Optimization of fermentation conditions for production of anti-TMV extracellular ribonuclease by Bacillus cereus using response surface methodology

Bacillus cereus ZH14 was previously found to produce a new type of antiviral ribonuclease, which was secreted into medium and active against tobacco mosaic virus. In order to enhance the ribonuclease production, in this study the optimization of culture conditions using response surface methodology was done. The fermentation variables including culture temperature, initial pH, inoculum size, sucrose, yeast extract, MgSO₄·7H₂O, and KNO₃ were considered for selection of significant ones by using the Plackett–Burman design, and four significant variables (sucrose, yeast extract, MgSO₄·7H₂O, and KNO₃) were further optimized by a 2⁴ factorial central composite design. The optimal combination of the medium constituents for maximum ribonuclease production was determined as 8.50 g/l sucrose, 9.30 g/l yeast extract, 2.00 g/l MgSO₄·7H₂O, and 0.62 g/l KNO₃. The enzyme activity was increased by 60%. This study will be helpful to the future commercial development of the new bacteria-based antiviral ribonuclease fermentation process.     

Statistical optimization of medium components for avilamycin production by <Emphasis Type="Italic">Streptomyces viridochromogenes</Emphasis> Tü57-1 using response surface methodology

A fermentation medium for avilamycin production by Streptomyces viridochromogenes Tü57-1 has been optimized. Important components and their concentrations were investigated using fractional factorial design and Box–Behnken Design. The results showed that soybean flour, soluble starch, MgSO₄·7H₂O and CaCl₂·2H₂O are important for avilamycin production. A polynomial model related to medium components and avilamycin yield had been established. A high coefficient of determination (R ² = 0.92) was obtained that indicated good agreement between the experimental and predicted values of avilamycin yield. Student’s T-test of each coefficient showed that all the linear and quadratic terms had significant effect (P > |T| < 0.05) on avilamycin yield. The significance of tested components was related to MgSO₄·7H₂O (0.37 g/L), CaCl₂·2H₂O (0.39 g/L), soybean flour (21.97 g/L) and soluble starch (37.22 g/L). The yield of avilamycin reached 88.33 ± 0.94 mg/L (p < 0.05) that was 2.8-fold the initial yield.     

Production of Alcaligenes xylosoxydans EMS33 in a Bench-scale Fermenter

Summary Optimization of medium composition and pH for chitinase production by the Alcaligenes xylosoxydans mutant EMS33 was carried out in the present study and the optimized medium composition and conditions were evaluated in a fermenter. The medium components screened initially using Plackett–Burman design were (NH₄)₂SO₄, MgSO₄ 7H₂O, KH₂PO₄, yeast extract, Tween 20 and chitin in shake flask experiments. The significant medium components identified by the Plackett–Burman method were MgSO₄ 7H₂O, Tween 20 and chitin. Central composite response surface methodology was applied to further optimize chitinase production. The optimized values of MgSO₄ 7H₂O, Tween 20, chitin and pH were found to be 0.6 g/l, 0.05 g/l, 11.5 g/l and 8.0, respectively. Chitinase and biomass production of Alcaligenes xylosoxydans EMS33, was studied in a 2-l fermenter containing (g/l): chitin, 11.5; yeast extract, 0.5; (NH₄)₂SO₄, 1; MgSO₄ 7H₂O, 0.6; KH₂PO₄, 1.36 and Tween 20, 0.05. The highest chitinase production was 54 units/ml at 60 h and pH 8.0 when the dissolved O₂ concentration was 60%, whereas the highest biomass production was achieved at 36 h and pH 7.5 without any dissolved O₂ control.     

Use of response surface methodology to study the effect of media composition on aflatoxin production by Aspergillus flavus

Aflatoxins are one of the most important secondary metabolites. These extrolites are produced by a number of Aspergillus fungi. In this study, we demonstrate the effect of media components and enhanced aflatoxin yield shown by A. flavus using response surface methodology in response to different nutrients. Different components of a chemically defined media that influence the aflatoxin production were monitored using Plackett–Burman experimental design and further optimized by Box–Behnken factorial design of response surface methodology in liquid culture. Interactions were studied with five variables, namely sorbitol, fructose, ammonium sulfate, KH₂PO₄, and MgSO₄.7H₂O. Maximum aflatoxin production was envisaged in medium containing 4.94 g/l sorbitol, 5.56 g/l fructose, 0.62 g/l ammonium sulfate, 1.33 g/l KH₂PO₄, and 0.65 g/l MgSO₄·7H₂O using response surface plots and the point prediction tool of the DESIGN EXPERT 8.1.0 (Stat-Ease, USA) software. However, a production of 5.25 μg/ml aflatoxin production was obtained, which was in agreement with the prediction observed in verification experiment. The other component (MgSO₄.7H₂O) was found to be an insignificant variable.     

