introduced into chemically competent Escherichia coli DH5α cells (Sambrook and Russell, 2001). Cells were spread onto solid LB supplemented with ampicillin (100 μg/mL) and X-gal (0.02 mg/mL 5-bromo-4-chloro-3-indolyl b-D-galactopyranoside (Invitrogen, CA). White transformant clones were selected and vector isolated (PureLink™ Quick Plasmid Miniprep Kit, Invitrogen, CA). Insert was sequenced using universal M13 primers, as well as a 16S rRNA internal primer (5’-GGATGACAGTACCTGAAGAATAAGCAC-3’). Sequences were deposited in the GenBank at NCBI. The homologies of the determined sequences of 16S rRNA gene were compared with full length sequences in the GenBank database by the BLAST method (Camacho et al. 2008).

Selection of amylolytic microorganisms and determination of enzymatic index

From the 72 strains inoculated in pure starch as the sole source of carbon, 55 were able to grow in this substrate with varied growth rates classified as poor or vigorous. When these 55 selected strains were inoculated in pure starch, 16 bacterial strains exhibited a higher growth rate after 24 hrs of culture. These 16 bacterial strains were then selected and their amylolytic potential for the use of cassava starch by-products was evaluated. From these 16 strains, 14 visually showed vigorous colony formation (data not shown) and were selected for the next selection step. The enzymatic index (ratio between average halo and colony diameters) was determined to verify the capacity of the enzyme secretion to hydrolyze the starch of the culture media (Table 2). Statistical analysis (Tukey test) of the enzymatic index showed significant variation among the different indexes.

The data in Table 2 show that the colony diameter of strain LAMA23 showed the highest value, whereas for the halo diameter strains LAMA23 and LAMA98 showed the highest values. Also, it can be observed that B. ehimensis (LAMA190), G. kaustophilus (LAMA098) and P. thiaminolyticus (LAMA133) had small colonies but high enzyme secretion levels, showing no direct relation between colony and halo diameters. Additionally, the results of Table 2 revealed that all microorganisms tested are able to secrete amylolytic enzymes at varied concentrations.

Thus, these results showed that the cassava starch by-product substrate enabled the microorganisms to grow in a minimal composition medium, and to secrete the enzymes responsible for PHA biosynthesis. This is a key feature for application in the industrial recycling processes, providing a rational way to reduce the environmental discharge of this by-product (Gupta et al. 2003). According to Reddy et al. (2003), three biochemical reactions are implicated in the PHA synthesis. In the first reaction, there is a condensation of two molecules of acetyl-coenzyme A (acetyl-CoA) to acetoacetyl-CoA (β-ketothiolase enzyme), followed by reduction of acetoacetyl-CoA to hydroxybutyryl-CoA (enzyme dependent of NADPH-acetoacetyl-CoA reductase), and finally, hydroxybutyryl-CoA monomers are polymerized to PHA (PHA-polymerase enzyme).

Determination of bacterial growth by the MTT assay

Bacterial growth of viable cells in this experimental context was determined by MTT assay. To verify the feasibility of this assay, a previous experiment was carried out with the strain LAMA 262 (selected randomly) grown in both raw and hydrolyzed cassava by-product.

The MTT assay results indicated no difference at the 5% level of significance between the microorganisms grown in the raw and hydrolyzed cassava by-product. Besides demonstrating the MTT assay feasibility, this finding confirms the versatility of the microorganisms to secrete all of the enzymes necessary for growth in a medium comprising this industrial residue. Therefore, the following tests were carried out with raw cassava by-product, considering minimization of the cost of PHA production on an industrial scale. The 14 selected bacterial strains were cultured in a medium containing raw cassava by-product and the results of the MTT assay for cellular growth are shown in Figure 1. It is possible to verify that some selected strains exhibited a vigorous growth in the limited medium composed by cassava starch by-product according to the MTT assay, except LAMA 261, which showed a lower growth potential than the other strains.

As MTT assay is proposed here as a new method for bacterial growth determination in insoluble media. Figure 2 show the relationship between formazan production and the number of viable cells of four strains cultivated in M9 media supplemented with glucose or cassava residue. A correlation of 0.97 between MTT reads and units forming colony average points was determined. This result clearly indicates the applicability of MTT for growth monitoring purposes.

