- Open Access
UV mutagenesis treatment for improved production of endoglucanase and β-glucosidase from newly isolated thermotolerant actinomycetes, Streptomyces griseoaurantiacus
© Kumar; licensee Springer. 2015
- Received: 21 February 2015
- Accepted: 30 April 2015
- Published: 19 May 2015
Bioconversion of cellulosic biomass into fuel ethanol involves several steps, among which enzymatic breakdown of cellulose into fermentable sugars play a significant role. The key enzymes involved in cellulosic breakdown are mainly endoglucanases and β-glucosidases. Even though the biochemical and molecular characterization of number of endoglucanases and β-glucosidases was extensively studied, still there is a demand for novel microbial cellulases for industrial applications. Among the group of actinomycetes, Streptomyces spp. are well known as a cellulase producer. The advantage of using actinomycetes is being that production process could be easily scaled-up to commercial levels. However, recent research studies have shown that the production of cellulases from actinomycetes could also be significantly improved by employing different types of strain improvement methods, thus achieving high yields of extracellular proteins. Besides this, highly thermostable and broad pH range cellulases are required for bioethanol application.
A lignocellulose degrading actinomycetes strain was newly isolated and identified as Streptomyces griseoaurantiacus. Strain improvement using UV mutagenesis developed two mutants (SGUV30 and SGUV5) with 57.4 % and 12.8 % higher endoglucanase and β-glucosidase activities. The cellulases (endoglucanases and β-glucosidases) were found to be highly thermostable with no loss in enzyme activities at 80 °C for 60 min and nearly 80 % of initial activity was retained at 90 °C. Enzyme assays in presence of additives showed that CoCl2, CaCl2, and FeSO4 increased β-glucosidase activity but showed negative effect on endoglucanase activity. However, both the enzyme activities were significantly enhanced by addition of PEG 8000, sodium azide and MnSO4.
Strain improvement of S. griseoaurantiacus was performed by UV mutagenesis where two mutant strains (SGUV30 and SGUV5) were developed with improved endoglucanase and β-glucosidase activities. Cellulase production in submerged fermentation was carried out using a cheap lignocellulosic biomass residue, rice straw as a sole source carbon. The results clearly show that the mutant strains produced high-efficient cellulases that are stable at a broad pH range at very high temperatures. Besides, the mutants also showed high extracellular protein secretions, which could be promising in reducing the overall cellulase production costs at large scale.
- Lignocellulosic biomass
Lignocellulosic biomass is available in plenty and is an inexpensive renewable bioresource for bioconversion to biofuels. However, this conversion is quite difficult owing to the complexity of the plant cell wall materials that are innately designed to resist microbial degradation. Among these, celluloses are one of the major structural components which could be converted into fuel ethanol with the help of microbial cellulose degrading enzymes, namely cellulases. This group of enzymes catalyze the hydrolysis of cellulose and other cellooligosaccharides into fermentable sugars. Primarily, but not limited to, the cellulase system consists of three major enzymes, viz. exoglucanase (avicelase), endoglucanase (CMCase) and β-glucosidase (cellobiase). In nature, a number of bacteria and fungi produce cellulases to hydrolyze the insoluble polysaccharides to soluble oligomers and subsequently to monomeric sugars . The production and application potential of diverse microbial cellulases is largely explored using a variety of growth substrates, and voluminous research reports and review articles are published in this area [2–5]. Actinomycetes, especially Streptomyces spp, are known to produce cellulose-degrading enzymes and have attracted considerable interest among the researchers due to their potential applications in the recovery of fermentable sugars from the hydrolysis mixture. Streptomyces spp. are also capable of producing an array of different extracellular enzymes including cellulases, xylanases, and chitinases . Synergistic action of these enzymes is vital for complete enzymatic hydrolysis of cellulose . During the recent past the intense basic and applied research studies revealed the commercial significance and industrial applicability of novel potential cellulases to a greater extent [8-10]. The production economics of cellulosic ethanol from lignocellulosic residues is largely dependent on the enzyme cost especially in the bioconversion processes . However, continuous efforts towards cost reduction have been directed for increasing the enzyme production levels, identification of hyperactive cellulose-degrading microbial strains, efficient ethanol fermentation techniques and enzyme recovery systems, etc. . Besides these, use of different mutagenic agents for microbial strain improvement and fermentation processes was demonstrated , where simultaneous treatments with N-methyl-N′-nitro-N-nitrosoguanidine (NTG), ethidium bromide, and UV either alone or in combination were employed for higher endoglucanases .
