Original Research Paper

Optimization of medium composition for alkaline protease productionby Marinobacter sp. GA CAS9 using response surfacemethodology – A statistical approach

Ramamoorthy Sathish Kumar, Gnanakkan Ananthan n, Antonyraj Selva PrabhuCentre of Advanced Study in Marine Biology, Faculty of Marine Sciences, Annamalai University, Parangipettai, 608502 TN, India

a r t i c l e i n f o

Article history:Received 29 October 2013Accepted 12 November 2013

Keywords:Marinobacter sp.Polyclinum glabrumAscidianAlkaline proteaseResponse surface methodology

a b s t r a c t

In the present study, the alkaline protease producing Marinobacter sp. GA CAS9 was isolated from themarine ascidian Polyclinum glabrum and identified by 16S rRNA analysis. Medium components andculture conditions for alkaline protease production were optimized using statistical optimization.Plackett–Burman design was employed to find out the optimal medium constituents and cultureconditions to enhance protease production. Central composite design revealed that four independentvariables, such as NaCl (60.53 g/l), beef extract (14.73 g/l), CuSO4 (4.73 g/l) and pH (10.7) significantlyinfluenced the protease production. Protease production obtained experimentally coincident with thepredicted value and the model was proven to be adequate. The enhancement of protease from 298.34 U/ml to 982.68 U/ml was achieved with the optimization procedure.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Microbial proteases are important class of enzymes, whichconstitute more than 65% of the total industrial enzyme market.Alkaline proteases have extensive applications in various indus-trial such as detergents, food, leather, silk and pharmaceuticals(Mukherjee et al., 2008; Rai and Mukherjee, 2009). Alkalineproteases are produced by a wide range of microbes, includingbacteria, molds, yeasts, and mammalian tissues (Mabrouk et al.,1999; Joo et al., 2002). Considering the commercial significanceof proteases, there were some attempts to study and maximizeprotease production and economize them in detergents(Chauhan and Gupta, 2004). For the prospective uses of pro-teases and their high demand, the need exists for the inventionof new strains of marine bacteria that produce enzymeswith novel properties and the development of low cost indus-trial media formulations (Annamalai et al., 2013; Esakkiraj et al.,2011).

Optimization of media components by classical methods whichinvolves the change of single variable optimization strategy hassome disadvantages, such as time consuming, requirement ofmore experimental data sets, and missing the interactions amongvariables (Cazetta et al., 2007; Li et al., 2008). Owing to thesedisadvantages, it has been replaced by statistical optimizationsuch as response surface methodology, which is an efficient

experimental strategy to seek optimal conditions for the multi-variable system. This method has been successfully applied for theoptimization of multiple variables in many fermentation processesand showed satisfactory results (Montgomeryd and Runger, 2002).

Marine environment provides various niche that are stillpristine and yet to be revealed by their biotechnological potential.One such pristine and diverse group of sessile marine inverte-brates is ascidians that serve as host for a range of microbes.Ascidians have a symbiotic relationship with different microor-ganisms which can be prosperous candidates for protease produc-tion (Tatian et al., 2002). In view of that, the present study wasaimed to optimize the significant variables such as NaCl, beefextract, CuSO4 and pH on yield of protease production fromascidian associated Marinobacter sp. GA CAS9 using Plackett–Burman design and the response surface methodology.

2. Materials and methods

2.1. Ascidian associated bacteria

Marine ascidian Polyclinum glabrum was collected from Tuti-corin coast, Southeast coast of India by SCUBA divers at a depth of10–15 m. The ascidan associated bacteria were isolated usingZobell marine agar with Amphotericin B (30 mg/ml) was added toinhibit the fungal contamination and the plates were incubated at40 1C for 5 days in dark. Isolated colonies were screened forprotease production on skim milk agar plates. The bacterial strainGA CAS9 produced the highest clear zone that was considered as

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/bab

Biocatalysis and Agricultural Biotechnology

1878-8181/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.bcab.2013.11.005

n Corresponding author. Tel.: þ91 4144 243223; fax: þ91 4144 243555.E-mail address: casananthan@yahoo.co.in (G. Ananthan).

