Pseudomonas and beyond

Polyamine metabolism, lignin degradation and potential applications in industrial biotechnology

Luaine Bandounas 2011

2

Pseudomonas and beyond

Polyamine metabolism, lignin degradation and potential applications in industrial biotechnology

Proefschrift

ter verkrijging van de graad van doctor aan de Technische Universiteit Delft, op gezag van de Rector Magnificus prof. ir.K.C.A.M. Luyben, voorzitter van het College voor Promoties, in het openbaar te verdedigen op dinsdag 8 november 2011 om 12:30 uur

door

Luaine BANDOUNAS

Master of Science In Molecular Cell Biology and Bioinformatics

geboren te Vanderbijlpark, South Africa

3

Dit proefschrift is goedgekeurd door de promoter: Prof. dr. J.H. de Winde

Copromotor Dr.ir. H.J. Ruijssenaars

Samenstelling promotiecommissie:

Rector Magnificus Voorzitter Prof. dr. J.H. de Winde Technische Universiteit Delft, promotor Dr. ir. H.J. Ruijssenaars BIRD Engineering, copromotor Prof. dr. I.W.C.E. Arends Technische Universiteit Delft Prof. dr. P. Verhaert Technische Universiteit Delft Prof. dr. W.R. Hagen Technische Universiteit Delft Prof. V. De Lorenzo Centro Nacional de Biotecnologia Dr. ir. J. Wery Dyadic Nederland B.V.

Dr. ir. H.J. Ruijsenaars heeft als begeleider in belangrijke mate aan de totstandkoming van het proefschrift bijgedragen.

This project is financially supported by TNO, the Netherlands Ministry of Economic Affairs and the B-Basic partner organizations ( www.b-basic.nl ) through B-Basic, a public-private NWO-ACTS programme (ACTS = Advanced Chemical Technologies for Sustainability). This project was carried out within the research programme of the Kluyver Centre for Genomics of Industrial Fermentation which is part of the Netherlands Genomics Initiative / Netherlands Organization for Scientific Research.

Cover image and layout: Luaine Bandounas ISBN: 978-84-90370-09-1 Printed by: Print Service Ede

4

In loving memory of my father, Harry Jongman

5

6

Table of Contents

Chapter 1 General Introduction 9 Chapter 2 Redundancy in putrescine catabolism in solvent tolerant Ps eudomonas putida S12. 32 Chapter 3 Isolation and characterization of novel bacterial strains exhibiting ligninolytic potential. 62 Chapter 4 Decolourization of the lignin-model dye, Azure B by a ligninolytic Bacillus sp. and initial identification of enzymes involved. 92 Chapter 5 Discussion 116

Supplemental data 132 Summary 153 Samenvatting 156 Curriculum Vitae 159 Publications 161 Acknowledgements 162

7

8

Chapter 1

General Introduction

Chapter 1

General Introduction

Global demand for renewable resources

Due to a growing world population, an average higher life expectancy and an increase in global industrialization, the world’s fossil reserves consumption has steadily been escalating over the last century. The demand for energy is expected to increase over 50 % by 2025 mainly due to worldwide development [1]. Current energy consumption places a huge strain on the environment in terms of the effect of increasing carbon dioxide emissions on the Earth’s climate. The resulting depletion of fossil resources [2, 3] will not only affect the global energy demand, but also the production of chemicals and materials. This necessitates the development of technologies to replace fossil-based resources by renewable resources.

First generation biofuels and chemicals are produced from starches, sugars and vegetable oils which gives rise to ethical and environmental issues [4]. The use of food crops as renewable resource has caused much apprehension and controversy regarding the competition for arable land and increasing food prices [3]. Lignocellulosic biomass, on the other hand, is an extremely promising renewable alternative as it is widespread and readily available to most countries, as well as not threatening the global food or feed supply. Dedicated crops for the production of lignocellulosic biomass may also be grown in combination with food crops, or on non-arable land [4]. Recently it was estimated that in the US approximately 1.3 billion tons of biomass could be sustainably produced annually from forestry and agricultural sources [5]. A conservative estimate of the global biomass production average is approximately 10 dry tons ha -1 year -1, however certain smaller scale field trials have produced significantly more than this amount [1, 6].

Emerging biorefinery industries focus on utilizing biomass as a renewable substrate for biochemicals and bioenergy production [4]. By 2030, the European Union plans to replace one quarter of the EU’s transportation fuels with biofuels and similarly, the United States Department of Energy strives to replace approximately one third of the 2004 liquid fuel 10 General Introduction demand with biofuels. In addition, the US government plans to substitute 25% of the industrially produced organic chemicals with biomass-derived alternatives [1, 5].

Fig. 1: Biotechnology vs. traditional fossil resource-based routes for energy, fuel and chemical production [7] .

Industrial Biocatalysis

Chemical industries are directing their focus towards developing sustainable processes for the biological production of (fine) chemicals [8]. The main driver for pursuing the bio- based production of chemicals is the shortage of fossil resources. Biotechnology is a potentially important approach for biobased production processes [9] and provides added benefits of wide substrate ranges and high enantio- and regio-selectivity of enzymatic biocatalysts [10, 11]. Biotechnological processes include biocatalysis, biotransformation and fermentation technology. These approaches are usually combined to convert a substrate molecule into the desired end product in a limited number of reaction steps [8, 10].

11 Chapter 1

Biocatalysis involves the use of enzymes or whole cells for the conversion of a substrate into a product of interest. The substrate may either be a (renewable) source of carbohydrates or other fermentable compounds (bioconversion), or a precursor molecule (biotransformation). The product is normally a metabolite or a dead-end product in the production strain [10]. Biocatalytic applications range from food and feed to the production of bulk or fine chemicals [11, 12]. The appropriate biocatalyst is often selected by intensive screening in which enzymes or whole cells are investigated for their abilities to catalyze a specific reaction. Whole cells are of particular interest for reactions requiring multiple reaction steps or specific co-factors. These can be regenerated by metabolically active cells, which is usually less expensive than exogenous addition of co-factors and less complicated than in-vitro regeneration [13] . High throughput screening, metabolic engineering and directed evolution techniques have made huge contributions to the development and implementation of biocatalytic systems [14].