Development of medium for enhanced production of glutaminase-free <Emphasis Type="SmallCaps">l</Emphasis>-asparaginase from <Emphasis Type="Italic">Pectobacterium carotovorum</Emphasis> MTCC 1428

Glutaminase-free l-asparaginase is known to be an excellent anticancer agent. In the present study, statistically based experimental designs were applied to maximize the production of glutaminase-free l-asparaginase from Pectobacterium carotovorum MTCC 1428. Nine components of the medium were examined for their significance on the production of l-asparaginase using the Plackett–Burman experimental design. The medium components, viz., glucose, l-asparagine, KH₂PO₄, and MgSO₄·7H₂O, were screened based on their high confidence levels (P < 0.04). The optimum levels of glucose, l-asparagine, KH₂PO₄, and MgSO₄·7H₂O were found to be 2.076, 5.202, 1.773, and 0.373 g L⁻¹, respectively, using the central composite experimental design. The maximum specific activity of l-asparaginase in the optimized medium was 27.88 U mg⁻¹ of protein, resulting in an overall 8.3-fold increase in the production compared to the unoptimized medium.     

Production of <Emphasis Type="SmallCaps">l</Emphasis>-lactic acid by <Emphasis Type="Italic">Rhizopus oryzae</Emphasis> using semicontinuous fermentation in bioreactor

Semicontinuous fermentation using pellets of Rhizopus oryzae has been recognized as a promising technology for l-lactic acid production. In this work, semicontinuous fermentation of R. oryzae AS 3.819 for l-lactic acid production has been developed with high l-lactic acid yield and volumetric productivity. The effects of factors such as inoculations, CaCO₃ addition time, and temperature on l-lactic acid yield and R. oryzae morphology were researched in detail. The results showed that optimal fermentation conditions for the first cycle were: inoculation with 4% spore suspension, CaCO₃ added to the culture medium at the beginning of culture, and culture temperature of 32–34°C. In orthogonal experiments, high l-lactic acid yield was achieved when the feeding medium was (g/l): glucose, 100; (NH₄)₂SO₄, 2; KH₂PO₄, 0.1; ZnSO₄·7H₂O, 0.33; MgSO₄·7H₂O, 0.15; CaCO₃, 50. Twenty cycles of semicontinuous fermentation were carried out in flask culture. l-lactic acid yield was 78.75% for the first cycle and 80–90% for the repeated cycles; the activities of lactate dehydrogenases (LDH) were 7.2–9.2 U/mg; fermentation was completed in 24 h for each repeated cycle. In a 7-l magnetically stirred fermentor, semicontinuous fermentation lasted for 25 cycles using pellets of R. oryzae AS 3.819 under the optimal conditions determined from flask cultures. The final l-lactic acid concentration (LLAC) reached 103.7 g/l, and the volumetric productivity was 2.16 g/(l·h) for the first cycle; in the following 19 repeated cycles, the final LLAC reached 81–95 g/l, and the volumetric productivities were 3.40–3.85 g/(l·h).     

Contributo Alla Flora Dell'alto Adige

In order to obtain a high ethanol yield from the Jerusalem artichoke raw extract and reduce the fermentation cost, we have engineered a new recombinant Saccharomyces cerevisiae strain that could produce ex-inulinase. The response surface methodology based on Plackett–Burman and Box–Behnken design was used to optimize the medium for the ethanol production from the Jerusalem artichoke raw extracts by the recombinant strain. In the first optimization step, Plackett–Burman design was employed to select significant factors, including concentrations of yeast extract, inoculum, and MgSO₄·7H₂O. In the second step, the steepest ascent experiment was carried out to determine the center point with the three significant factors; the selected combinations were further optimized using the Box–Behnken design. The maximum ethanol production rate was predicted at 91.1 g/l, which was based on a medium consisting of yeast extract 9.24 g/l, inoculum 39.8 ml/l, and MgSO₄·7H₂O 0.45 g/l. In the validating experiment, the ethanol fermentation rate reached 102.1 g/l, closely matching the predicted rate.     