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

potential strain and used for further study. The strain GA CAS9was identified by employing morphological and biochemicalcharacteristics (Holt et al., 1994) and confirmed through molecularcharacterization. Briefly, the genomic DNAwas extracted (Marmur,1961) and nearly full length 16S rRNA sequences were amplifiedby using primers 8F (5′-AGA GTTTGA TCC TGG CTC AG-3′) and1492R (5′-GGT TAC CTT GTT ACG ACTT-3′). PCR was performedunder the following conditions; 35 cycles consisting of initialdenaturation at 95 1C for 5 min, denaturation at 95 1C for 30 s,annealing at 55 1C for 30 s and followed by final extension of 5 minat 72 1C. The 16S rRNA gene sequences were obtained by anautomated DNA sequencer (Megabace, GE) and homology wasanalyzed with sequences in the Gene Bank using the CLUSTAL Xsoftware. The phylogenetic tree was constructed by the neighbor-joining method (Saitou and Nei, 1987). The 16S rRNA genesequence obtained from the GA CAS9 was deposited into GenBank(NCBI).

2.2. One factor at a time experiments

The media optimization experiment was initiated by enrichingin Zobell marine broth supplemented with casein (0.5%) for 3 daysat 40 1C. Ten percent of the enriched culture was inoculated in250 ml flask containing 45 ml basal medium (%w/v) (Casein 1.0,glucose 0.5, K2HPO4 0.1, NaCl 3.0, MgSO4 �7 H2O 0.01 and NH4NO3

0.1 at pH 9) and incubated in a shaker (150 rpm) for three days at40 1C. The cells were harvested by centrifugation at 10,000 rpm for15 min and the supernatant was further used for protease assay.

Initial screenings of the most significant carbon, nitrogensources and metal ions to maximize protease production wereperformed by one-variable-at-a-time approach. Carbon sources

such as fructose, lactose, sucrose, xylose, soluble starch andmaltose and nitrogen sources such as peptone, skim milk powder,beef extract, meat extract, potassium nitrate, ammonium chlorideand sodium nitrate were also individually tested at a concentrationof 0.5% in basal medium and various metal ions (zinc sulfate, fericchloride and copper sulfate) were tested for protease production.

2.3. Plackett–Burman design

Plackett–Burman design is an excellent way for screen the mainphysicochemical parameters from the large number of processvariables, which is required for prominent protease production. Inthe present study, the nutrients were selected based on the resultsobtained in one variable at time experiments, NaCl, lactose, beefextract, CuSO4 and additionally initial pH, incubation time andincubation temperature were selected to be the major variables inthe protease production. The Plackett–Burman method allowsevaluation of N variables in Nþ1 experiment, each factor wasexamined in two levels: �1 for a low level and þ1 for a highlevel (Table 1) and the seven variables were evaluated in 12experimental trials (Table 2). The design was run in a single blockand the order of the experiments was fully randomized. Thedesign was developed by the Minitab package version 15.

2.4. Response surface methodology

Based on the selection of the significant variables for proteaseproduction by Plackett–Burman design experiment, the significantvariables were selected as follows: NaCl, beef extract, CuSO4

and pH. Once the ranges of relevant variables were selected, theresponse surface methodology, using a central composite design,

Table 1Variables and test levels for Plackett–Burman experiment.

No Variables Levels Effect Coefficient t-Value p-Value

�1 þ1

Constant 373.73 185.21 0.000X1 NaCl (g/l) 30 50 �39.43 �19.72 �9.77 0.001X2 Lactose (g/l) 5 10 11.93 5.97 2.96 0.042X3 Beef extract (g/l) 5 10 72.21 36.11 17.89 0.000X4 CuSo4 (g/l) 0.5 1.0 32.93 16.47 8.16 0.001X5 Incubation temperature (1C) 30 50 �22.17 �11.09 �5.49 0.005X6 pH 7 11 55.86 27.93 13.84 0.000X7 Incubation time (h) 40 72 �9.38 �4.69 �2.32 0.081

Table 2Plackett–Burman design for seven variables with coded values along with the predicted and observed results.