Metabolic engineering strategies

Microorganisms are a vast source of novel and diverse enzymes and metabolic pathways, as well as valuable chemicals or metabolites. They possess a range of metabolic pathways, which serve to provide them with necessary metabolites. Usually, these metabolites are not produced in excess and their formation is strictly regulated [15]. Thus, the performance of the organism commonly has to be improved by metabolic engineering to optimize product formation, eliminate product degradation and enhance substrate import and product export [11].

Various strategies are available for the overproduction of a metabolite or product of interest by a microorganism. These include increasing the amount of precursor; adding, deleting or modifying regulatory genes or sequences; increasing the copy number of genes encoding enzymes involved in bottleneck reactions; and removing unnecessary or competing pathways or reactions [15]. Directed evolution can be utilized to introduce specific gene mutations which influence enzyme activity or functionality [16].

12 General Introduction

Much information regarding novel metabolic pathways, genes of unknown function and gene regulatory elements, can be obtained from the multitude of microbial genomes that have been completely sequenced. To investigate gene expression patterns and to identify novel metabolic pathways, comparative genomic analysis and microarray technology can be used [17]. These methods can offer information regarding the changes of the metabolic and regulatory networks in microbial cells under various conditions and their response to environmental changes [17].

Solvent tolerant Pseudomonas putida as a whole-cell biocatalyst

Pseudomonas putida is a soil bacterium capable of degrading a wide range of chemical compounds [18]. P. putida KT2440 is the best characterized strain of this species and is considered the workhorse of Pseudomonas research. Much knowledge and understanding has been acquired from its 6.18 Mbp genome sequence [18, 19]. This strain has also been used to study the effects of genome reconstruction. The investigation and alteration of metabolic pathways have lead to increased fluxes towards specific metabolites, resulting in increased product formation [18]. The metabolic diversity of P. putida has been exploited for the production of value-added compounds such as polyhydroxyalkanoates, epoxides, substituted catechols, alcohols and heterocyclic compounds [11, 18]. The catabolic versatility of P. putida KT2440 may be related to the high number of insertion elements present in this strain [18], as many of these elements are related to resistance or accessory functions which bacteria have acquired via gene rearrangements or horizontal gene transfer [20]. Many P. putida strains are able to utilize xenobiotic or toxic compounds which property has been exploited to eliminate environmental pollutants [21].

Our laboratory has genetically engineered P. putida strain S12 to produce a range of aromatic compounds such as cinnamic acid, p-coumarate, p-hydroxybenzoate and p- hydroxystyrene. This P. putida strain was specifically selected for the production of toxic hydrophobic chemicals in view of its ability to tolerate organic solvent-like compounds. Its solvent tolerance is based on several adaptation mechanisms: it can modify the composition of the inner and outer membrane, but it also disposes of a solvent efflux

13 Chapter 1 pump that extrudes uncharged lipophilic compounds from the cell [22, 23]. The solvent- tolerance properties of P. putida S12 allow this strain to be cultured in a two-phase (water-solvent) system [24]. In such a two-phase system, very toxic products like p- hydroxystyrene can be extracted in-situ , resulting in reduced product inhibition and improved overall yields [25].

Engineered P. putida S12 produces the aromatic products from model renewables like sugar and glycerol, via the amino acids phenylalanine or tyrosine [26-28]. To enable the production of such aromatic compounds, extensive metabolic engineering was required, usually via a combination of rational, targeted approaches and classical strain improvement techniques [26-30]. Although P. putida S12 has a broad substrate specificity, it is unable to grow on the pentose sugars xylose and arabinose that commonly comprise up to 25 % of the total sugars present in lignocellulosic hydrolysate [31]. Therefore, pathways for the utilization of D-xylose were introduced and, surprisingly, in one case this also resulted in the efficient utilization of L-arabinose [31]. Thus, a P. putida S12 strain was obtained via metabolic engineering and laboratory evolution, which was able to utilize glucose, xylose and arabinose, the three main sugars present in lignocellulosic hydrolysate [31, 32]. Since the carbon substrate is not only utilized for product formation but also to generate reducing equivalents [33], a (cheap) co-substrate may be added to improve the overall yield. A promising auxiliary substrate which can be obtained from biomass (via syngas) is methanol [34]. Methanol is converted by dehydrogenases or oxidases to yield reducing equivalents, thus saving the primary substrate for product formation [35, 36]. A key oxidation intermediate of methanol, however, is the toxic compound formaldehyde, which must be efficiently metabolized to prevent accumulation. P. putida S12 was genetically engineered to efficiently metabolize formaldehyde to allow the use of methanol as an auxiliary substrate, which considerably increased the biomass yield on the primary substrate glucose [33].

Thus, P. putida S12 has been engineered at multiple levels to construct a useful platform organism for the production of chemicals from renewable, biobased feedstocks. Several of the improved P. putida S12 strains have furthermore been studied at the systems level in

14 General Introduction order to identify the crucial changes underlying their improved performance. The analyses were based on the P. putida KT2440 genome sequence as a reference standard [19], in view of the high level of conservation of catabolic and biosynthetic pathways among different strains of P. putida [18, 37, 38]. A microarray-based genomotyping study furthermore demonstrated that P. putida S12 was genetically most similar to P. putida KT2440, directly after P. putida DSM3931 (which is a subculture of the mt-2 parent of strain KT2440): 86.9% of the genes with assigned functions in KT2440 were also identified in P. putida S12 [39]. Due to this high level of genetic conservation, KT2440-based microarrays could be used successfully for comparative transcriptome and proteome studies on P. putida S12 [18, 39-43].

Lignocellulosic biomass as sustainable feedstock for the production of bio-based chemicals and fuels.

The conversion of lignocellulosic biomass such as grasses, crop residues, sawdust, wood chips, pulp and paper waste has been highlighted as an important sustainable and renewable alternative means for energy, fuel and chemicals production [3, 44]. Lignocellulosic biomass is an abundant renewable resource; large amounts are available from the agricultural, forestry, food, pulp and paper industries, as well as in the form of municipal solid waste [2]. To date, biobased feedstocks are used in many applications such as plastics, solvents and lubricants. Current bio-based processes, however, usually rely on purified feedstocks e.g carbohydrates. Therefore purification and separation of components is extremely important, which presents a huge challenge especially for lignocellulosic feedstocks [1].