Medium optimization for the production of a novel bioflocculant from Halomonas sp. V3a′ using response surface methodology

The novel exopolysaccharide bioflocculant HBF-3 is produced by Halomonas sp. V3a′, which is a mutant strain of the deep-sea bacterium Halomonas sp. V3a. Response surface methodology (RSM) was employed to optimize the production medium for increasing HBF-3 production. Using a Plackett–Burman experimental design to aid in the first step of optimization, edible glucose, MgSO₄·7H₂O, and NH₄Cl were found to be significant factors affecting HBF-3 production. To determine the optimal concentration of each significant variable, a central composite design was employed. Based on response surface and canonical analysis, the optimum concentrations of the critical components were obtained as follows: edible glucose, 16.14 g/l; MgSO₄·7H₂O, 2.73 g/l; and NH₄Cl, 1.97 g/l. HBF-3 production obtained by using the optimized medium was 4.52 g/l, which was in close agreement with the predicted value of 4.55 g/l. By scaling up fermentation from flask to fermenter, HBF-3 production was further increased to 5.58 g/l.     

A new medium for spore production of <Emphasis Type="Italic">Blakeslea trispora</Emphasis> using response surface methodology

Optimization of the medium components which enhance sporulation of the two mating types of the fungus Blakeslea trispora ATCC 14271 and ATCC 14272 (a heterothallic Zygomycota producing carotene) was achieved with the aid of response surface methodology (RSM). Glucose, corn steep liquor, yeast extract, and ammonium sulfate were investigated as carbon and nitrogen sources in a basal medium. RSM was adopted to optimize the medium in order to obtain a good growth of the fungus as a prerequisite for enhanced sporulation. In the second step, the basal medium was supplemented with different trace elements which significantly affect sporulation (i.e. CuSO₄·5H₂O, FeCl₃·6H₂O, Co(NO₃)₂·6H₂O, and MnCl₂·4H₂O). Central composite design proved to be valuable in optimizing a chemically defined solid medium for spore production of B. trispora. The composition of the new solid medium to enhance spore production by B. trispora (ATCC 14271) is as follows (per liter): 7.5 g glucose, 3.2 g corn steep liquor, 1.7 g yeast extract, 4.1 g ammonium sulfate, 6 mg CuSO₄·5H₂O, 276 mg FeCl₃·6H₂O, 2 mg Co(NO₃)₂·6H₂O, and 20 g agar (pH 6.0). Practical validation of this optimum medium gave spore number of 1.2 × 10⁸ spores/dish which is 77% higher than that produced in Potato Dextrose Agar (PDA). In the case of B. trispora (ATCC 14272) the new solid substrate for enhanced sporulation consists of (per l) 6.4 g glucose, 3.3 g corn steep liquor, 1.4 g yeast extract, 4.3 g ammonium sulfate, 264 mg CuSO₄·5H₂O, 485 mg FeCl₃·6H₂O, 223 mg MnCl₂^.4H₂O, and 20 g agar (pH 6.0). Spore numbers of 2 × 10⁷ spores/dish were obtained on the new medium by B. trispora (ATCC 14272), which is 95% higher than that produced on PDA. The results corroborated the validity and the effectiveness of the models. The new media considerably improved sporulation of both strains of B. trispora compared to the production of spores on PDA, which is the medium usually used for sporulation of the fungus.     

Ca<Superscript>2+ </Superscript>and Cu<Superscript>2+ </Superscript>supplementation increases mannitol production by <Emphasis Type="Italic">Candida magnoliae</Emphasis>

Supplementation with CaCl₂·2H₂O (50 mg l⁻¹) or CuSO₄·5H₂O (10 mg l⁻¹) improved mannitol production by Candida magnoliae by 14.5 and 18.6% (25 and 32 g/L), respectively. When used in combination, they acted synergistically: Ca²⁺ decreased the intracellular concentration of mannitol 30%, whereas Cu²⁺ increased the intracellular activity of mannitol dehydrogenase 1.6-times more than control. Ca²⁺ probably works by altering the permeability of cells to mannitol, whereas, Cu²⁺ increases the activity of an enzyme responsible for mannitol biosynthesis.     