Run X1 X2 X3 X4 X5 X6 X7 Protease yield (U/ml)

Observed Predict

1 1 �1 1 �1 �1 �1 1 354.018 346.1592 1 1 �1 1 �1 �1 �1 326.681 328.1893 �1 1 1 �1 1 �1 �1 384.230 384.7284 1 �1 1 1 �1 1 �1 440.441 444.3285 1 1 �1 1 1 �1 1 298.146 296.6386 1 1 1 �1 1 1 �1 401.654 401.1567 �1 1 1 1 �1 1 1 487.240 486.3158 �1 �1 1 1 1 �1 1 391.453 396.3509 �1 �1 �1 1 1 1 �1 397.234 389.37410 1 �1 �1 �1 1 1 1 303.162 307.63211 �1 1 �1 �1 �1 1 1 380.245 381.17012 �1 �1 �1 �1 �1 �1 �1 320.291 322.755

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

was used to determine the optimum concentration of thesevariables for protease production. A total of 31 experiments wereformulated using the Minitab package version 15. The centralvalues of all variable were coded as zero. The minimum andmaximum ranges of the variables were used, and the full experi-mental plan with regard to their values in actual and coded formis provided in Table 3. The data obtained from the RSM on proteaseproduction were subjected to analysis of variance (ANOVA). Theexperimental results of RSM were fitted with the response surfaceregression procedure using the following second order polynomialequation:

Y ¼ β0þ∑βiXiþ∑βiiX2i þ∑βijXiXj

where Y is the predicted response, xi and xj are independentfactors, β0 is the intercept, βi is the linear coefficient, βii is thequadratic coefficient, and βij is the interaction coefficient.

2.5. Protease assay

The protease assay was carried out based on the tyrosinereleased under standard assay conditions. The amount of proteaseproduced was measured with the help of a tyrosine standardgraph (Takami et al., 1989).

3. Results and discussion

In the present study, a total of 28 ascidian associated marinebacterial strains were isolated and screened for protease production.Among these, the strain GA CAS9 was considered as potentialprotease producer and used for further identification study. The

strain GA CAS9 was characterized as a gram negative, motile, rodshaped, glucose, sucrose, maltose and catalase positive (Table 4).Based on these characteristics, the isolate was identified as genusMarinobacter. The phylogenetic analysis (Fig. 1) and BLAST searchconfirmed that the isolate as Marinobacter sp. 16S rRNA sequence ofGA CAS9 has been deposited in the Gen-Bank database (Gen Bank ID:JX627399).

3.1. Selection of physical and nutritional parameters on proteaseproduction

A series of experiments were carried out to study the effects ofvarious simple and complex carbon, nitrogen and metal ions onprotease production by Marinobacter sp. GA CAS9. Cultures wereconducted using basal medium containing casein 1.0%, glucose0.5%, K2HPO4 0.1%, NaCl 3.0%, MgSO4 �7H2O 0.01% and NH4NO3

0.1% at pH 9. The optimization of suitable carbon sources forprotease production revealed that lactose (298.34 U/ml) was foundas an excellent source for maximum protease production (Fig. 2).Vijayaraghavan et al. (2012) reported that lactose and xylose werethe best carbon sources for protease production by Halobacteriumsp. while glucose and sucrose were not very effective.

In the present study, the organic and inorganic nitrogensources were used to evaluate their ability for higher proteaseproduction. The results revealed that organic nitrogen sourcesregistered higher protease production than inorganic nitrogensources. The highest protease production was found with beefextract (249.18 U/ml), with peptone and skim milk powder(Fig. 3). The reason may be that inorganic nitrogen compoundsserve only as a nitrogen source, whereas organic nitrogensources were containing multinutrient sources such as freeamino acids, carbohydrates, essential fatty acids, etc. Theseconstituents serve as multigrowth inducer for protease produc-tion (Esakkiraj et al., 2011). Prakasham et al. (2006) and Anandanet al. (2007) suggested that the organic nitrogen source highlyinduced protease production for marine bacteria. The traceelements and metal ions are one of the required cofactors forenzyme production. The present results confirmed that thecopper sulfate (232.87 U/ml) and potassium nitrate also inducedthe protease production (Fig. 4).

Table 3Experimental design and results of central composite designn optimizationexperiment.