Recently, important progress has been made with respect to the utilization of the sugar fraction of lignocellulosic biomass for bio-based chemicals and fuels production. However, the potential for utilizing the recalcitrant lignin fraction as a renewable feedstock still needs to be explored. At present, lignin is mainly incinerated for heat and power generation, although in view of its structural richness it could also be suited as feedstock for the production of value-added chemicals such as substituted aromatics [1]. Far more

15 Chapter 1 value could be gained from lignocellulosic biomass if the lignin fraction could be converted into products of interest by efficient, controlled depolymerization reactions. At present, only a small percentage (1 - 2 %) of lignin is isolated from spent pulping liquors and utilised in specialty products, amounting to 1 million tons per year worldwide [45-47]. The availability of efficient enzymes for controlled depolymerization of lignin could greatly increase the extent to which this compound can be utilized. The vast microbial diversity should therefore be studied extensively for ligninolytic capacities, in order to obtain an enzymatic toolkit for lignin valorisation.

Table 1 : Content of cellulose, hemicellulose and lignin commonly found in agricultural waste and residues [48-51].

Lignocellulosic biomass Cellulose (%) Hemicellulose (%) Lignin (%) Newspaper 40-55 25-40 18-30 Paper 85-99 0 0-15 Corn cobs 45 35 15 Nut shells 25-30 25-30 30-40 Grasses 25-40 35-50 10-30 Hardwood stems 40-55 24-40 18-25 Softwood stems 45-50 25-35 25-35 Wheat straw 30 50 15

Lignocellulose composition

Lignocellulosic materials such as wood, consist of three main components: cellulose, hemicellulose and lignin. These constituents occur in varying degrees depending on the source. Furthermore, small amounts of pectin, ash and proteins are present [2]. Cellulose comprises approximately 45 % of the dry weight of wood and consists of D-glucose subunits linked by β-1,4 glycosidic bonds [52]. Adjacent cellulose chains interact via hydrogen bonds, Van der Waal’s forces and hydrophobic interactions to form crystalline microfibrils, which are covered by hemicellulose and lignin (Fig. 3) [2]. The second most abundant component of lignocellulose, comprising approximately 25 – 30 % of the dry wood weight, is hemicellulose [52, 53]. This carbohydrate polymer comprises various

16 General Introduction polysaccharides consisting of numerous monomers linked by β-1,4- and sometimes β-1,3- glycosidic bonds [52].

Fig. 2 : General overview of renewable lignocellulosic biomass pretreatment and conversion to fuels and chemicals.

Lignin is a recalcitrant aromatic polymer comprising 10 – 25 % of lignocellulosic biomass [56]. It consists mainly of three hydroxycinnamyl-derived alcohol monomers: p-coumaryl (p-hydroxyphenyl alcohol), coniferyl (guaiacyl propanol) and sinapyl (syringyl) alcohols

17 Chapter 1

(Fig. 4), which differ in degree of methoxylation [57]. Approximately half of the lignin structure is composed of β-Ο-4-linked ethers (arylglycerol-β-aryl), followed by phenylcoumarans, resinols and other minor subunits [58]. Since the aryl ethers are more difficult to oxidize than the 10% phenolic content of lignin [58], degradation of the β-Ο-4 substructure is considered a crucial step in lignin degradation [59]. Besides the β-aryl ether (β-O-4) linkage, which can relatively easily be cleaved chemically, more chemically resistant C-C bond linkages are present, such as β-1, β-5, β-β, 5-5 and 5-O-4 (Fig. 5) [57, 60].

Fig. 3 : A simplified representation of a plant macrofibril consisting of bundles of cellulose microfibrils encased in hemicellulose and lignin [54, 55].

Hardwood lignin mainly consists of guaiacyl and syringyl subunits, while softwood, which usually has a higher lignin content, predominantly contains guaiacyl subunits with more resistant linkages ( β-5, 5-5 and 5-O-4) due to the available C 5 position [52, 61]. Lignin serves to reinforce plant stems offering stability and strength, make the plant’s vascular tissue waterproof to allow water transport, and provide protection against pathogens [61, 62].

Fig. 4: Predominant monomers found in lignin [63].

18 General Introduction

Pretreatment of lignocellulosic biomass

The major challenge for efficient utilization of biomass for the production of biofuels or value-added chemicals is its conversion to fermentable sugars. The initial steps usually involve the destruction of various interactions between lignin and hemicellulose to make the cellulose fibrils accessible to hydrolysing chemicals or enzymes [64]. This is generally accomplished by physico-chemical pretreatment, most commonly heating under acidic conditions [56, 65]. Cellulose and hemicellulose are then depolymerized by chemical or enzymatic hydrolysis to monomeric sugars, which can be fermented to ethanol or other products [56]. Table 2 presents an overview of various lignocellulosic pretreatment methods, as well as their effect on lignocellulose.

One of the problems arising from some of the various lignocellulosic pretreatment methods is the release or formation of (toxic) inhibitors such as furfural, 5- hydroxymethylfurfural, aromatics and organic acids which may affect the overall microbial fermentation yield [65]. The production of such toxic by-products could be prevented by replacing the thermochemical pretreatment with a biological pretreatment [52], e.g ., by incubating lignocellulosic biomass with white-rot fungi. Such a pretreatment has been reported to be comparable to an alkaline pretreatment in efficiency, allowing enzymatic hydrolysis of both hemicellulose and cellulose to fermentable sugars [44], to an even higher final glucose concentration [66]. Biological pretreatment of lignocellulose requires less energy input and milder conditions, but the rate of hydrolysis is usually quite low [3].

19 Chapter 1

Fig. 5: Linkages commonly found in lignin [67].

20 General Introduction

Table 2: Examples of lignocellulose pretreatment methods

Biomass pretreatment Result Reference method

Physico-chemical pretreatment methods

Steam explosion (high temperature and rapidly Degrades hemicellulose and lignin [3, 68] reduced pressure)

Hemicellulose remains untouched, but Ammonia fibre expansion cells walls more receptive to enzyme [5] (AFEX) hydrolysis

CO 2 explosion Effective for pretreatment of cellulose [69]

Thermochemical Makes cellulose in cell walls more pretreatment (H SO at 140- [5] 2 4 accessible to saccharifying enzymes 200 °C)

(Bio-)Chemical pretreatment methods

Acid pretreatment (Sulfuric, Removes hemicellulosic components, [70] nitric or hydrochloric acid) exposing cellulose for enzymatic digestion

Removes lignin and uronic acid Alkali pretreatment [71] substitutions on hemicellulose

Oxidative delignification and reduction of Peroxide pretreatment [72] cellulose crystallinity

