Introduction

The poultry industry is continuously expanding to attend the increasing worldwide demand for poultry meat. Particularly, Brazil occupies an important position in the global ranking of poultry meat production, with approximately 5.4 billion chickens slaughtered in 2014, totaling 12.5 million tons of carcasses [1]. This activity, through the processing of raw materials, intrinsically generates high amounts of organic wastes, such as feathers and viscera. Feathers are mainly composed by keratins, structural proteins possessing high cysteine contents, which lead to extensive disulfide bonding and determines feather resistance [2]. Feathers are the most common keratinous residues produced during poultry processing and, assuming that feathers account for 5–7 % of the poultry weight, approximately 600 thousand tons of feathers were produced as waste from the Brazilian poultry meat industry in 2014. Worldwide, it is estimated that 5 million tons of feathers are generated annually from the production of chicken meat [3].

The recalcitrance of keratin-rich wastes represents a challenge for its correct disposal and/or management, since local accumulation could lead to environmental problems. Feathers are usually landfilled or incinerated, processes associated with ecological concerns, such as the generation of hydrogen sulfide and ammonia, and also financial issues. Additionally, feathers could be converted through hydrothermal treatment into feather meal that could be used as an animal feed ingredient; however, this process also raises questions from safety, nutritional efficiency and economical standpoints [4]. Considering these options, alternative methods to manage these abundant keratinous wastes are needed to reduce potential monetary, environmental and health risks.

Keratins and other recalcitrant proteins are degraded in nature, indicating the active and essential role of microorganisms in recycling processes. Thus, microbial conversion is investigated as an adequate management approach since it generally occurs at mild conditions, has a low cost, and is considered to be an ecologically safe process [5]. Beyond feather degradation, microbial proteases are among the valuable products that could be recovered from such bioprocess [2]. In this sense, the exploration of microbial diversity is considered as a powerfull strategy aiming bioprocess development and also the obtainment of biotechnologically relevant enzymes [6]. Hence, the present study evaluated the proteolytic and keratinolytic potentials of bacterial isolates, and characterized the feather-degrading ability of a selected strain as a promising approach aiming feathers valorization.

Materials and Methods

Isolation of Bacteria

Decomposing feathers were aseptically collected from a site where organic residues were discarded, and transported under refrigeration to the laboratory. The feathers were employed to inoculate Erlenmeyer flasks (250 mL) containing 100 mL of feather meal broth (FMB) composed by a mineral medium (MM; 0.5 g/L NaCl, 0.3 g/L K2HPO4, 0.4 g/L KH2PO4) and feather meal (10 g/L) [7]. After 4 days of incubation in an orbital shaker (30 °C, 125 rpm), samples were collected and serially diluted in sterile NaCl solution (8.5 g/L). The dilutions were spread-plated onto tryptone soy agar (TSA) plates, and incubated at 30 °C for up to 5 days. Bacterial colonies showing distinct morphologies were isolated through successive streaking onto fresh TSA plates.

Proteolytic Potential on Solid Media

Isolated bacteria were initially screened for extracellular protease production in skim milk agar (SMA) plates [8]. After incubation at 30 °C for 24 h, plates were observed for the presence of clear zones around the colonies, which indicates the production of proteolytic enzymes. Then, clearance zone (H) and colony diameter (C) were measured.

Bacteria displaying protease production in SMA tests were inoculated into feather meal agar (FMA) plates [9], prepared with different pH values (6, 7, 8, 9, and 10), which was adjusted with 1 M NaOH or 1 M HCl before autoclaving. After inoculation, plates were incubated for 48 h at distinct temperatures (25, 30, 37, and 45 °C). Bacterial growth was checked and colony diameters were measured. The SMA and FMA testes were performed in triplicates.

Qualitative Evaluation of Feather Degradation

Proteolytic bacteria able to grow on FMA plates were evaluated for feather-degrading potential on test tubes containing 10 mL of MM and a single chicken feather. After inoculation with each bacterial isolate, the tubes were incubated for up to 10 days at 30 °C and 125 rpm. The duplicate tubes were checked on a daily basis, and the isolate demonstrating higher feather degradation was selected for subsequent studies.

Identification of the Feather-Degrading Isolate

The total DNA of the selected isolate was extracted and the 16S rRNA gene was amplified by PCR using the following universal primers: 27F (5′-AGAGTTTGATCMTGGCTCAG-3′) and 1525R (5′-AAGGAGGTGWTCCARCC-3′) [10]. The sequencing reactions were carried out using the same primers above and the sequences obtained were submitted to the BLAST search program of the National Center for Biotechnology Information website (NCBI, http://www.ncbi.nlm.nih.gov) in order to search for homologous sequences. BioEdit [11] was used to edit sequences and for the construction of 16S rRNA gene contig.