Trials Coded variable level Protease yield (U/ml)

X1 X3 X4 X6 Observed Predict

1 50 10 1.0 8.0 425.450 425.3052 70 10 1.0 8.0 438.322 435.8243 50 20 1.0 8.0 411.765 412.4864 70 20 1.0 8.0 378.621 380.8995 50 10 10.0 8.0 362.345 362.9566 70 10 10.0 8.0 372.377 372.9677 50 20 10.0 8.0 395.207 394.1948 70 20 10.0 8.0 362.654 362.1009 50 10 1.0 13.0 420.421 421.489

10 70 10 1.0 13.0 431.381 432.33911 50 20 1.0 13.0 400.183 399.53812 70 20 1.0 13.0 368.379 368.28213 50 10 10.0 13.0 449.456 447.12314 70 10 10.0 13.0 457.672 457.46515 50 20 10.0 13.0 466.216 469.22816 70 20 10.0 13.0 437.374 437.46417 40 15 5.5 10.5 193.379 192.97018 80 15 5.5 10.5 171.774 171.72419 60 5 5.5 10.5 327.207 328.41520 60 25 5.5 10.5 297.263 295.59621 60 15 �3.5 10.5 338.431 337.84022 60 15 14.5 10.5 344.543 344.67423 60 15 5.5 5.5 773.297 773.53124 60 15 5.5 15.5 845.774 845.08025 60 15 5.5 10.5 823.617 909.36826 60 15 5.5 10.5 923.132 909.36827 60 15 5.5 10.5 923.311 909.36828 60 15 5.5 10.5 923.617 909.36829 60 15 5.5 10.5 924.521 909.36830 60 15 5.5 10.5 923.688 909.36831 60 15 5.5 10.5 923.688 909.368

Table 4Biochemical characteristics of Marinobacter sp.GA CAS9.

Biochemical test Results

Shape RodGram stain –

Spore formation –

Motility þGlucose þSucrose þGlycerol –

Maltose þStarch hydrolysis þGelatin hydrolysis þUrease –

Lipase –

Casein hydrolysis þAmylase –

Catalase activity þOxidase þIndole þCitrate þ

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

3.2. Plackett–Burman design

As described above, protease production is greatly influencedby the media composition and physical factors. A total of sevenfactors such as lactose, beef extract, CuSO4, NaCl, pH, incubationtemperature and incubation time were chosen for initial screeningby Plackett–Burman design. Each factor was made at the highlevel (þ1) and at the low level (�1) of that factor (Table 1).The protease production had a wide variation from 298.146 to487.240 U/ml (Table 2). This variation reflects the significance ofmedium optimization to attain good productivity. Factors having aconfidence level greater than 95% and significant ‘p’ value (lessthan 0.05) were considered to have a significant effect on theprotease production (Table 1). Based on the results, NaCl, beefextract, CuSO4 and initial pH were found to be the most significantvariables affecting protease productivity.

Fig. 1. Phylogenetic tree of isolate Marinobacter sp. GA CAS9 and their closest NCBI (BLAST) strains based on the 16S rRNA gene sequences.

0

50

100

150

200

250

300

350

Fructose Lactose Sucrose Xylose soluble starch Maltose

Prot

ease

pro

duct

ion

(U/m

l)

Carbon sources

Fig. 2. Effect of different carbon sources on protease production.

0

50

100

150

200

250

300

Peptone Skim milk powder

Beef extract Meat extract

Potassium nitrate

Ammonium chloride

Sodium nitrate

prot

ease

pro

duct

ion

(U/m

l)

Nitrogen sources

Fig. 3. Effect of various nitrogen sources on protease production.

0

50

100

150

200

250

300

Zinc sulphate

Feric chloride

Copper sulphate

Potassium chloride

Barium chloride

Mercuric chloride

Zinc chloride

Prot

ease

pro

duct

ion

(U/m

l)

Metal ions

Fig. 4. Effect of different metal ions on protease production.

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

3.3. Response surface methodology

Among all these number of other studies response surfacemethodology is a very powerful technique in the optimizationof bioprocesses technology (Adinarayana and Ellaiah, 2002; Puriet al., 2002). In the present study, central composite design withfour factors and six levels, including five replicates at the centerpoint, was used for fitting a second-order response surface. Thedesign matrix and the corresponding experimental data wereshown in Table 3. The mathematical model relating to the proteaseproduction with the independent process variables, X1, X2, X3 andX4, is given in the second-order polynomial equation.