In this study, the determination of bacterial growth in a turbid medium could be performed only by means of culture analysis and microscopy microorganism counts, which are laborious and time-consuming. The MTT assay is rapid and easy to perform, although it can be significantly affected by a number of conditions (Sieuwerts et al. 1995). In this respect, MTT assay can be affected directly by: i) the sample dilution, since compound incubated with cells must be reduced by succinate dehydrogenase prior to spectrophotometric measurement; ii) exposure time of MTT to the cells, which could be variable according to the cell physiology. Furthermore, MTT solution conservation is important to preserve reaction effectiveness. Thus, this compound must be stored in a freezer, as well as the solution prepared in PBS buffer (5 mg/mL). In our experiments the MTT assay was found to be a reliable methodology to measure bacterial growth rates, without interference due to the culture medium or the industrial residue composition.

Production, purification and characterization of PHA

To evaluate the PHA production capacity, two microorganisms LAMA262 and LAMA268 (Bacillus megaterium and Paenibacillus macquariensis) were selected from MTT assay and were inoculated in limited medium containing 20 g/L of pure glucose or 20 g/L of raw cassava starch by-product, both with the addition of Nile Red Dye. As a positive control, a recombinant strain of E. coli (DH10B) transformed with plasmid pRLC2, which contains the genes for PHA production (gene phaC of Chromobacterium violaceum and genes phbA and phbB of Cupriavidus necator, all regulated by Lac promoter of the plasmid) was used. After culture, observation under an epifluorescence microscope indicated that the strains grew in both carbon sources; however, they did not produce PHA from the cassava by-products media (data not shown), perhaps the few starch hydrolysis products were used to the cellular multiplication, only.

In order to identify which among the 14 pre-selected amylolytic strains are the most PHA producers, they were cultivated in rich glucose (20 g/L) complex media (10 g/L tryptone; 5 g/L yeast extract; 10 g/L NaCl; 2.5 g/L Na₂HPO₄·7H₂O and 0.83 g/L KH₂PO₄) and hydrolyzed cassava by-product (20% of hydrolyzed cassava (v/v), equivalent to 2 g of glucose/L media). After 48 hrs of incubation, the cultures were observed under an epifluorescence microscope and the results showed that LAMA073, LAMA262 and LAMA265 produced a higher degree of granule fluorescence than the other bacterial strains (Figure 3).

The four selected strains (LAMA095, LAMA073, LAMA262 and LAMA265) were then cultivated in pure glucose and cassava starch by-product in order to quantify the enzyme biomass and polyhydroxybutyrate [P(3HB)] polymer production. Through methanolysis and using benzoic acid as the internal standard, samples were analyzed by gas chromatography (Table 3).

According to Table 3, the LAMA095 strain produced the highest biomass amount when cultivated in the cassava starch by-product. However, LAMA073 produced a higher amount of P(3HB) when cultivated in glucose, where the polymer corresponds to approximately 41.95 wt.% of its biomass. Interestingly, LAMA265 showed a yield of 25% (w/w dry cell) in the production of P(3HB) when cultivated with 20% of hydrolyzed cassava starch by-product (v/v), which increased to 31.6% when cultivation was carried out in pure glucose. Nevertheless, the LAMA095 strain should be highlighted because it showed a higher biomass production when cultivated with cassava starch by-product than when cultivated with pure glucose. This strain produced the highest amount of P(3HB), while the polymer recuperation percentage was similar for the two substrates (24.1% of P(3HB) for glucose culture, and 23.9% for hydrolyzed cassava starch by-product).

To improve the P(3HB) production yield, the proportion of hydrolyzed cassava starch by-product in the culture medium was increased to 40% (v/v), equivalent to 4 g/L of sugar. As expected, an increase in the strain biomass and in P(3HB) production were observed when the carbon source was increased in the medium, most notably for the strain LAMA095, which reached 1.48 g P(3HB)/L. The highest P(3HB) yield was achieved by the LAMA073 strain, 30.4% (w/w dry cell), which is 57% higher than the amount produced by the strain cultivated in the medium with 20% hydrolyzed cassava starch by-product.

The percentage of P(3HB) observed in this study (around 30% based on w/w dry cells) can be compared with results of other studies available in the literature. Dalcanton et al. (2010) obtained 35% of PHB (w/w dry cell) when Ralstonia eutropha (recently renamed Cupriavidus necator) was cultivated in hydrolyzed rice processing by-product. Recombinant E. coli showed 34.8% yield of PHB (w/w dry cell) when this bacterium was cultivated in hydrolyzed corn starch (Fonseca et al. 2008), which is the same magnitude found in the present work. Better results were obtained by Zhang et al. (1994) who cultivated recombinant E. coli and Klebsiella aerogenes in 6% of cane molasses. These authors showed that the bacteria accumulated 45% of P(3HB) (w/w dry cell).