Feather-Degrading Potential of the Selected Isolate

Quantitative analyses were performed during growth of the selected bacterial isolate in feather broth (FB). This medium was composed by MM and contained whole chicken feathers (10 g/L) as the sole source of carbon, nitrogen, and energy. The initial pH of FB was adjusted to 7.0 before autoclaving. Erlenmeyer flasks (250 mL) containing 50 mL of FB were inoculated with 1 mL of a bacterial suspension (O.D.600 of 0.9) prepared as described elsewhere [12]. Cultivations were perfomed at 30 °C in an orbital shaker (125 rpm) for 11 days, and duplicate Erlenmeyers were withdrawn every 24 h to determine the degradation of feathers, medium pH, soluble protein concentration, and protease production.

Feather degradation was determined by the dry weight of feathers remaining on FB during incubation. Cultures were filtered through pre-weighted filter paper and dried at 60 °C until constant weight; results were expressed as percentage of the initial feather weight (100 %) of the respective keratinous waste. Culture filtrates were centrifuged (10,000g for 5 min) and the resulting supernatants were utilized for determination of proteolytic activity using azocasein as substrate [12], and measurement of soluble protein concentration by the Folin-phenol method using bovine serum albumin as standard. All determinations were performed in triplicates.

Protease Production with Different Growth Substrates

Extracellular protease production by the selected isolate was also investigated during submerged cultivations with different organic growth substrates. Feather meal, human hair, casein, peptone, soy protein isolate, or cheese whey powder, were investigated at 10 g/L in MM. Initial medium pH was adjusted to 7.0 before autoclaving. Erlenmeyer flasks (250 mL) containing 50 mL of the respective medium were inoculated and incubated as previously described for FB. Samples were collected every 24 h, centrifuged (10,000g for 5 min), and the supernatants utilized for determination of proteolytic activity using azocasein as substrate.

Results and Discussion

Proteolytic and Keratinolytic Potentials of Bacterial Isolates

Functional screening of microorganisms could provide important informations regarding microbial ecology, but also constitutes one of the foundations of industrial biotechnology [13]. In the present investigation, bacteria were isolated from a site where organic wastes, including feathers, were discarded. Through this culture-dependent approach, 15 morphologically distinct bacterial colonies were obtained, named CL23 to CL28, CL30, CL32B, CL33A, CL33B, CL35A, CL36, and CL38 to CL40.

Isolates were initially screened for proteolytic potential on SMA plates, a strategy commonly employed for this purpose. Although all bacteria were able to grow in this medium, seven isolates demonstrated protease production (Table 1). Growth and clearance zones were not measured for the CL39 isolate, since it exhibited a filamentous growth. Ghosh et al. [14] investigated enzyme production by 38 bacteria isolated from Indian soils, and observed that 76 % displayed proteolytic potential on casein agar, representing the most abundant enzyme activity among those evaluated. An evaluation of 100 soil bacteria isolated from a detergent industry in South Korea indicated that 80 % produced extracellular proteases on SMA plates [15]. From Table 1, hydrolysis zones ranged from 2.5 to 6.1 mm. Barros et al. [16] observed clearance zones ranging from 2.5 to 10.0 mm for 10 Bacillus subtilis strains grown in SMA plates at 30 °C for 24 h. Five proteolytic bacteria isolated from feather waste displayed hydrolysis zones of 2–5 mm in this medium [8]. Singh et al. [17] reported that among 70 proteolytic bacteria isolated from soil, 40 % were considered as good protease producers, exhibiting clearance zones higher than 3 mm on milk agar plates incubated for 20–30 h at 37 °C.

Table 1 Growth and proteolytic potential of bacterial isolates evaluated in skim milk agar plates (24 h, 30 °C)
Full size table

FMA is considered a useful indicator of the microbial ability to utilize keratin as a growth substrate. Bach et al. [7] reported the isolation of 34 bacterial strains capable of growing on FMA from soils of the Brazilian Atlantic Forest; and a functional screening performed by Ghosh et al. [14] indicated that six out of 38 bacterial strains degraded keratin on feather meal medium. Therefore, proteolytic strains were submitted to growth tests on FMA plates prepared with distinct pH values (6.0–10.0) and incubation was carried out at different temperatures (25–45 °C). It is worth mentioning that each bacterial strain has intrinsic colonial features, and thus comparisons among isolates are inappropriate. However, it is possible to compare growth profiles for the same isolate cultured at different conditions.