Ozone pretreatment Reduces lignin content [70]

Breaks internal lignin and hemicellulose Organosolv process [3, 73] bonds

Biological treatment with Delignification liberates cellulose and white-rot fungi or [52] hemicellulose from lignin ligninolytic microorganisms

Physical pretreatment methods

Mechanical treatment Reduces cellulose crystallinity [3, 74] (chipping, grinding, milling)

Pyrolysis, also Cellulose decomposes [3] thermochemical pretreatment

21 Chapter 1

Lignin biodegradation

The key challenge for the bioconversion of lignin into high-value products usually lies in the recalcitrant and complex nature of the lignin polymer. However, specialized organisms have evolved that are capable of depolymerizing lignin, such as white rot and soft rot fungi. These micoorganisms attack lignin and cellulose, while brown rot fungi specifically attack cellulose [3]. Lignin degradation by white-rot fungi involves numerous, mostly oxidative reactions. In addition, demethoxylation or demethylation reactions take place, as well as propyl side-chain cleavage between C α and C β [75]. Brown rot fungi can degrade cellulose and hemicellulose, and at advanced stages of decay, a slightly modified lignin residue is left behind [76]. To date, white rot basidiomycetes are the most efficient lignin degraders due to their extracellular oxidative enzymes, collectively called ligninases [2, 61]. Usually, these ligninolytic enzymes cannot penetrate the intact wood structure. Therefore, mediator molecules such as activated oxygen species are often required to transfer the oxidizing potential of the enzyme to the lignin substrate. These, usually small, molecules can diffuse into the structure and initiate the degradation process [77]. Generally, lignin is depolymerized by the combined efforts of extracellular laccases, lignin peroxidases (LiP’s), manganese peroxidases (MnP’s), secondary secreted metabolites and reactive oxygen species [78]. The oxidoreductase enzymes transfer electrons from a substrate to an electron acceptor, which is hydrogen peroxide in the case of peroxidases and oxygen for laccases and tyrosinases [79]. Due to the large, highly complex and diverse structure of lignin, microbial depolymerization generally occurs extracellularly, after which the smaller resulting monomers are mineralized intracellularly [5, 78].

Laccases oxidize complex substrates such as lignin, which makes them particularly interesting for diverse industrial applications [80, 81]. Laccases belong to the multi-copper oxidase family [55, 79] and have been widely studied in fungi, although various bacterial laccases have also been described. Several potential bacterial laccases have been identified in Bacillus species, such as Bacillus halodurans [82], Bacillus sphaericus and Bacillus subtilis , either or not associated with spores [79]. Similarly, laccase-like enzymes were identified in E. coli ( yacK ), Pseudomonas putida ( cumA ), Streptomyces griseus ( epoA )

22 General Introduction and Xanthomonas campestris ( copA ) [79, 83]. Laccases are found both intracellularly, as well as extracellularly as spore coat proteins. Although their functions are quite different in various organisms, they commonly serve to catalyze either polymerization or depolymerization reactions [81].

Peroxidases are peroxide-dependent oxidative enzymes, capable of oxidizing many organic and inorganic compounds. Prokaryotic peroxidases are usually located intracellularly, while fungal peroxidases are extracellular [84]. Peroxidases can react with manganese (Mn 2+ ) ions, methoxylated aromatics and phenolic compounds [85]. Lignin peroxidase (LiP) can directly oxidize lignin, 1,2,4,5-tetramethoxybenzene, phenolic compounds, anilines and certain non-phenolic structures. Although manganese peroxidases (MnP) are also strong oxidizing enzymes, they are not able to directly oxidize non-phenolic lignin-related structures [58]. MnP transfers its oxidizing potential to Mn 3+ which diffuses into the cell wall and attacks lignin [58]. The main difference between LiP’s and MnPs is that LiP’s are not very efficient at oxidizing the non-phenolic lignin compounds, probably due to their inability to penetrate the pores in lignocellulose. MnPs produce stronger oxidizing agents which do penetrate the substrate, but the overall yields are quite low [58]. A versatile peroxidase (VP) has been described in certain fungi, which is able to oxidize phenolic compounds by a combination of catalytic activities similar to Lip, MnP and plant or microbial peroxidases [61, 86]. Glyoxal oxidases, aryl-alcohol oxidases, aryl-alcohol dehydrogenases and quinone reductases have also been reported to be involved in lignin degradation [61]. Fenton-based OH radical-producing reactions could also be involved in the degradation of phenolic and non-phenolic lignin related compounds [87].

In addition to fungi, certain soil bacteria such as Nocardia and Rhodococcus, have been reported to degrade lignin [88]. Bacterial enzymes that potentially play a role in lignin degradation could be laccases, monooxygenases, multiple ring-cleaving dioxygenases and phenol oxidases [89]. The Gram-negative Sphingobium sp. SYK-6 (formerly known as Sphingomonas paucimobilis SYK-6) has the ability to degrade various lignin monoaryls and biaryls, such as vanillin, vanillate, syringaldehyde, syringate, phenylcoumarane and diarylpropane arising from lignin degradation [90, 91]. Sphingobium sp. SYK-6 is also able

23 Chapter 1 to cleave the most abundant intermolecular linkage in lignin, the β-aryl ether linkage [90, 91]. The biphenyl component of lignin, which comprises approximately 10% of the structure depending on the source of lignin, can be degraded by Pseudomonas, Sphingomonas, Burkholderia, Rhodococcus, Bacillus, Ralstonia, Acinetobacter, Comamonas and Achromobacter genera [92], as well as Pandoraea pnomenusa B-356 [93].

With regard to the potential application of ligninolytic enzymes for lignin valorization, bacterial enzymes may not need the (expensive) mediators commonly required by fungal depolymerizing enzymes in industrial applications. Furthermore, many of the fungal ligninolytic enzymes such as LiPs, MNPs and VPs contain a heme prosthetic group [94]. In an industrial environment, it may not be feasible to utilize these enzymes, as the prosthetic group may be difficult to incorporate or they may require complicated reactivation procedures [95]. Thus, the identification of enzymes involved in bacterial lignin degradation would possibly provide important alternative systems for lignin conversion to products of interest. Recent insight suggests that bacterial enzymes may be more specific, exhibit higher thermostability and may be more suited for alkaline pH conditions than those present in fungi [56, 82, 96]. Ligninolytic bacteria or their enzymes may be exploited in industrial applications for the controlled depolymerization of waste lignin, instead of it being incinerated for heat.