The present investigation is on isolation of cellulase-producing actinomycete strains from agricultural and plant waste residues such as decaying woods and aged tree trunk burrows and to screen the microbe for robust utilization of different locally available lignocellulosic biomass residues. We also report on increasing cellulase production levels by applying traditional UV mutagenesis treatment and biochemical characterization of the crude cellulase enzymes (endoglucanase and β-glucosidase) from two mutant strains.
Substrate, chemicals, and media
Rice straw was obtained from local agricultural fields and was used as carbon substrate for cellulase production. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) molecular weight markers were purchased from Bio-Rad Laboratories, Hercules, CA. Bovine serum albumin (BSA), p-nitrophenyl-β-d-glucopyranoside (PNPG), carboxymethyl cellulose (CMC), cellobiose, and 4-methylumbelliferyl-β-d-glucopyranoside (MUG) were purchased from Sigma (St. Louis, MO, USA). All other reagents were obtained from commercial sources and were of analytical grade.
Microbial isolation and screening
Cellulase producing microbes were isolated from diverse environmental places such as soil and water samples, decaying carton box scrapings, degrading wood pieces, and salt pans near the coastal marine areas of state of Andhra Pradesh, India. The samples (1 g each) were crushed with mortar and pestle and were then suspended in 10 ml of sterile distilled water. A 103 dilution of the samples was made and spread plated on IGA medium agar plates comprising 0.5 % peptone, 0.5 % yeast extract, 0.5 % malt extract, 0.5 % glycerol, 3 % 10× mineral salt medium, and 1.5 % agar. The composition of mineral salt medium is given in Additional file 1: Table S2. All plates were then incubated at 40 °C for at least 2–3 weeks. Further, the microbes are sub-cultured onto fresh agar plates in order to ensure the purity of the selected colonies. All the cultures were periodically maintained on sterile actinomycete isolation agar plates (HiMedia Laboratories, Mumbai, India) and stored at 4 °C for further use.
Cellulose-degrading microbes were screened using actinomycete isolation agar plates containing 0.5 % CMC and 10 mM MUG, separately. The plates were then incubated for 48 h at 40 °C. For endoglucanase production, the plates were stained with 0.05 % congo red dye, and for β-glucosidase production, the plates were directly viewed under UV light. A potential cellulase-producing microbe for the current investigation was selected based on the maximum cellulolytic index and intensity of fluorescence. The cellulolytic index value of the actinomycetes was calculated using the microbial colony size and the size of zone of clearance.
Among the potential cellulase-producing microorganisms, an actinomycete strain denoted by SG14 was selected in the present study. The genomic DNA was isolated using genomic DNA isolation kit (Bangalore Genei, India). Consensus primers (forward primer: 5′CGCGGCCTATCAGCTTGTTG′3 and reverse primer: 5′CCGTACTCCCCAGGCGGGG′3) were used for PCR amplification of 16S rRNA. The PCR product was then purified and subjected to DNA sequencing. The 16 SrRNA was submitted to NCBI database with GenBank ID KR094972. The nucleotide sequence was analyzed using Blastn, and multiple sequence analysis was performed using clustalX software using the top 20 consensus sequences obtained from the NCBI databank (>98 % sequence identity). A phylogenetic tree was prepared using neighbor-joining method with 1000 bootstrap replicates. Mega 5.0 software was used for tree analysis .