Y ¼ 7715:85þ220:71X1þ189:43X2

þ49:83X4þ82:99X6�1:82X21þ5:97X2

3

�7:01X24�4:00X2

6�0:21X1X3�0:00X1X4þ0:00X1X6

þ0:49X3X4�0:18X3X6þ1:96X4X6

where Y is the predicted protease yield, X1 is NaCl, X2 is beef extract,X3 is CuSO4 and X4 is initial pH. The statistical significance wasevaluated by performing F-test and ANOVAwith the Minitab packageversion 15. The high F-value and non-significant lack of fit indicatethat the model is a good fit. Also significant p-values (0.000)suggested that the obtained experimental data was a good fit withthe model and the fit of the model was also checked by determina-tion of coefficient (R2) with R2 (multiple correlation coefficient) of99.53%. The predicted R2 and the adjusted R2 was about 99.35% and99.12% respectively (Table 5). The regression coefficients of all linear,quadratic terms and two cross products are significant at 1% level inTable 6. Fig. 5 shows the response surface plots of protease produc-tion. The main aim of response surface is to efficiently hunt for the

optimum values of the variables, such that the response is max-imized. Each figure presents the effect of two factors while the otherfactor was held at zero level. The central optimum point wasevaluated using the gradient method in the direction of the steepestascend of media for the protease production evaluated from thesurface plots. Ramirez et al., (2001) reported using cost effectivemedia formulation and optimizing it at its minimum requirements formaximum enzyme production is extremely important in industrialscale for economic reasons. Therefore, using common industrialingredients such as beef extract, soybean meal, maltose and tween80,which are known for their low cost and then optimizing the processusing response surface methodology will serve as a potential examplefor the applications used in industrial microbial fermentations(Rodrigues et al., 1998).

The results predicted by the model showed that the maximumprotease production could be achieved when NaCl, beef extract,CuSO4 and initial pH were set at 60.53, 14.73, 4.73 and 10.7 g/lrespectively. The maximum predicted value of protease yield obtainedwas 905.42 U/ml. In order to confirm the optimization results, thesuggested medium components and condition was performed. Underthese suggested conditions, the mean value of the protease yield was982.68 U/ml, which was in agreement with the predicted value. Thisoptimization strategy led to the enhancement of protease from298.34 U/ml to 982.68 U/ml. Moreover, the protease production byMarinobacter sp. GA CAS9 using the response surface methodology inthe present study was considerably higher than the those producedby Bacillus subtilis DM-04 (Ashis et al., 2011), Bacillus sp. JER02(Badoei-Dalfard and Zahra, 2013), Serratia marcescens P3 (Bach et al.,2012). Similar protease production was obtained by Bacillus sp. AS-S20-1 (Rai and Mukherjee, 2010). Thus, only few reports are availablefor optimization of culture condition in case of haloalkaliphilicproteases viz. 4.6 U/ml from Salinivibrio sp. strain AF-2004(Amoozegar et al., 2007), 61.0 U/ml from B. halodurans (Ibrahimet al., 2007) and 104 U/ml from Chromohalobacter sp. TVSP10(Vidyasagar et al., 2007).

4. Conclusion

To the best of our knowledge it is the first report regarding theoptimization of protease production from ascidian associatedMarinobacter sp. In this study, the Plackett–Burman design andthe response surface methodology were employed to optimizemedium composition and culture conditions for production ofalkaline protease. The optimal combinations of media constituentsand culture conditions for maximum protease production were

Table 6Results of regression analysis of the second-order polynomial model for optimization of protease production.

Factor Coefficient Estimate coefficient t-Value p-Value

Constant �7715.85 276.601 �27.895 0.000X1 220.71 6.067 36.378 0.000X3 189.43 10.111 18.734 0.000X4 49.83 10.525 4.734 0.000X6 82.99 21.600 3.842 0.001

X21

�1.82 0.043 �41.883 0.000

X23

5.97 0.174 34.414 0.000

X24

�7.01 0.214 �32.728 0.000

X26

�4.00 0.694 �5.764 0.000

X1� 3 �0.21 0.116 �1.814 0.088X1� 4 �0.00 0.129 �0.022 0.983X1� 6 0.00 0.232 0.014 0.989X3� 4 0.49 0.258 1.899 0.006X3�6 �0.18 0.464 �0.394 0.699X4� 6 1.96 0.516 3.791 0.002

Table 5Analysis of variance for the fitted quadratic polynomial model for optimization ofprotease production.