With regard to Bacillus studies, Gouda et al. (2001) cultivated B. megaterium in cane molasses, obtaining a yield of 46.2% of P(3HB) (w/w dry cell), similar to the yield obtained by Omar et al. (2001) (around 50%) using beet molasses and the same bacterial species. In another study, Law et al. (2003) carried out assays with recombinant B. subtilis cultivated in hydrolyzed malt waste and obtained 15.3% (w/w dry cell) of yield, while B. megaterium CQPBA036-07 DMP01 cultivated in 50 g/L of sucrose yielded 75% of PHB (w/w dry cell) (Faccin et al. 2009). More recently, Kulpreecha et al. (2009) achieved a yield of 61.6% of PHB (w/w dry cell) cultivating B. megaterium BA-019 for 12 hrs. The production of polyhydroxybutyrate (PHB) using agro-industrial residues as the carbon source has also been studied using seven substrates, viz., wheat bran, potato starch, sesame oil cake, groundnut oil cake, cassava powder, jackfruit seed powder and corn flour (Ramadas et al. 2009). These residues were hydrolyzed using commercial enzymes and the hydrolysates assessed to select the best substrate for PHB production. Jackfruit seed powder allowed the maximum production of PHB under submerged fermentation using Bacillus sphaericus (19%) at an initial pH of 7.5 (Ramadas et al. 2009). These literature data show that the P(3HB) production yield presented in this study is acceptable, despite the poor medium content and low cost substrates used in our experiments. In this regard, it is important to note that the quantity of hydrolyzed cassava starch by-product used in these experiments was lower than the quantity of glucose, and thus it may be possible to generate more significant results using equivalent quantities of substrate. Also, the P(3HB) yield can be increased in purified/hydrolyzed substrate raw material.

Table 4 shows the molecular sequencing of four selected bacterial strains based on the 16S ribosomal RNA gene amplification and a comparison of the results with NCBI GenBank data. The segments isolated from the strains LAMA073, LAMA095, LAMA262 and LAMA265 were compared to nucleotides accessed in GenBank. After comparison, the bacterial strains were identified as B. megaterium and registered in the Genbank (NCBI) under the following access numbers: LAMA 073 - HM104230, LAMA 095 - HM104231, LAMA 262 - HM104232 and LAMA 265 - HM104233. Although we obtained the same identification for the four strains, it is clear that this species presents great diversity and plasticity concerning its metabolism, which can be exploited in several industrial processes including, more recently, the recycling waste industry.

Figure 4 shows the FTIR spectra for the P(3HB) produced by LAMA095 (Figure 4a) and LAMA265 (Figure 4b). Both spectra exhibited the main characteristic bands of PHAs, which confirmed the presence of the polymer: an intense band at 1,726 cm^-1, related to carbonyl stretching, which is typical of PHA (Hong et al. 1999). Several other bands with varied intensity, between 1,450 and 1,000 cm^-1, are also present, which are attributed to deformation of methyl and methylene groups and due to C-O-C stretching. It is important to point out that the exact wave number of the peaks varies with the crystallinity and with the chain length of the PHA (Hong et al. 1999; Kansiz et al. 2000). These results are in agreement with the P(3HB) characterization data available in the literature (Kansiz et al. 2000).

The cassava processing by-product possesses high levels of starch and reducing sugar when hydrolyzed, which will generate an environmental imbalance if discarded to the environment without prior treatment (FAO, 2001). In this perspective, production of PHA from cassava starch residues meets with two goals: i) recycling of cassava processing waste to give an appropriate destination for this industrial by-product, and ii) development of high value polymeric products for use in areas such as medicine, agriculture, tissue engineering, nanocomposites, polymer blends and chiral synthesis (Philip et al. 2007). The results of this study revealed that PHA production by microorganisms is dependent on the culture conditions and substrate concentration, as well as the addition of supplements. Most microorganisms, including Bacillus, require nutrient limitation and carbon excess in the medium in which they are grown, allowing them to conserve energy by means of metabolism regulation and PHA accumulation. This processes starts with carbon absorption, followed by PHA synthesis and accumulation (Reddy et al. 2003, Philip et al. 2007).