Bacterial metabolism of lignin-related and lignin-derived compounds

Many bacteria may not be able to degrade lignin, but some are capable of utilizing lignin- derived compounds released by other lignin-degraders, such as fungi. Bacteria employ O- demethylation systems in the utilization of lignin-derived compounds, such as vanillin demethylase, syringate O-demethylase and tetrahydrofolate-dependent aromatic O- demethylase, in conjunction with C 1-metabolism [96]. Pseudomonads are not known as very efficient lignin degraders, but some are able to efficiently degrade several low molecular weight lignin model compounds, such as anisoin, benzoin, biphenyls and chlorobiphenyls [97]. An important catabolic route for lignin-related aromatic compounds is the β-ketoadipate pathway which has been identified in various bacterial genera, such

24 General Introduction as Bacillus, Pseudomonas, Rhodococcus, Streptomyces and Alcaligenes [98]. β-Ketoadipate pathway enzymes convert lignin-derived aromatic compounds to protocatechuate or catechol. These are then converted to β-ketoadipate, followed by a two-step conversion to tricarboxylic acid intermediates [99, 100]. The decarboxylation of lignin-related aromatic compounds like ferulic acid and p-coumaric acid has been described for several microorganisms, including Bacillus species [101].

Many ligninolytic organisms are also able to utilize or degrade lignin model compounds such as lignin mimicking dyes. Synthetic industrial dyes can be applied as effective lignin model compounds to screen for the presence of ligninolytic enzymes, as certain enzymes active towards lignin are also capable of decolourizing these recalcitrant dyes [102]. Several Bacillus sp. isolates have been reported to possess ligninolytic and industrial dye degrading capabilities [89, 103-106]. It appears that certain bacteria have various enzyme systems to deal with lignin-derived compounds, although extensive research is required to identify and characterize these systems.

Fig. 6: Simplified representation of the complex lignin polymer [52].

25 Chapter 1

Scope and outline of this thesis

This research was carried out within the framework of the B-Basic (Bio-Based Sustainable Industrial Chemistry) programme and focused on the genetic engineering of microorganisms for the production of bio-based chemicals from renewable resources, as well as the identification of novel microorganisms and enzymes involved in lignin degradation for eventual application in lignin valorisation.

Chapter 1 provides a general overview of the use of microorganisms, especially the solvent-tolerant Pseudomonas putida S12 for the production of chemicals from renewable resources. The use of lignocellulose as a sustainable alternative to fossil resources is discussed, including the composition and pretreatment of lignocellulose. An overview is presented of enzyme systems and microorganisms reported to be involved in the degradation or utilization of lignin.

Chapter 2 discusses the polyamine metabolism of P. putida S12. Of particular interest is the polyamine putrescine, which has been associated with general stress conditions and may play a role in the solvent stress response of P. putida S12. The metabolic pathway of putrescine was unknown in P. putida S12; and it was therefore investigated in order to gain a better understanding of the role of polyamines in (solvent) stress.

Although Pseudomonas putida S12 has a broad substrate affinity, it cannot degrade or utilize lignin efficiently. Therefore, it was necessary to investigate other bacteria as potential sources of novel ligninolytic enzymes for waste lignin valorisation. Chapter 3 describes the isolation and characterization of three soil isolates enriched on Kraft lignin. The ligninolytic potential of Pandoraea norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003 was evaluated by investigating their ability to degrade lignin model dyes and to grow on Kraft lignin or aromatic monomers as sole carbon source.

The Bacillus sp. LD003 that demonstrated ligninolytic potential, as described in Chapter 3, also displayed a particular affinity for decolourizing the dye Azure B, which is frequently

26 General Introduction used to assess lignin-degrading activity. Bacillus sp. LD003 was therefore considered to be a promising source of novel ligninolytic enzymes. Chapter 4 describes the isolation and identification of enzyme(s) involved in Azure B decolourization in Bacillus sp. LD003.

Chapter 5 provides a summary of the findings described in Chapters 2 – 4. The importance of regulating and maintaining polyamine homeostasis in the solvent-tolerant Pseudomonas putida S12 is discussed, as well as the challenges involved in the screening of ligninolytic microorganisms and the bioconversion of lignin into products of interest.