Strains CL25, CL30, CL33A, and CL36 grew in all the tested conditions of pH and temperature (Supplementary Fig. S1). The isolate CL33B was unable to grow at pH 6.0, and at pH 10.0 with incubation temperature of 45 °C; and strain CL23 did not grew at pH 6.0 when incubated at 45 °C. Growth profiles of the CL39 isolate are not shown, but it was able to grow at all conditions tested, except at pH 10.0 when incubated at 45 °C. In general, the isolates tended to grow better at neutral to slightly alkaline conditions (pH 7.0–9.0), and at temperatures of 30 and 37 °C, reflecting the environmental conditions from where they were isolated [9]. Beyond the ecological perspective, from which the growth on SMA and FMA plates could indicate the participation of these bacterial isolates in the turnover of protein-rich organic matter, thus performing important ecosystem services, the observed proteolytic activities opens the possibility of utilizing such potential for technological purposes.

Therefore, the six proteolytic bacteria were evaluated for their ability to degrade whole chicken feathers as an indicative of their keratinolytic potential. In these qualitative tests, only the CL33A isolate demonstrated extensive feather degradation (Fig. 1). Although much research is dedicated to the feather-degrading capacity of thermophilic and/or alkalophilic microorganisms, bioprocesses involving mesophilic and neutrophilic ones, such as the CL33A isolate, would imply in lower energy inputs, also potentially avoiding the generation of highly alkaline effluents to be treated [8, 18]. Considering the proteolytic and keratinolytic potentials of isolate CL33A, it was selected for further investigations. Through 16S rRNA gene sequencing, the isolated was identified as a member of the genus Bacillus, sharing 99 % of sequence similarity with B. aerius (accession number NR_118439), a member of the B. pumilus group, being therefore denominated Bacillus sp. CL33A.

Fig. 1
figure 1

Feather-degrading potential of the CL33A isolate. Cultivations were performed at 30 °C, initial pH 7.0, for 2 (b), 4 (c), 6 (d) and 8 (e) days. a Non-inoculated control, incubated at the same conditions for 8 days

Full size image

Feather Degradation by Bacillus sp. CL33A

The feather-degrading potential of Bacillus sp. CL33A was investigated in FB. During incubation, feather degradation, soluble protein content, medium pH and proteolytic activity were determined (Fig. 2). After 4 days of growth, the isolate solubilized 29 % of the initial feather mass, reaching 75 % and 95 % of degradation after 6 and 9 days of cultivation, respectively (Fig. 2a). The keratinolytic potential of CL33A is comparable to that of Bacillus pumilus F3-4, which degraded 75 % of the feathers after 7 days [19], and higher than that of B. polymyxa B20 and B. cereus B5esz [20]. Queiroga et al. [21] evaluated feather degradation by five Bacillus spp., and reported values ranging from 12 to 67 % after 4 days of growth. Thus, the feather-degrading potential of keratinolytic strains is highly variable [7, 22]. For instance, Bacillus sp. P45 demonstrated efficient feather degradation (90 %) after 3 days [12], and Paenibacillus woosongensis TKB2 degraded 87 % of feathers (0.8 %, w/v), after culture optimization, in 2 days [23].

Fig. 2
figure 2

Quantitative analyses during growth of the CL33A isolate on feather broth (FB) at 30 °C, initial pH 7.0, 125 rpm. a Residual feathers (%); b soluble protein concentration (mg/mL); c medium pH; d proteolytic activity (U/mL). The values are mean ± SEM of two independent experiments repeated twice (n = 4)

Full size image

Feather degradation by Bacillus sp. CL33A was corroborated by the increase in soluble protein concentration on culture supernatants, reaching approximately 5.0 mg/mL after 7 days of cultivation (Fig. 2b), and also by the increment on medium pH, from 7.0 (initial) to 8.7 at day 6 (Fig. 2c). Since FB contains feathers as the only organic substrate to sustain microbial growth, the accumulation of soluble proteins shows the efficiency of feather bioconversion beyond bacterial needs. Also, as the pH increase usually relates to the deamination of peptides and amino acids, these results indicate the utilization of feather proteins as carbon, nitrogen and energy sources by Bacillus sp. CL33A. In fact, medium alkalinization is often employed as an indication of the keratinolytic potential of microbial strains [24].

Microbial keratinolysis is a complex process, usually depending on the synergistic action of sulfitolytic mechanisms and extracellular proteolytic enzymes [25]. Protease production by Bacillus sp. CL33A was accompanied during cultivations on FB. A gradual increase on protease production was observed, especially from day 3, reaching maximal values at day 8 (Fig. 2d). The increase on protease production was similar to the feather solubilization profile (Fig. 2a, b), indicating the crucial role of extracellular proteases for feather degradation and, thus, microbial nutrition.