From the different species tested in this study, B. amyloliquefaciens, B. stearothermophilus, B. subtilis, B. megaterium and B. licheniformis were effective in P(3HB) biosynthesis. Furthermore, four B. megaterium strains were selected as natural PHAs producers when cultivated in media containing cassava starch by-product as the sole carbon source, which is consistent with other studies on PHA production by this species (Gouda et al. 2001; Faccin et al. 2009; Kulpreecha et al. 2009). PHA was first discovered in a B. megaterium culture, but this species was subsequently neglected by researchers, probably because it utilizes the polymer in the presence of carbon and nutrient sources as energy for the sporulation process, while other bacteria consume the polymer only when there are no carbon and nutrient sources (Wu et al. 2001). Microorganisms that undergo the sporulation process are easily preserved and replicated, and are more stress-tolerant. Thus, B. megaterium is found in abundance in soil and water, and has important industrial advantages, e.g., it is non-pathogenic, it can grow in different carbon sources and it produces PHAs from industrial by-products (Gouda et al. 2001; Law et al. 2003).

From a total of 72 strains, four isolates of B. megaterium were selected based on their capacity to bioconvert hydrolyzed cassava starchy by-product into P(3HB). One of them reached 4.97 g dry biomass/L with 29.7% of P(3HB), demonstrating its efficiency to produce biodegradable polymers from residues. The bacterial growth was monitored thought the MTT assay, which proved to be a reliable methodology for this purpose. Finally, the alternative treatment proposed for cassava by-products will minimizes environmental impacts, ensuring a sustainable life-cycle for polymeric products.

BENGTSSON, S.; WERKER, A. and WELANDER, T. (2008). Production of polyhydroxyalkanoates by glycogen accumulating organisms treating a paper mill wastewater. Water Science & Technology, vol. 58, no. 2, p. 323-330. [CrossRef]

BEUN, J.J.; PALETTA, F.; VAN LOOSDRECHT, M.C.M. and HEIJNEN, J.J. (2000). Stoichiometry and kinetics of poly-β-hydroxybutyrate metabolism in aerobic, slow growing, activated sludge cultures. Biotechnology and Bioengineering, vol. 67, no. 4, p. 379-389. [CrossRef]

BOONNOP, K.; WANAPAT, M.; NONTASO, N. and WANAPAT, S. (2009). Enriching nutritive value of cassava root by yeast fermentation. Scientia Agricola (Piracicaba, Brazil), vol. 66, no. 5, p. 629-633. [CrossRef]

CAMACHO, C.; COULOURIS, G.; AVAGYAN, V.; MA, N.; PAPADOPOULOS, J.; BEALER, K. and MADDEN, T.L. (2008). BLAST+: Architecture and applications. BMC Bioinformatics, vol. 10, p. 421. [CrossRef]

CHAKRAVARTY, P.; MHAISALKAR, V. and CHAKRABARTI, T. (2010). Study on poly-hydroxyalkanoate (PHA) production in pilot scale continuous mode wastewater treatment system. Bioresource Technology, vol. 101, no. 8, p. 2896-2899. [CrossRef]

CHIA, K.-H.; OOI, T.-F.; SAIKA, A.; TSUGE, T. and SUDESH, K. (2010). Biosynthesis and characterization of novel polyhydroxyalkanoate polymers with high elastic property by Cupriavidus necator PHB⁻4 transformant. Polymer Degradation and Stability, vol. 95, no. 12, p. 2226-2232. [CrossRef]

COSTA, S.G.V.A.O.; LÉPINE, F.; MILOT, S.; DÉZIEL, E.; NITSCHKE, M. and CONTIERO, J. (2009). Cassava wastewater as a substrate for the simultaneous production of rhamnolipids and polyhydroxyalkanoates by Pseudomonas aeruginosa. Journal of Industrial Microbiology & Biotechnology, vol. 36, no. 8, p. 1063-1072. [CrossRef]

DALCANTON, F.; IENCZAK, J.L.; FIORESE, M.L. and ARAGÃO, G.M.F. (2010). Production of poly(3-hydroxybutyrate) by Cupriavidus necator in hydrolyzed rice starch medium with soybean oil supplementation at different temperatures. Química Nova, vol. 33, no. 3, p. 552-556. [CrossRef]

DOBROTH, Z.T.; HU, S.; COATS, E.R. and MCDONALD, A.G. (2011). Polyhydroxybutyrate synthesis on biodiesel wastewater using mixed microbial consortia. Bioresource Technology, vol. 102, no. 3, p. 3352-3359. [CrossRef]