MEDIUM OPTIMIZATION OF ANTIFUNGAL ACTIVITY PRODUCTION BY Bacillus amyloliquefaciens USING STATISTICAL EXPERIMENTAL DESIGN

In order to overproduce biofungicides agents by Bacillus amyloliquefaciens BLB371, a suitable culture medium was optimized using response surface methodology. Plackett–Burman design and central composite design were employed for experimental design and analysis of the results. Peptone, sucrose, and yeast extract were found to significantly influence antifungal activity production and their optimal concentrations were, respectively, 20 g/L, 25 g/L, and 4.5 g/L. The corresponding biofungicide production was 250 AU/mL, corresponding to 56% improvement in antifungal components production over a previously used medium (160 AU/mL). Moreover, our results indicated that a deficiency of the minerals CuSO₄, FeCl₃ · 6H₂O, Na₂MoO₄, KI, ZnSO₄ · 7H₂O, H₃BO₃, and C₆H₈O₇ in the optimized culture medium was not crucial for biofungicides production by Bacillus amyloliquefaciens BLB371, which is interesting from a practical point of view, particularly for low-cost production and use of the biofungicide for the control of agricultural fungal pests.     

Optimization of nutrient medium containing agricultural waste for xylanase production by Bacillus pumilus B20

We aimed to optimize a nutrient medium containing agricultural waste for xylanase production by Bacillus pumilus B20. Xylanase production with lignocellulosic material was optimized in two steps using DeMeo’s fractional factorial design. A 3.4-fold increase in xylanase production (313.3 U/mL) was achieved using the optimized culture medium consisting of (g/L): K₂HPO₄, 2; MgSO₄·7H₂O, 0.3; CaCl₂·2H₂O, 0.01; NaCl, 2; peptone, 5 yeast extract, 4; and wheat bran, 50. _{B. pumilus} B20 produced a high level of xylanase, which may have potential industrial application.     

Acetylene reduction and indoleacetic acid production by Azospirillum isolates from Cactaceous plants

Azospirillum isolates were obtained from rhizosphere soil and roots of three cactaceae species growing under arid conditions. All Azospirillum isolates from rhizosphere and roots ofStenocereus pruinosus andStenocereus stellatus were identified asA. brasilense; isolates of surface-sterilized roots fromOpuntia ficus-indica were bothA. brasilense andA. lipoferum. Azospirilla per g of fresh root in the three species ranged from 70×10³ to 11×10³. The most active strains in terms of C₂H₂ reduction (25–49.6 nmol/h·ml) and indoleacetic acid (IAA) production (36.5–77 μg/ml) were those identified asA. brasilense and isolated from Stenocereus roots.A. lipoferum isolated from Opuntia roots produced low amounts of IAA (6.5–17.5 μg/ml) and low C₂H₂-reduction activity (17.8–21.2 nmol/h·ml).     

Production of mycelial biomass and exo-polymer by <Emphasis Type="Italic">Hericium erinaceus</Emphasis> CZ-2: Optimization of nutrients levels using response surface methodology

The Doehlert experimental design was used to optimize the production of mycelial biomass and exopolymer from Hericium erinaceus CZ-2 in this study. Statistical analysis showed that the linear and quadric terms of 3 variables: corn flour, yeast extract, and corn steep liquor had significant effects. The optimized combination of these 3 variables was confirmed through validation experiments. The optimal conditions for higher production of mycelial biomass (19.92 g/L) were estimated when the media composition concentrations were set as: 30.85 g/L, corn flour; 2.81 g/L, yeast extract; 16.9 mL/L, corn steep liquor; 10 g/L, glucose; 1 g/L, KH₂PO₄; and 0.5 g/L, MgSO₄·7H₂O; while a maximal exo-polymer yield (1.653 g/L) could be achieved when setting concentrations of: 32.71 g/L, corn flour; 2.35 g/L, Yeast extract; 14.42 mL/L, Corn steep liquor; 10 g/L, glucose; 1 g/L, KH₂PO₄; and 0.5 g/L, MgSO₄·7H₂O. The upscale production was also investigated using a 15 L fermentor using the optimized medium.     