Media optimization and growth on different lignocellulosic residues
For cellulase production in submerged fermentation, seed inoculum was prepared in actinomycete broth medium (HiMedia Laboratories, Mumbai, India). Preliminary cellulase productions were tested using 1 % (w/v) CMC as carbon source in different growth media compositions viz. minimal salt media, Mandels mineral media , Czapek media , and Vogel’s mineral media . The detailed composition of growth media were given in Additional file 1: Tables S2 to S5. Production of cellulase was tested on different ligocellulosic biomass residues. The carbon sources tested were rice straw (RS), wheat bran (WB), rice husk (RH), wheat husk (WH), coco pith (CP), saw dust (SD), Walseth cellulose (WC), cardboard waste (CW), and newspaper waste (NW), respectively. Except WC, other carbon sources were desized to 0.5–1 cm using a hammer mill machine and subjected to extensive washing with distilled water and finally dried at 60 °C overnight until the moisture content was <5 %, while the WC substrate was prepared using commercial cellulose powder (Sigma Aldrich, USA) and pre-treated with 83 % (w/v) phosphoric acid for 25 h at 40 °C. The precipitate obtained from the Buchner funnel filtration method was then collected and washed twice with 70 % (v/v) ethanol and neutralized to pH 7.0. Finally, the precipitate was dehydrated by acetone rinsing and dried under vacuum. For submerged fermentation, 100 ml of minimal media was placed in a 500-ml Erlenmeyer flask supplemented with 1 % (w/v) rice straw. The growth media were inoculated with 4 % spore suspension and incubated at 40 °C at 160 rpm for 5 days. The culture broth was filtered through Whatman filter paper (9 μm size), and the filtrate was then centrifuged at 10,000 rpm for 30 min. The clear supernatant was collected and treated as the crude cellulase enzyme for biochemical analysis.
Determination of cellulase activities
Endoglucanase activity was assayed by incubating 0.5 ml of crude enzyme with 0.5 ml of 2 % CMC in citrate buffer (50 mM, pH 5.0) for 30 min at 60 °C. The reducing sugars generated were determined by standard dinitrosalicylic acid (DNSA) method . β-Glucosidase assay was carried out by mixing 0.2 ml of culture supernatant with 0.2 ml of 0.01 M PNPG and 1.6 ml of citrate buffer (50 mM, pH 5.0). The reaction mixture was then incubated for 30 min at 60 °C. Reaction was stopped by the addition of 4 ml NaOH-glycine buffer (0.2 M, pH 10.6) and measured by colorimetric method . The amount of enzyme required for liberating either 1 μg of glucose or 1 μg p-nitrophenol per minute under the standard assay conditions was considered as one unit of endoglucanase or β-glucosidase activity unit, respectively. All the assay experiments were carried out in triplicates and mean values are given.
Zymogram analysis using non-denaturing SDS-PAGE
Non-denaturing SDS-PAGE was performed for identification of endoglucanase and β-glucosidase proteins. Endoglucanase protein bands were detected using the congo red dye staining method of an overlaying gel containing 2 % CMC, while β-glucosidase active protein bands were detected in the gel by incubating with 10 mM MUG for 30 min at 45 °C, and then the gel was viewed directly under UV light. Briefly, a 10 % separating gel was prepared, and 30 μg of crude cellulase enzyme protein from native and UV-mutated samples was loaded. After electrophoresis, the gel was equilibrated with sodium citrate buffer (50 mM, pH 5.0) for 30 min and was overlayered on a CMC-containing polyacrylamide gel. This cassette was then enclosed in a plastic wrap and incubated at 60 °C for 2 h. Post incubation, the CMC-containing gel was peeled off and was stained with 0.2 % congo red solution for 1 h. To visualize the zone of clearance corresponding to the endoglucanase activity, destaining was performed with 1 M NaCl. The developed gels were visualized and digitally imaged. Protein concentration was determined using BSA as standard following the method described by Bradford .
Effect of UV treatment on cellulase production
Spore suspension from an overnight grown culture was prepared by serial dilution method. A 0.5-ml culture sample with a 106 dilution (approximately 105–106 spores ml−1) was spread plated onto a 0.5 % CMC agar plate under sterile conditions. Each petri plate was then exposed to a UV tube (6 W, 240 nm) with a constant 10-cm distance for a treatment period of 15 s to 60 min. The UV-treated petri plates were taken out at regular intervals (15, 30, and 45 ; 1,5,15,30, and 60 min) and incubated at 40 °C for 48 h. The survivor microbial colonies after the UV treatment were then selected for cellulase production by a similar zone of clearance method either using 0.05 % congo red dye as mentioned earlier. The selected colonies were further screened by analyzing the endoglucanase and β-glucosidase activities using rice straw as sole source of carbon in submerged fermentation.