FACCIN, D.J.L.; MARTINS, I.; CARDOZO, N.S.M.; RECH, R.; AYUB, M.A.Z.; ALVES, T.L.M.; GAMBETTA, R. and RESENDE SECCHI, A. (2009). Optimization of C:N ratio and minimal initial carbon source for poly(3-hydroxybutyrate) production by Bacillus megaterium. Journal of Chemical Technology and Biotechnology, vol. 84, no. 12, p. 1756-1761. [CrossRef]

Food and Agriculture Organization of the United Nations, International Fund for Agricultural Development (FAO). (2001). Strategic environmental assessment: An assessment of the impact of Cassava production and processing on the environment and biodiversity. In: Proceedings of the validation forum on the global cassava development strategy. (26^th - 28^th April, Rome), vol. 5.

FONSECA, G.G.; DE ARRUDA-CAULKINS, J.C. and VASCONCELLOS ANTONIO, R. (2008). Production and characterization of poly-(3-hydroxybutyrate) from recombinant Escherichia coli grown on cheap renewable carbon substrates. Waste Management & Research, vol. 26, no. 6, p. 546-552. [CrossRef]

FRESHNEY, R.I. (2005). Culture of Animal Cells. New York, Wiley-Liss, 696 p. ISBN 100471453293

GOUDA, M.K.; SWELLAM, A.E. and OMAR, S.H. (2001). Production of PHB by a Bacillus megaterium strain using sugarcane molasses and corn steep liquor as sole carbon and nitrogen sources. Microbiological Research, vol. 156, no. 3, p. 201-207. [CrossRef]

GUPTA, R.; GIGRAS, P.; MOHAPATRA, H.; GOSWAMI, V.K. and CHAUHAN, B. (2003). Microbial α-amylases: A biotechnological perspective. Process Biochemistry, vol. 38, no. 11, p. 1599-1616. [CrossRef]

HIRAISHI, A. (1992). Direct automated sequencing of 16S rDNA amplified by polymerase chain reaction from bacterial cultures without DNA purification. Letters in Applied Microbiology, vol. 15, no. 5, p. 210-213. [CrossRef]

HONG, K.; SUN, S.; TIAN, W.; CHEN, G.Q. and HUANG, W. (1999). A rapid method for detecting bacterial polyhydroxyalkanoates in intact cells by Fourier transform infrared spectroscopy. Applied Microbiology and Biotechnology, vol. 5, no. 4, p. 523-526. [CrossRef]

JENDROSSEK, D.; SELCHOW, O. and HOPPERT, M. (2007). Poly(3-hydroxybutyrate) granules at the early stages of formation are localized close to the cytoplasmic membrane in Caryophanon latum. Applied and Environmental Microbiology, vol. 73, no. 2, p. 586-593. [CrossRef]

KANSIZ, M.; BILLMAN-JACOBE, H. and MCNAUGHTON, D. (2000). Quantitative determination of the biodegradable polymer poly(β-hydroxybutyrate) in a recombinant Escherichia coli strain by use of mid-infrared spectroscopy and multivariative statistics. Applied and Environmental Microbiology, vol. 66, no. 8, p. 3415-3420. [CrossRef]

KHARDENAVIS, A.A.; KUMAR, M.S.; MUDLIAR, S.N. and CHAKRABARTI, T. (2007). Biotechnological conversion of agro-industrial wastewaters into biodegradable plastic, poly β-hydroxybutyrate. Bioresource Technology, vol. 98, no. 18, p. 3579-3584. [CrossRef]

KULPREECHA, S.; BOONRUANGTHAVORN, A.; MEKSIRIPORN, B. and THONGCHUL, N. (2009). Inexpensive fed-batch cultivation for high poly(3-hydroxybutyrate) production by a new isolate of Bacillus megaterium. Journal of Bioscience and Bioengineering, vol. 107, no. 3, p. 240-245. [CrossRef]

LAW, K.H.; CHENG, Y.C.; LEUNG, Y.C.; LO, W.H.; CHUA, H. and YU, H.F. (2003). Construction of recombinant Bacillus subtilis strains for polyhydroxyalkanoates synthesis. Biochemical Engineering Journal, vol. 16, no. 2, p. 203-208. [CrossRef]

LOGAN, N. and DE VOS, P. (2009). Genus I. Bacillus Cohn 1872, 174^AL. In: DE VOS, P.; GARRITY, G.M.; JONES, D.; KRIEG, N.R.; LUDWIG, W.; RAINEY, F.A. SCHLEIFER, K.-H. and WHITMAN, W.B. eds. Bergey’s Manual of Systematic Bacteriology. 2^nd ed. Maryland, Williams and Wilkins, vol. 3, p. 21-127.