Improvement of English walnut somatic embryo germination and conversion by desiccation treatments and plantlet development by lower medium salts

Summary Well-developed somatic embryos were selected from a repetivively somatic embryo line derived from embryonic axes of immature zygotic embryos of English walnut ‘No. 120’ (Juglans regia L.) for germination and conversion studies. In germinating dishes, somatic embryos germinated into only shoots, only roots, or both shoots and roots. Without any pretreatment, 28% somatic embryos germinated, while those treated with 2.5–5.0 mg 1⁻¹ (7.2–14.4 μmol) gibberellic acid (GA₃) germinated at 25–28% and those receiving a cold treatment of 2–3 mo. at 3–4°C germinated at 30–43%. However, only 4–19% of the germinating embryos showed both shoots and roots. Treated with desiccation, either with CaCl₂·6H₂O or Ca(NO₃)₂·4H₂O at 20°C in the dark for 3 d, somatic embryos germinated at 85–91%, 57–69% of which had both shoots and roots. Treatment with 2 mo. cold storage in combination with desiccation using Ca(NO₃)₂·4H₂O resulted in 92% of somatic embryos germinating, 70% of which showed both shoots and roots. No significant differences were observed between solid and liquid germination media. After transferring the germinating embryos to plantlet development media, 52–63% of those with both shoots and roots developed into plantlets while 11% with only shoots or 9% with only roots converted into plantlets. Plantlet development was improved by using lower medium salts and sucrose concentrations. The addition of activated charcoal enhanced root development, particularly root branching. Of 131 plants transplanted, 91 plants were acclimatized to a greenhouse.     

A NEWLY ANTI-Streptococcus suis BACTERIOCIN PRODUCING STRAIN FROM UNWEANED PIGLETS FECAL MATTER: ISOLATION,PRELIMINARY IDENTIFICATION,AND OPTIMIZATION OF MEDIUM COMPOSITION FOR ENHANCED BACTERIOCIN PRODUCTION

A newly isolated anti-Streptococcus suis bacteriocin-producing strain LPL1-5 was obtained from healthy unweaned piglets' fecal matter, and was designated as Lactobacillus pentosus LPL1-5 based on morphology, biochemical properties, and 16S rDNA sequencing analysis. The medium composition for enhanced bacteriocin production by L. pentosus LPL1-5 was optimized by statistical methodology. Yeast extract, K₂HPO₄ · 3H₂O, and MnSO₄ · H₂O were identified as significant components influencing pentocin LPL1-5 production using the Plackett–Burman method. Response surface methodology was applied for further optimization. The concentrations of medium components for enhanced pentocin LPL1-5 production were as follows (g/L): lactose 20.00, tryptone 10.00, beef extract 10.00, yeast extract 14.00, MnSO₄ · H₂O 0.84, K₂HPO₄ · 3H₂O 4.92, triammonium citrate 2.00, Na-acetate 5.00, MgSO₄ · 7H₂O 0.58, Tween 80 1.00. Under the optimized condition, a value of 3154.65 ± 27.93 IU/mL bacteriocin activity was achieved, which was 4.2-fold that of the original medium.     

High level phytase production by <Emphasis Type="Italic">Aspergillus niger</Emphasis> NCIM 563 in solid state culture: response surface optimization,up-scaling,and its partial characterization

Phytase production by Aspergillus niger NCIM 563 was optimized by using wheat bran in solid state fermentation (SSF). An integrated statistical optimization approach involving the combination of Placket–Burman design (PBD) and Box–Behnken design (BBD) was employed. PBD was used to evaluate the effect of 11 variables related to phytase production, and five statistically significant variables, namely, glucose, dextrin, NaNO₃, distilled water, and MgSO₄·7H₂O, were selected for further optimization studies. The levels of five variables for maximum phytase production were determined by a BBD. Phytase production improved from 50 IU/g dry moldy bran (DMB) to 154 IU/g DMB indicating 3.08-fold increase after optimization. A simultaneous reduction in fermentation time from 7 to 4 days shows a high productivity of 38,500 IU/kg/day. Scaling up the process in trays gave reproducible phytase production overcoming industrial constraints of practicability and economics. The culture extract also had 133.2, 41.58, and 310.34 IU/g DMB of xylanase, cellulase, and amylase activities, respectively. The partially purified phytase was optimally active at 55°C and pH 6.0. The enzyme retained ca. 75% activity over a wide pH range 2.0–9.5. It also released more inorganic phosphorus from soybean meal in a broad pH range from 2.5 to 6.5 under emulated gastric conditions. Molecular weight of phytase on Sephacryl S-200 was approximately 87 kDa. The K _m and V _max observed were 0.156 mM and 220 μm/min/mg. The SSF phytase from A. niger NCIM 563 offers an economical production capability and its wide pH stability shows its suitability for use in poultry feed.     