Source DF Seq SS Adj SS MS F-value p-value

Regression 14 1817539 1817539 129824 241.08 0.000Linear 4 10042 818033 204508 379.76 0.000Square 4 1795959 1795959 448990 833.75 0.000Interaction 6 11539 11539 1923 3.57 0.019Residual error 16 8616 8616 539Lack-of-fit 10 36 36 4 0.00 1.000Pure error 6 8580 8580 1430Total 30 1826155

R2¼99.53%, R2 (adj)¼99.12%, R2 (pred)¼99.35%, SS, sum of squares; DF, degrees offreedom; MS, mean square.

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

determined as NaCl (60.53 g/l), beef extract (14.73 g/l), CuSO4

(4.73 g/l) and pH (10.7). The alkaline protease yield increased from298.34 U/ml to 982.68 U/ml in the shake flask system. The resultsobtained in this work indicated that RSM is a reliable method fordeveloping the model, optimizing factors, and analyzing interac-tion effects, before using protease in fermentation industry.

Acknowledgment

The authors are very much greatful to the Dean and Directorand authorities of Annamalai University for providing the facilities.The author also thanks to the Ministry of Environmental andForest (MoEnF) for fianancial support.

Fig. 5. Three dimensional response surface plot for protease production showing the interactive effects of the copper sulfate and pH (A), beef extract and copper sulfate (B),beef extract and pH (C), NaCl and copper sulfate (D), NaCl and pH (E), NaCl and beef extract (F).

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎6

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i

References

Adinarayana, K., Ellaiah, P., 2002. Response surface optimization of the criticalmedium components for the production of alkaline protease by a newlyisolated Bacillus sp. J. Pharm. Sci. 5, 272–276.

Amoozegar, M.A., Fatemi, A.Z., Karbalaei-Heidari, H.R., Razavi, M.R., 2007. Produc-tion of an extracellular alkaline metalloprotease from a newly isolated,moderately halophile Salinivibrio sp. Strain AF-2004. Microbiol. Res. 62,369–377.

Anandan, D., Marmer, W.N., Dudley, R.L., 2007. Isolation, characterization andoptimization of culture parameters for production of an alkaline proteaseisolated from Aspergillus tamari. Ind. Microbiol. Biotechnol. 34, 339–347.

Annamalai, N., Rajeswari, M.V., Thavasi, R., 2013. Optimization, purification andcharacterization of novel thermostable, haloalkaline, solvent stable proteasefrom Bacillus halodurans CAS6 using marine shellfish wastes: a potentialadditive for detergent and antioxidant synthesis. Bioprocess Biosyst. Eng. 7,873–883.

Ashis, K., Mukherjee, Sudhir, K., Rai., 2011. A statistical approach for the enhancedproduction of alkaline protease showing fibrinolytic activity from a newlyisolated Gram-negative Bacillus sp. strain AS-S20-I. N. Biotechnol..

Bach, E., Anna, V.S., Daroit, D.J., Corrêa, A.P.F., Segalin, J., Brandelli, A., 2012.Production, one-step purification, and characterization of a keratinolyticprotease from Serratia marcescens P3 Process. J. Biochem., 2455–2462.

Badoei-Dalfard, A., Zahra, K., 2013. Screening and isolation of an organic solventtolerant-protease from Bacillus sp. JER02: Activity optimization by responsesurface methodology. J. Mol. Catal. B: Enzym. 89, 15–23.

Cazetta, M.L., Celligoi, M.A., Buzato, J.B., Scarmino, I.S., 2007. Fermentation ofmolasses by Zymomonas mobilis: effects of temperature and sugar concentra-tion on ethanol production. Bioresour. Technol. 98, 2824–2828.

Chauhan, B., Gupta, R., 2004. Application of statistical experimental design foroptimization of alkaline protease production from Bacillus sp. RGR-14. ProcessBiochem. 39, 15–22.

Esakkiraj, P., Sankaralingam, S., Usha, R., Palavesam, A., Immanuel, G., 2011. Solid-state protease production using anchovy waste meal by moderate halophileSerratia proteamaculans AP-CMST isolated from fish intestine. Ann. Microbiol.61, 749–755.

Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Bergey’s Manualof Determinative Bacteriology, 9th ed. Williamsons and Wilkins, Baltimore.

Ibrahim, A.S.S., Nefisa, M.A., Shayeb, E.I., Mabrouk, S.S., 2007. Isolation andidentification of alkaline protease producing alkaliphilic bacteria from anEgyptian soda lake. J. Appl. Sci. Res. 3, 1363–1368.