Determination of optimum pH, temperature, and stability
The enzyme reactions were carried out separately in buffer solutions at varying pH (pH 2.0, pH 5.0, pH 7.0, pH 9.0, and pH 11.0) and temperature (40 to 100 °C with 10 °C increments). The endoglucanase and β-glucosidase activities were measured using the assay methods as mentioned in the above section. For pH stability studies, the enzyme samples were incubated in different test buffer solutions with varying pH (pH 2.0, pH 5.0, pH 7.0, pH 9.0, and pH 11.0) for 30 to 60 min at 60 °C, and then the residual enzyme activities were measured in sodium citrate buffer (50 mM, pH 5.0). For temperature stability studies, the enzyme samples were incubated at different test temperatures (60, 80, 90, and 100 °C) in sodium citrate buffer (50 mM, pH 5.0) for different time periods, and residual activities were measured.
Effect of metal ions and chemical reagents
The enzyme samples were mixed with different metal ions, chemical reagents, and solvents, and then the effect of these additives on the enzyme activity was analyzed. The test compounds used were (10 mM each) MgSO4, MgCl2, CoCl2, CaCl2, CuSO4, MnSO4, FeSO4, 0.02 % sodium azide, 0.8 % Tween 20, and 20 % PEG 8000. Control experiments were also analyzed voiding either the chemical agent or the enzyme to nullify the background absorbance on the activity measurements.
Microbial growth on rice straw
Cellulase production in different growth media
Enzyme activity (U ml−1)
2.7 ± 0.2
13.4 ± 0.5
Mandels mineral media
1.8 ± 0.4
6.0 ± 0.8
1.9 ± 0.2
10.5 ± 0.3
Vogel’s mineral media
2.2 ± 0.1
11.2 ± 0.6
UV mutagenesis and zymogam analysis
Mutational analysis was performed using traditional UV treatment of the actinomycete for increasing cellulase production. Based on the endoglucanase and β-glucosidase enzyme activities the prominent UV mutant strains were selected for further biochemical studies. However, initial screening of the mutants was established on the index ratio of the zone of clearance. The zone of clearance is indicated by the visible growth of the microbial colony to the degraded CMC agar plate as described in the “Methods” section. The isolated UV mutants were then tested subsequently in submerged fermentation for a minimum of 20 sub-culturing experiments. The variation in the production of endoglucanase and β-glucosidase and reproducibility of the results were evaluated. Similar results were obtained with all the tested sub-cultured experiments, and no significant alteration in cellulase production was observed. Maximum cellulase production from the mutant cells was obtained within 94 h as compared to the longer incubation period (120 h) with the native strain. One of the most intriguing features of S. gresioaurantiacus is its high resistance to the UV radiation for a maximum treatment period of 60 min. After 48 h growth incubation, >40 % of microbial cells was found to survive and form colonies after UV treatment. These data clearly indicated that the mutants were not only able to survive harsh conditions but also stable after several generations without any back mutation. Very few microbes in nature could survive such harsh treatments, and to date, no reports are available on such high-UV-tolerant actinomycetes. In most of the microbial mutagenesis studies, UV treatment was generally followed by a chemical treatment to avoid back mutation . However, our studies showed that these mutagens were stable without any drastic changes in the cellulase production profiles and corresponding enzyme activities for several subsequent generations indicating that the developed mutants were highly stable.
Effect of UV treatment on cellulase activities and protein production
UV treatment time (min)
Enzyme activity (U ml−1)
Protein (mg ml−1)
Effect of pH, temperature and additives on cellulase activity
Although ~36.6 % higher endoglucanase activity was observed in SGUV5 cells, the β-glucosidase activity was not found to be increased. This variation in activity levels could be mostly due to the UV-dependent effect on the cellulase-encoding genes, since endoglucanase and β-glucosidase are two different proteins and are expressed by separate genes. It clearly indicated that UV mutagenesis is a random mutation process. As the mutation occurs at the gene level and mostly involved in thymine dimerization, the translation process of some of the genes might have inhibited by the UV treatment. Further detailed molecular investigation and proteomic studies are required to understand these expressed endoglucanase and β-glucosidase proteins. Moreover, UV treatment showed significant effect on overall increase in the extracellular protein production. The protein concentration in the culture supernatant was increased from 0.081 to 0.115 mg ml−1 after UV treatment (Table 2). Besides this, the reason for low protein production with decreased enzyme activities at higher UV treatment periods may be due to the longer exposures of the microbe to harsh UV radiation which resulted in undesirable genetic modifications inside the cells .