27 Chapter 1

References

[1] Ragauskas, A.J., et al., The path forward for biofuels and biomaterials. Science, 2006. 311 (5760): p. 484-9. [2] Dashtban, M., H. Schraft, and W. Qin, Fungal bioconversion of lignocellulosic residues; opportunities & perspectives. Int J Biol Sci, 2009. 5(6): p. 578-95. [3] Sun, Y. and J. Cheng, Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresour Technol, 2002. 83 (1): p. 1-11. [4] Cherubini, F. and A.H. Stroemman, Production of Biofuels and Biochemicals from Lignocellulosic Biomass: Estimation of Maximum Theoretical Yields and Efficiencies Using Matrix Algebra. Energy Fuels, 2010. 24 : p. 2657–2666. [5] Himmel, M.E., et al., Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science, 2007. 315 (5813): p. 804-7. [6] Berndes, G., et al., The feasibility of large-scale lignocellulose-based bioenergy production Biomass and Bioenergy, 2001. 20 (5): p. 371-383. [7] van Maris, A.J., et al., Alcoholic fermentation of carbon sources in biomass hydrolysates by Saccharomyces cerevisiae: current status. Antonie Van Leeuwenhoek, 2006. 90 (4): p. 391-418. [8] Bruggink, A., A.J.J. Straathof, and L.A.M. van der Wielen, A "fine" chemical industry for Life Science Products: green solutions to chemical challenges. Advances in Biochemical Engineering/ Biotechnology, 2003. 80 : p. 69-113. [9] Hatti-Kaul, R., et al., Industrial biotechnology for the production of bio-based chemicals--a cradle-to- grave perspective. Trends Biotechnol, 2007. 25 (3): p. 119-24. [10] Held, M., Schmid, A., van Beilen, J.B. and Witholt, B., Biocatalysis. Biological systems for the production of chemicals. Pure Appl. Chem., 2000. 72 (7): p. 1337-1343. [11] Schmid, A., et al., Industrial biocatalysis today and tomorrow. Nature, 2001. 409 (6817): p. 258-68. [12] Straathof, A.J., S. Panke, and A. Schmid, The production of fine chemicals by biotransformations. Curr Opin Biotechnol, 2002. 13 (6): p. 548-56. [13] Hudlicky, T. and J.W. Reed, Applications of biotransformations and biocatalysis to complexity generation in organic synthesis. Chem Soc Rev, 2009. 38 (11): p. 3117-32. [14] Zaks, A., Industrial biocatalysis. Curr Opin Chem Biol, 2001. 5(2): p. 130-6. [15] Adrio, J.L. and A.L. Demain, Genetic improvement of processes yielding microbial products. FEMS Microbiol Rev, 2006. 30 (2): p. 187-214. [16] Joo, H., Z. Lin, and F.H. Arnold, Laboratory evolution of peroxide-mediated cytochrome P450 hydroxylation. Nature, 1999. 399 (6737): p. 670-3. [17] Torsvik, V. and L. Ovreas, Microbial diversity and function in soil: from genes to ecosystems. Curr Opin Microbiol, 2002. 5(3): p. 240-5. [18] Wu, X., et al., Comparative genomics and functional analysis of niche-specific adaptation in Pseudomonas putida. FEMS Microbiol Rev, 2010. [19] Nelson, K.E., Weinel, C., Paulsen, I. T., Dodson, R. J., Hilbert, H., Martins dos Santos, V. A., Fouts, D. E., Gill, S. R., Pop, M., Holmes, M., Brinkac, L., Beanan, M., DeBoy, R. T., Daugherty, S., Kolonay, J., Madupu, R., Nelson, W., White, O., Peterson, J., Khouri, H., Hance, I., Chris Lee, P., Holtzapple, E., Scanlan, D., Tran, K., Moazzez, A., Utterback, T., Rizzo, M., Lee, K., Kosack, D., Moestl, D., Wedler, H., Lauber, J., Stjepandic, D., Hoheisel, J., Straetz, M., Heim, S., Kiewitz, C., Eisen, J. A., Timmis, K. N., Dusterhoft, A., Tummler, B., Fraser, C. M., Complete genome sequence and comparative analysis of the metabolically versatile Pseudomonas putida KT2440. Environ Microbiol, 2002. 4(12): p. 799-08. [20] Mahillon, J. and M. Chandler, Insertion sequences. Microbiol Mol Biol Rev, 1998. 62 (3): p. 725-74. [21] Kurbatov, L., et al., Analysis of the proteome of Pseudomonas putida KT2440 grown on different sources of carbon and energy. Environ Microbiol, 2006. 8(3): p. 466-78. [22] Kieboom, J., Dennis, J. J., de Bont, J. A., Zylstra, G. J., Identification and molecular characterization of an efflux pump involved in Pseudomonas putida S12 solvent tolerance. J Biol Chem, 1998. 273 (1): p. 85-91. [23] Ramos, J.L., Duque, E., Godoy, P., Segura, A., Efflux pumps involved in toluene tolerance in Pseudomonas putida DOT-T1E. J Bacteriol, 1998. 180 (13): p. 3323-9. [24] de Bont, J.A.M., Solvent-tolerant bacteria in biocatalysis Trends in Biotechnology, 1998. 16 (12): p. 493- 499.

28 General Introduction

[25] Verhoef, S., Wierckx, N., Westerhof, R. G., de Winde, J. H., Ruijssenaars, H. J., Bioproduction of p- hydroxystyrene from glucose by the solvent-tolerant bacterium Pseudomonas putida S12 in a two- phase water-decanol fermentation. Appl Environ Microbiol, 2009. 75 (4): p. 931-6. [26] Nijkamp, K., et al., The solvent-tolerant Pseudomonas putida S12 as host for the production of cinnamic acid from glucose. Appl Microbiol Biotechnol, 2005. 69 (2): p. 170-7. [27] Verhoef, S., et al., Bioproduction of p-hydroxybenzoate from renewable feedstock by solvent-tolerant Pseudomonas putida S12. J Biotechnol, 2007. 132 (1): p. 49-56. [28] Wierckx, N.J., et al., Engineering of solvent-tolerant Pseudomonas putida S12 for bioproduction of phenol from glucose. Appl Environ Microbiol, 2005. 71 (12): p. 8221-7. [29] Nijkamp, K., et al., Optimization of the solvent-tolerant Pseudomonas putida S12 as host for the production of p-coumarate from glucose. Appl Microbiol Biotechnol, 2007. 74 (3): p. 617-24. [30] Verhoef, S., et al., Bioproduction of p-hydroxystyrene from glucose by the solvent-tolerant bacterium Pseudomonas putida S12 in a two-phase water-decanol fermentation. Appl Environ Microbiol, 2009. 75 (4): p. 931-6. [31] Meijnen, J.P., J.H. de Winde, and H.J. Ruijssenaars, Engineering Pseudomonas putida S12 for efficient utilization of D-xylose and L-arabinose. Appl Environ Microbiol, 2008. 74 (16): p. 5031-7. [32] Meijnen, J.P., J.H. de Winde, and H.J. Ruijssenaars, Construction of an oxidative D-xylose metabolism in Pseudomonas putida S12. Appl Environ Microbiol, 2009. [33] Koopman, F.W., de Winde, J. H., Ruijssenaars, H. J., C(1) compounds as auxiliary substrate for engineered Pseudomonas putida S12. Appl Microbiol Biotechnol, 2009. 83 (4): p. 705-13. [34] Chmielniak, T. and M. Sciazko, Co-gasification of biomass and coal for methanol synthesis. Appl Energy, 2003. 74 : p. 393-403. [35] Anthony, C., Bacterial oxidation of methane and methanol. Adv Microb Physiol, 1986. 27 : p. 113-210. [36] Sahm, H., Metabolism of methanol by yeast. Adv Biochem Eng Biotechnol, 1977. 6: p. 77-103. [37] Molina-Henares, M.A., et al., Identification of conditionally essential genes for growth of Pseudomonas putida KT2440 on minimal medium through the screening of a genome-wide mutant library. Environ Microbiol. 12(6): p. 1468-85. [38] Winsor, G.L., Van Rossum, T., Lo, R., Khaira, B., Whiteside, M. D., Hancock, R. E., Brinkman, F. S., Pseudomonas Genome Database: facilitating user-friendly, comprehensive comparisons of microbial genomes. Nucleic Acids Res, 2009. 37(Database issue): p. D483-8. [39] Ballerstedt, H., Volkers, R. J., Mars, A. E., Hallsworth, J. E., Santos, V. A., Puchalka, J., van Duuren, J., Eggink, G., Timmis, K. N., de Bont, J. A., Wery, J., Genomotyping of Pseudomonas putida strains using P. putida KT2440-based high-density DNA microarrays: implications for transcriptomics studies. Appl Microbiol Biotechnol, 2007. 75 : p. 1133-42. [40] Wierckx, N.J., Ballerstedt, H., de Bont, J. A., de Winde, J. H., Ruijssenaars, H. J., Wery, J., Transcriptome analysis of a phenol-producing Pseudomonas putida S12 construct: genetic and physiological basis for improved production. J Bacteriol, 2008. 190 (8): p. 2822-30. [41] Verhoef, S., Ballerstedt, H., Volkers, R. J., de Winde, J. H., Ruijssenaars, H. J., Comparative transcriptomics and proteomics of p-hydroxybenzoate producing Pseudomonas putida S12: novel responses and implications for strain improvement. Appl Microbiol Biotechnol, 2010. 87 (2): p. 679-90. [42] Volkers, R.J., de Jong, A. L., Hulst, A. G., van Baar, B. L., de Bont, J. A., Wery, J., Chemostat-based proteomic analysis of toluene-affected Pseudomonas putida S12. Environ Microbiol, 2006. 8(9): p. 1674-9. [43] Wijte, D., et al., Probing the proteome response to toluene exposure in the solvent tolerant Pseudomonas putida S12. J Proteome Res, 2011. 10 (2): p. 394-403. [44] Ding, S.Y., et al., A biophysical perspective on the cellulosome: new opportunities for biomass conversion. Curr Opin Biotechnol, 2008. 19 (3): p. 218-27. [45] Chum, H.L., et al., The Economic Contribution of Lignins to Ethanol Production from Biomass. , in SERI/TR 1985. p. 1-86. [46] Gargulak, J.G. and S.E. Lebo, Commercial Use of Lignin-Based Materials , in Lignin: Historical, Biological, and Materials Perspectives , W.G. Glasser, R.A. Northey, and T.P. Schultz, Editors. 2000, American Chemical Society. p. 304-320. [47] Lora, J.H. and W.G. Glasser, Recent Industrial Applications of Lignin: A Sustainable Alternative to Nonrenewable Materials. J Polym Environ, 2002. 10 (1/2): p. 39-48. [48] Boopathy, R., Biological treatment of swine waste using anaerobic baffled reactors. . Bioresour Technol, 1998. 64 : p. 1-6. [49] Cheung, S.W. and B.C. Anderson, Laboratory investigation of ethanol production from municipal primary waste water solids. Bioresour Technol, 1997. 59 : p. 81-96. 29 Chapter 1