MENGMENG, C.; HONG, C.; QINGLIANG, Z.; SHIRLEY, S.N. and JIE, R. (2009). Optimal production of polyhydroxyalkanoates (PHA) in activated sludge fed by volatile fatty acids (VFAs) generated from alkaline excess sludge fermentation. Bioresource Technology, vol. 100, no. 3, p. 1399-1405. [CrossRef]

OMAR, S.; RAYES, A.; EQAAB, A.; VOß, I. and STEINBÜCHEL, A. (2001). Optimization of cell growth and poly(3-hydroxybutyrate) accumulation on date syrup by a Bacillus megaterium strain. Biotechnology Letters, vol. 23, no. 14, p. 1119-1123. [CrossRef]

PHILIP, S.; KESHAVARZ, T. and ROY, I. (2007). Polyhydroxyalkanoates: Biodegradable polymers with a range of applications. Journal of Chemical Technology and Biotechnology, vol. 82, no. 3, p. 233-247. [CrossRef]

RAMADAS, N.V.; SINGH, S.K.; SOCCOL, C.R. and PANDEY, A. (2009). Polyhydroxybutyrate production using agro-industrial residue as substrate by Bacillus sphaericus NCIM 5149. Brazilian Archives of Biology and Technology, vol. 52, no. 1, p. 17-23. [CrossRef]

RAMSAY, B.A.; LOMALIZA, K.; CHAVARIE, C.; DUBÉ, B.; BATAILLE, P. and RAMSAY, J.A. (1990). Production of poly-(β-hydroxybutyric-co-β-hydroxyvaleric) acids. Applied and Environmental Microbiology, vol. 56, no. 7, p. 2093-2098.

REDDY, C.S.K.; GHAI, R.; RASHMI, T. and KALIA, V.C. (2003). Polyhydroxyalkanoates: An overview. Bioresource Technology, vol. 87, no. 2, p. 137-146. [CrossRef]

SAMBROOK, J. and RUSSELL, D. (2001). Molecular Cloning: A Laboratory Manual. New York, Cold Spring Harbor Laboratory Press. 3 ed., 2344 p. ISBN 978-087969577-4.

SANKHLA, I.S.; BHATI, R.; SINGH, A.K. and MALLICK, N. (2010). Poly(3-hydroxybutyrate-co-3-hydroxyvalerate) co-polymer production from a local isolate, Brevibacillus invocatus MTCC 9039. Bioresource Technology, vol. 101, no. 6, p. 1947-1953. [CrossRef]

SHISHATSKAYA, E.I. and VOLOVA, T.G. (2004). A comparative investigation of biodegradable polyhydroxyalkanoate films as matrices for in vitro cell cultures. Journal of Materials Science: Materials in Medicine, vol. 15, no. 8, p. 915-923. [CrossRef]

SIEUWERTS, A.M.; KLIJN, J.G.; PETERS, H.A. and FOEKENS, J.A. (1995). The MTT tetrazolium salt assay scrutinized: How to use this assay reliably to measure metabolic activity of cell cultures in vitro for the assessment of growth characteristics, IC50-values and cell survival. European Journal of Clinical Chemistry and Clinical Biochemistry, vol. 33, no. 11, p. 813-823.

VERLINDEN, R.A.; HILL, D.J.; KENWARD, M.A.; WILLIAMS, C.D. and RADECKA, I. (2007). Bacterial synthesis of biodegradable polyhydroxyalkanoates. Journal of Applied Microbiology, vol. 102, no. 6, p. 1437-1449. [CrossRef]

WU, Q.; HUANG, H.; HU, G.H.; CHEN, J.; HO, K.P. and CHEN, G.Q. (2001). Production of poly-3-hydroxybutyrate by Bacillus sp. JMa5 cultivated in molasses media. Antonie Van Leeuwenhoek, vol. 80, no. 2, p. 111-118. [CrossRef]

ZHANG, H.; OBIAS, V.; GONYER, K. and DENNIS, D. (1994). Production of polyhydroxyalkanoates in sucrose-utilizing recombinant Escherichia coli and Klebsiella strains. Applied and Environmental Microbiology, vol. 60, no. 4, p. 1198-1250.