The stability of the protein in different pH buffer systems was analyzed, and it was found that >70 % endoglucanase and >80 % β-glucosidase residual activities were retained after 60 min incubation period in pH 2.0 to pH 11.0 (Fig. 5c, d), except where no significant loss of endoglucanase and β-glucosidase activity was observed at pH 5.0 and pH 7.0, respectively. The data indicated that these proteins were highly stable in broad pH solutions. In contrary, the endoglucanase and β-glucosidase proteins from the native strain were not stable either above or below pH 5.0. The reason behind this drastic improvisation in pH stability from the mutant strains is not clearly known.
Since the cellulases are stable in a broad pH range, this enzyme cocktail could be used in diverse industrial applications where a specific acidic, basic, or neutral cellulase protein is needed to be supplemented during an individual process step. Moreover, single cellulase enzyme cocktail is preferred over separate cellulase protein with a specific pH condition, thus reducing the overall cost of the protein. In addition, the crude cellulase protein cocktail also showed broad pH stability with high potential in cellulosic ethanol production.
S. griseoaurantiacus was mutated by UV treatment, and two mutant strains (SGUV30 and SGUV5) were developed with improved endoglucanase and β-glucosidase production and activity. Rice straw was used as a cheap lignocellulosic residue for microbial growth for cellulase production in submerged fermentation. Both the native and mutant strains were able to utilize rice straw very efficiently. The UV mutants showed 57.4 % and 12.8 % higher endoglucanase and β-glucosidase activities compared to the wild-type enzymes. There was no loss in endoglucanase and β-glucosidase activities at 80 °C were found to be highly thermostable with no loss in enzyme activities at 80 °C for 60 min and nearly 80 % of initial activity was retained at 90 °C. Studies on purification of the endoglucanases and β-glucosidases from these mutant strains and their synergistic action on enzymatic saccharification will provide the industrial applicability in cellulosic ethanol production.
The author is thankful to the Director, SPRERI for providing the necessary facilities to carry out this research work. The financial support from the start-up research fund (SPRERI/AKK/RP-2) of SPRERI, Gujarat is highly acknowledged.
- Rathan RK, Ambili M (2011) Cellulase enzyme production by Streptomyces sp using fruit waste as substrate. Aust J Basic Appl Sci 5:1114–1118Google Scholar
- Horn SJ, Vaaje-Kolstad G, Westereng B, Eijsink VG (2012) Novel enzymes for the degradation of cellulose. Biotechnol Biofuel 5:45View ArticleGoogle Scholar
- Miettinen-Oinonen A, Londesborough J, Joutsjoki V, Lantto R, Vehmaanpera J (2004) Three cellulases from Melanocarpus albomyces for textile treatment at neutral pH. Enzym Microbial Technol 34:332–341View ArticleGoogle Scholar
- Sukumaran RK, Singhania RR, Pandey AK (2005) Microbial cellulases -production, application and challenges. J Sci Indus Res 64:832–844Google Scholar
- Bhat MK (2000) Cellulases and related enzymes in biotechnology. Biotechnol Adv 18:355–383View ArticleGoogle Scholar
- Manivasagan P, Gnanan S, Sivakumar K, Thangaradjou T, Vijaylaxmi S, Balasubramanian T (2010) Isolation, identification and characterization of multiple enzyme producing actinobacteria from sediment samples of Kodiyakarai coast, the Bay of Bengal. Afr J Microbiol Res 4:1550–1559Google Scholar
- Berg B, Laskowski R (2006) Litter decomposition: a guide to carbon and nutrient turnover, vol 38, CA, Adv. Ecol. Res. Elsevier Academic Press, San DiegoGoogle Scholar
- Bajpai P (1999) Applications of enzymes in the pulp and paper industry. Biotechnol Prog 15:147–157View ArticleGoogle Scholar