[50] Dewes, T. and E. Hunsche, Composition and microbial degradability in the soil of farmyard manure from ecologically managed farms. Biol Agric Hortic, 1998. 16 : p. 251-268. [51] Reshamwala, S., B.T. Shawky, and B.E. Dale, Ethanol production from enzymatic hydrolysates of AFEX- treated coastal bermudagrass and switchgrass. Appl Biochem Biotechnol, 1995. 51/52 : p. 43-55. [52] Perez, J., et al., Biodegradation and biological treatments of cellulose, hemicellulose and lignin: an overview. Int Microbiol, 2002. 5(2): p. 53-63. [53] Rubin, E.M., Genomics of cellulosic biofuels. Nature, 2008. 454 (7206): p. 841-5. [54] Morrison, L., Second-generation biofuels may offer a way forward in Creamer Media's Engineering News Online , L. Tyrer, Editor. 2008. [55] Sharma, P., R. Goel, and N. Capalash, Bacterial Laccases. World J Microbiol Biotechnol, 2007. 23 : p. 823-832. [56] Kumar, R., S. Singh, and O.V. Singh, Bioconversion of lignocellulosic biomass: biochemical and molecular perspectives. J Ind Microbiol Biotechnol, 2008. 35 (5): p. 377-391. [57] Boerjan, W., J. Ralph, and M. Baucher, Lignin biosynthesis. Annu Rev Plant Biol, 2003. 54 : p. 519-46. [58] Hammel, K.E. and D. Cullen, Role of fungal peroxidases in biological ligninolysis. Curr Opin Plant Biol, 2008. 11 (3): p. 349-55. [59] Sato, Y., et al., Identification of three alcohol dehydrogenase genes involved in the stereospecific catabolism of arylglycerol-beta-aryl ether by Sphingobium sp. strain SYK-6. Appl Environ Microbiol, 2009. 75 (16): p. 5195-201. [60] Tejado, A., et al., Physico-chemical characterization of lignins from different sources for use in phenol- formaldehyde resin synthesis. Bioresour Technol, 2007. 98 (8): p. 1655-1658. [61] Martinez, A.T., et al., Biodegradation of lignocellulosics: microbial, chemical, and enzymatic aspects of the fungal attack of lignin. Int Microbiol, 2005. 8(3): p. 195-204. [62] Vikman, M., et al., The influence of lignin content and temperature on the biodegradation of lignocellulose in composting conditions. Appl Microbiol Biotechnol, 2002. 59 (4-5): p. 591-8. [63] Ralph, J., et al., Lignins: Natural polymers from oxidative coupling of 4-hydroxyphenylpropanoids. Phytochem Rev, 2004. 3: p. 29-60. [64] Singh, D. and S. Chen, The white-rot fungus Phanerochaete chrysosporium: conditions for the production of lignin-degrading enzymes. Appl Microbiol Biotechnol, 2008. 81 (3): p. 399-417. [65] Wierckx, N., et al., Isolation and characterization of Cupriavidus basilensis HMF14 for biological removal of inhibitors from lignocellulosic hydrolysate Microb Biotech, 2009. 3(3): p. 336-343. [66] Hatakka, A.I., Biological Pretreatment of Lignocellulose for Enzymatic Hydrolysis of Cellulose. Appl Biochem Biotechnol, 1984. 9: p. 363-364. [67] Chakar, F.S. and A.J. Ragauskas, Review of current and future softwood kraft lignin process chemistry Industrial Crops and Products 2004. 20 (2): p. 131-141. [68] Zhang, L.H., et al., Effect of steam explosion on biodegradation of lignin in wheat straw. Bioresour Technol, 2008. 99 (17): p. 8512-5. [69] Zheng, Y., H. Lin, and G.T. Tsao, Pretreatment for cellulose hydrolysis by carbon dioxide explosion. Biotechnol Prog, 1998. 14 (6): p. 890-6. [70] Silverstein, R.A., et al., A comparison of chemical pretreatment methods for improving saccharification of cotton stalks. Bioresour Technol, 2007. 98 (16): p. 3000-11. [71] Chang, V.S. and M.T. Holtzapple, Fundamental factors affecting biomass enzymatic reactivity. Appl Biochem Biotechnol, 2000. 84-86 : p. 5-37. [72] Gould, J.M., Studies on the mechanism of alkaline peroxide delignification of agricultural residues. Biotechnol Bioeng, 1985. 27 (3): p. 225-31. [73] Thring, R.W., E. Chornet, and R.P. Overend, Recovery of a Solvolytic Lignin: Effects of Spent Liquor/Acid Volume Ratio, Acid Concentration and Temperature. Biomass, 1990. 23 : p. 289-305. [74] Millett, M.A., A.J. Baker, and L.D. Satter, Physical and chemical pretreatments for enhancing cellulose saccharification. Biotech. Bioeng. Symp., 1976. 6: p. 125-153. [75] Chen, C.L., H. Chang, and T.K. Kirk, Carboxylic Acids Produced Through Oxidative Cleavage of Aromatic Rings During Degradation of Lignin in Spruce Wood by Phanerochaete Chrysosporium. J Wood Chem Technol, 1983. 3(1): p. 35-57. [76] Kleman-Leyer, K., et al., Changes in Molecular Size Distribution of Cellulose during Attack by White Rot and Brown Rot Fungi. Appl Environ Microbiol, 1992. 58 (4): p. 1266-70. [77] Regalado, V., et al., Activated oxygen species and two extracellular enzymes: laccase and aryl-alcohol oxidase, novel for the lignin-degrading fungus Fusarium proliferatum. Appl Microbiol Biotechnol, 1999. 51 : p. 388-390.