Protease Production by Bacillus sp. CL33A During Submerged Cultivations

Proteolytic enzymes are valuable biocatalysts employed by the food, feed, leather, detergent, pharmaceutical, and other industries. Microbial proteases dominate the enzyme market due to their diverse physicochemical and catalytic properties, and also because microorganisms provide a regular supply of enzymes [26]. More recently, microbial keratinolytic proteases are gaining momentum due to their versatility; however, commercial availability is still limited [27]. Considering the commercial and industrial applicability of these enzymes and its usually high production costs, increasing interest is focused both on protease-producing microorganisms and on the evaluation of appropriate substrates for protease production [6].

Thus, production of extracellular protease by Bacillus sp. CL33A was investigated using different organic substrates (Fig. 3). Protease yield on FB (Fig. 2d) was added for comparison purposes. Peptone resulted in higher protease production as a function of cultivation time (192 U/mL at day 2), followed by feather meal (160 U/mL at day 3), soy protein isolate (207 U/mL at day 4), chicken feathers (248 U/mL at day 8), and casein (167 U/mL at day 10) (Fig. 3). In these media, after the protease production peak was reached, proteolytic activity decreased as the incubation time progressed, which could be a consequence of enzyme autolysis. Similar results were observed elsewhere [9, 12, 17, 22].

Fig. 3
figure 3

Protease production by Bacillus sp. CL33A during submerged cultivations (30 °C, initial pH 7.0, 125 rpm) on mineral medium containing different organic substrates (10 g/L). Peptone (open triangle), feather meal (filled square), soy protein isolate (open circle), whole feathers (filled circle), casein (open square), cheese whey powder (filled diamond), and human hair (filled inverted triangle). Values are mean ± SEM of three independent experiments

Full size image

Although feathers and human hair contain sufficient carbon and nitrogen to support growth, its macromolecular character and insolubility are not appropriate to meet microbial needs; hence, nutrient deprivation could act as a trigger for protease production and secretion [28]. However, Bacillus sp. CL33A was not able to produce proteases nor to degrade human hair during cultivations (results not shown), which might have been the result of strain inability to grow using this substrate [29]. Hair and feathers are mainly composed by keratins, but in the former these proteins are mainly found as α-keratins, possessing higher cysteine content than the latter, resulting in a more compact strucuture and higher stability against microbial attack [2, 30].

When comparing protease production on feathers and feather meal, higher yields on the latter could have resulted from its enhanced accessibility [18, 29]. It could also be noted a relatively lower protease production during growth on casein (Fig. 3), even though azocasein was the substrate employed to detect protease activity on culture supernatants. Also, protease production was negligible during cultivations on cheese whey powder (Fig. 3). Production of extracellular proteases by Bacillus spp. is extremely complex, since it is controlled by different mechanisms, including substrate induction, nutritional stresses, and catabolite repression [2]. Nevertheless, the obtained results indicate that specific organic substrates were needed for extracellular protease production by Bacillus sp. CL33A, and that the substrate type affected protease yield.

The assessment of substrates for microbial protease production, besides evaluating specific conditions for higher enzyme yields, should also consider economical and environmental aspects. Agroindustrial residues represent valuable alternatives to be exploited, even if these substrates do not result in higher enzyme production [6]. Particularly, feather meal is used on a limited basis as an ingredient in animal feed. Therefore, chicken feathers and feather meal might be adequate substrates for cost-effective enzyme production due to their wide availability and low cost. Considering feathers composition (90 % keratin protein), bioprocessing technologies could be appropriate to avoid the wastage of an interesting protein source [4]. Hence, bioconversion could represent a strategy to obtain valuable feather/feather meal hydrolysates with increased nutritional value and digestibility for animal feed, for utilization as nitrogen-rich fertilizers, and also as substrates for biohydrogen and methane production [5, 22, 31]. Additionally, feather hydrolysates obtained through microbial processing were recently characterized as potential sources of bioactive peptides [32].

Conclusion

The largely unknown microbial diversity provides the framework for searching useful bacterial strains aiming technological applications. In this study, proteolytic and keratinolytic potentials were focused. Functional screening indicated that protease production was a common feature of culturable bacteria, and that the proteolytic strains were able to grow on feather meal agar at different pH (6.0–10.0) and temperature conditions (25–45 °C), suggesting their participation on cycling of protein-rich materials. However, only the isolate identified as Bacillus sp. CL33A possessed the ability to solubilize whole chicken feathers. This neutrophilic and mesophilic isolate produced proteolytic enzymes during submerged cultivations on feathers, feather meal, soy protein isolate, casein, and peptone. The keratinolytic potential of Bacillus sp. CL33A could be employed for the bioconversion of waste feathers and feather meal to obtain invaluable biotechnological products. Characterization of the crude protease is under way to postulate biocatalyst applications.