Joo, H.S., Kumar, C.G., Park, G.C., Kim, K.T., Paik, S.R., Chang, C.S., 2002. Optimizationof the production of an extracellular alkaline protease from Bacillus horikoshi.Process Biochem. 38, 155–159.

Li, X., Xu, T., Ma, X., Guo, K., Kai, L., Zhao, Y., Jia, X., Mad, Y., 2008. Optimization ofculture conditions for production of cis-epoxysuccinic acid hydrolase usingresponse surface methodology. Bioresour. Technol. 99, 5391–5396.

Mabrouk, S.S., Hashem, A.M., El-Shayeb, N.M.A., Ismail, A.S., Abdel-Fattah, A.F.,1999. Optimization of alkaline protease productivity by Bacillus licheniformisATCC 21415. Bioresour. Technol. 69, 155–159.

Marmur, J., 1961. A procedure for the isolation of deoxyribonucleic acid frommicroorganisms. Mol. Biol. J. 3, 208–218.

Montgomeryd, D.C., Runger, G.C., 2002. Applied Statistics and Probability forEngineers, 3rd ed. John Wiley and Sons, New York.

Mukherjee, A.K., Mukherjee, Ashis K., Sudhir, H.A., Rai, K., 2008. Production ofprotease by a thermophilic Bacillus subtilis under solid-state fermentation (SSF)condition using Imperata cylindrical grass and potato peel as low-cost medium:characterization and application of enzyme in detergent formulation. Biochem.Eng. J. 39, 353–361.

Prakasham, R.S., Subba Rao, C.h, Sarma, P.N., 2006. Green gram husk-an inexpensivesubstrate for alkaline protease production by Bacillus sp. in solid-state fermen-tation. Bioresour. Technol. 97, 449–1454.

Puri, S., Khali, O., Gupta, R., 2002. Optimization of alkaline protease production fromBacillus sp. by response surface methodology. Curr. Microbiol. 44, 286–290.

Rai, S.K., Mukherjee, A.K., 2009. Ecological significance and some biotechnologicalapplication of an organic solvent stable alkaline serine protease from Bacillussubtilis strain DM-04. Bioresour. Technol. 100, 2642–2645.

Rai, S.K., Mukherjee, A.K., 2010. Statistical optimization of production, purificationand industrial application of a laundry detergent and organic solvent-stablesubtilisin like serine protease (Alzwiprase) from Bacillus subtilis DM-04.Biochem. Eng. J. 48, 172–180.

Ramirez, R., Gutierrez, H., Gschaedler, A., 2001. Optimization of astaxantinproduction by Phaffia rhodozyma through factorial design and response surfacemethodology. J. Biotechnol. 88, 259–268.

Rodrigues, D.C., Silva, S.S., Felipe, M.G.A.., 1998. Using response surface methodol-ogy to evaluate xylitol production by Candida guilliermondii by the fed-batchprocess with exponential feeding rate. J. Biotechnol. 62, 73–77.

Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method forreconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406–425.

Takami, H., Akiba, T., Horikaoshi, K., 1989. Production of extremely thermostablealkaline protease from Bacillus Sp. no. AH-101. Appl. Microbiol. Biotechnol. 30,120–124.

Tatian, M., Sahade, R., Kowalke, J., Kivatinitz, S.C., Esnal, G., 2002. Food availabilityand gut contents in the ascidian cnemidocarpa verrucosa at potter cove,antarctica. Polar Biol. 25, 58–64.

Vidyasagar, M., Prakash, S., Jayalakshmi, S.K., Sreeramulu, K., 2007. Optimization ofculture conditions for the production of halothermophilic protease fromhalophilic bacterium Chromohalobacter sp TVSP101. World J. Microbiol. Bio-technol. 23, 655–662.

Vijayaraghavan, P., Jebamalar, T., Prakash Vincent, S.G., 2012. Biosynthesis optimi-zation and purification of a solvent stable alkaline serine protease fromHalobacterium sp. Ann. Microbiol. 62, 403–410.

S.K. Ramamoorthy et al. / Biocatalysis and Agricultural Biotechnology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Please cite this article as: Sathish Kumar, R., et al., Optimization of medium composition for alkaline protease production by.... Biocatal.Agric. Biotechnol. (2013), http://dx.doi.org/10.1016/j.bcab.2013.11.005i