响应面法优化麻疯树花粉离体萌发培养基的研究

麻疯树因其种子含油率较高,种子油提炼的生物柴油可部分替代汽油,而成为一种极具潜能的能源作物,但由于产量低,麻疯树在热带、亚热带的发展受到极大限制。杂交育种是提高产量的重要手段,杂交亲本花粉生活力的高低直接影响到育种的成效。因此,寻求麻疯树离体花粉萌发的最适培养基配方,探明花粉萌发培养基中各主要培养基成分间的交互作用对生产上麻疯树杂交结实率和种子产量的提高具有重要意义。该研究以麻疯树开花初期雄花上花药刚散粉时的成熟花粉粒为材料,采用Box-Behnken设计(Box-Behnken design,BBD)的响应面法,对麻疯树花粉离体萌发培养基中各主要培养基成分的浓度配比及各主要培养基成分的交互作用进行了研究。以花粉萌发率为响应指标,建立了4种营养成分(蔗糖、硼酸、硝酸钙、硝酸钾)与花粉萌发率的响应面模型,并对各主要培养基成分的浓度配比进行了优化。通过R软件进行响应面分析的结果表明:4因素对花粉萌发率的影响顺序为蔗糖硼酸硝酸钙硝酸钾;蔗糖与硼酸、蔗糖与硝酸钙、蔗糖与硝酸钾之间的交互作用显著。响应面建模优化后的最佳培养基为13.77%蔗糖+32.14 mg·L~(-1)硼酸+22.21 mg·L~(-1)硝酸钙+19.95 mg·L~(-1)硝酸钾+200 mg·L~(-1)硫酸镁,在此条件下的理论萌发率为99.73%。采用此培养基成分配比得到麻疯树花粉离体试验萌发率为98.97%,与理论响应值相吻合,同时也表明利用BBD设计的响应面模型进行麻疯树花粉离体萌发培养条件优化方法的有效性。     

Optimization of polyhydroxybutyrate production by <Emphasis Type="Italic">Bacillus</Emphasis> sp. CFR 256 with corn steep liquor as a nitrogen source

Polyhydroxyalkanotes (PHAs), the eco-friendly biopolymers produced by many bacteria, are gaining importance in curtailing the environmental pollution by replacing the non-biodegradable plastics derived from petroleum. The present study was carried out to economize the polyhydroxybutyrate (PHB) production by optimizing the fermentation medium using corn steep liquor (CSL), a by-product of starch processing industry, as a cheap nitrogen source, by Bacillus sp. CFR 256. Response surface methodology (RSM) was used to optimize the fermentation medium using the variables such as corn steep liquor (5–25 g l⁻¹), Na₂HPO₄ 2H₂O (2.2–6.2 g l⁻¹), KH₂PO₄ (0.5–2.5 g l⁻¹), sucrose (5–55 g l⁻¹) and inoculum concentration (1–25 ml l⁻¹). Central composite rotatable design (CCRD) experiments were carried out to study the complex interactions of the variables. The optimum conditions for maximum PHB production were (g l⁻¹): CSL-25, Na₂HPO₄ 2H₂O-2.2, KH₂PO₄ − 0.5, sucrose − 55 and inoculum − 10 (ml l⁻¹). After 72 h of fermentation, the amount of PHA produced was 8.20 g l⁻¹ (51.20% of dry cell biomass). It is the first report on optimization of fermentation medium using CSL as a nitrogen source, for PHB production by Bacillus sp.     

Nutritional regulation of keratinolytic activity in Bacillus pumilis

Bacillus pumilis F3-4 utilized feather as a sole source of carbon, nitrogen and sulfur. Supplementation of the feather medium with glucose or MgSO₄ · 7H₂O increased keratinolytic protease production (14.6–16.7 U/mg). The synthesis of keratinolytic protease was repressed by an exogenous nitrogen source. Keratinolytic protease was produced in the absence of feather (9.4 U/mg). Feather degradation resulted in sulfhydryl group formation (0.8–2.6 μM). B. pumilis F3-4 effectively degraded chicken feather (75%), duck feather (81%) and feather meal (97%), whereas human nails, human hair and sheep wool under went less degradation (9–15%). An erratum to this article can be found at