- Harman GE, Kubicek CP (1998) Trichoderma and Gliocladium: enzymes, biological control and commercial applications, vol 2. Taylor and Francis Ltd, London, pp 393Google Scholar
- Uhlig H (1998) Industrial enzymes and their applications. In: John Wiley & Sons, Inc, New York, pp 435Google Scholar
- Reith JH, den Uil H, van Veen H, de Laat WTAM, Niessen JJ, de Jong E, Elbersen HW, Weusthuis R, van Dijken JP, Raamsdonk L (2002) Co-production of bio-ethanol, electricity and heat from biomass residues. In: Proceedings of the 12th European conference on biomass for energy, industry and climate protection. 17–21 June 2002, Amsterdam, The Netherlands, 1118–1123.Google Scholar
- Xu ZH, Bail YL, Xu X, Shi JS, Tao WI (2005) Production of alkali-tolerant cellulase-free xylanase by Pseudomonas sp. UN024 with wheat bran as the main substrate. World J Microbiol Biotechnol 21:575–581View ArticleGoogle Scholar
- Parekh S, Vinci VA, Strobel RJ (2000) Improvement of microbial strains and fermentation processes. Appl Microbiol Biotechnol 54:287–301View ArticleGoogle Scholar
- Chand P, Aruna A, Maqsood AM, Rao LV (2005) Novel mutation method for increased cellulase production. J Appl Microbiol 98:318–323View ArticleGoogle Scholar
- Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28:2731–2739View ArticleGoogle Scholar
- Mandels M, Hontz I, Nystrom J (1974) Enzymatic hydrolysis of waste cellulose. Biotechnol Bioeng 16:1471–1493View ArticleGoogle Scholar
- Czapek F (1902) Beitr Chem Physiol Pathol 1:540Google Scholar
- Vogel HJ (1956) A convenient growth medium for Neurospora (medium N). Microbial Genetics Bulletin 13:42–43Google Scholar
- Dashtban M, Maki M, Leung KT, Mao C, Qin W (2010) Cellulase activities in biomass conversion: measurement methods and comparison. Crit Rev Biotechnol 30:302–309View ArticleGoogle Scholar
- Kubicek CP (1982) Beta-glucosidase excretion by Trichoderma pseudokoningii: correlation with cell wall bound beta-1,3-glucanase activities. Arch Microbiol 132:349–354View ArticleGoogle Scholar
- Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254View ArticleGoogle Scholar
- Jaradat Z, Dawagreh A, Ababneh Q, Saadoun I (2008) Influence of culture conditions on cellulase production by Streptomyces sp. strain J2. Jordan J Biol Sci 1:141–146Google Scholar
- Irfan M, Javed J, Syed Q (2011) UV mutagenesis of Aspergillus niger for enzyme production in submerged fermentation. Pak J Biochem Mol Biol 44:137–140Google Scholar
- Yao R, Li M, Deng S, Hu H, Wang H, Li F (2012) Mutagenesis of Trichoderma viride by ultraviolet and plasma. Plasma Sci Tech 14:353–356View ArticleGoogle Scholar
- Lui j, Feng Y, Yu Y, Zhou X, He W (2011) Selection of fungus with high ability of cellulase activity production using UV mutagenesis. International Conference on Consumer Electronics, Communications and Networks (CECNet), 1643–1645Google Scholar
- Ikehata H, Ono T (2011) The mechanisms of UV mutagenesis. J Radiat Res 52:115–125View ArticleGoogle Scholar
- Kluepfel D, Shareck F, Mondou F, Morosoli R (1986) Characterization of cellulase and xylanase activities of Streptomyces lividans. Appl Microbiol Biotechnol 24:230–234Google Scholar
- Jang HD, Chen KS (2003) Production and characterization of thermostable cellulases from Streptomyces transformant T3-1. World J Microbiol Biotechnol 19:263–268View ArticleGoogle Scholar
- Schrempf H, Walter S (1995) The cellulolytic system of Streptomyces reticuli. Int J Macromolecules 15:353–355View ArticleGoogle Scholar
- Maki M, Leung KT, Qin W (2009) The prospects of cellulase-producing bacteria for the bioconversion of lignocellulosic biomass. Int J Biol Sci 5:500–516View ArticleGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.