30 General Introduction

[78] Shary, S., et al., Differential expression in Phanerochaete chrysosporium of membrane-associated proteins relevant to lignin degradation. Appl Environ Microbiol, 2008. 74 (23): p. 7252-7. [79] Claus, H., Laccases and their occurrence in prokaryotes. Arch Microbiol, 2003. 179 (3): p. 145-50. [80] Dube, E., et al., Decolourization of recalcitrant dyes with a laccase from Streptomyces coelicolor under alkaline conditions. J Ind Microbiol Biotechnol, 2008. 35 (10): p. 1123-9. [81] Riva, S., Laccases: blue enzymes for green chemistry. Trends Biotechnol, 2006. 24 (5): p. 219-26. [82] Ruijssenaars, H.J. and S. Hartmans, A cloned Bacillus halodurans multicopper oxidase exhibiting alkaline laccase activity. Appl Microbiol Biotechnol, 2004. 65 (2): p. 177-82. [83] Francis, C.A. and B.M. Tebo, cumA multicopper oxidase genes from diverse Mn(II)-oxidizing and non- Mn(II)-oxidizing Pseudomonas strains. Appl Environ Microbiol, 2001. 67 (9): p. 4272-8. [84] Ferreira-Leitao, V.S., J. Godinho da Silva, and E.P.S. Bon, Methylene blue and azure B oxidation by horseradish peroxidase:a comparative evaluation of class II and class III peroxidases. Applied Catalysis B: Environmental, 2003. 42 : p. 213-221. [85] Ullrich, R. and M. Hofrichter, Enzymatic hydroxylation of aromatic compounds. Cell Mol Life Sci, 2007. 64 (3): p. 271-93. [86] Heinfling, A., et al., A study on reducing substrates of manganese-oxidizing peroxidases from Pleurotus eryngii and Bjerkandera adusta. FEBS Lett, 1998. 428 (3): p. 141-6. [87] Arantes, V. and A.M. Ferreira Milagres, The synergistic action of ligninolytic enzymes (MnP and Laccase) and Fe3+-reducing activity from white-rot fungi for degradation of Azure B Enzyme Microb Technol, 2007. 42 : p. 17-22. [88] Ahmad, M., et al., Development of novel assays for lignin degradation: comparative analysis of bacterial and fungal lignin degraders. Mol. BioSyst, 2010. 6: p. 815-821. [89] Perestelo, F., et al., Bioalteration of Kraft Pine Lignin by Bacillus rnegaterium Isolated from Compost Piles. J Fermen Bioeng, 1989. 68 (2): p. 151-153. [90] Allocati, N., et al., Glutathione transferases in bacteria. Febs J, 2009. 276 (1): p. 58-75. [91] Masai, E., et al., Roles of the enantioselective glutathione S-transferases in cleavage of beta-aryl ether. J Bacteriol, 2003. 185 (6): p. 1768-75. [92] Pieper, D.H., Aerobic degradation of polychlorinated biphenyls. Appl Microbiol Biotechnol, 2005. 67 (2): p. 170-91. [93] Gomez-Gil, L., et al., Characterization of biphenyl dioxygenase of Pandoraea pnomenusa B-356 as a potent polychlorinated biphenyl-degrading enzyme. J Bacteriol, 2007. 189 (15): p. 5705-15. [94] Lundell, T.K., M.R. Makela, and K. Hilden, Lignin-modifying enzymes in filamentous basidiomycetes-- ecological, functional and phylogenetic review. J Basic Microbiol, 2010. 50 (1): p. 5-20. [95] Wever, R., Application of peroxidases. , in Peroxidases and Catalases: Biochemistry, Biophysics, Biotechnology and Physiology H.B. Dunford, Editor. 2010, John Wiley &Sons. [96] Masai, E., Y. Katayama, and M. Fukuda, Genetic and biochemical investigations on bacterial catabolic pathways for lignin-derived aromatic compounds. Biosci Biotechnol Biochem, 2007. 71 (1): p. 1-15. [97] Zimmermann, W., Degradation of lignin by bacteria. J Biotechnol, 1990. 13 : p. 119-130. [98] Buchan, A., E.L. Neidle, and M.A. Moran, Diversity of the ring-cleaving dioxygenase gene pcaH in a salt marsh bacterial community. Appl Environ Microbiol, 2001. 67 (12): p. 5801-9. [99] Davis, J.R. and J.K. Sello, Regulation of genes in Streptomyces bacteria required for catabolism of lignin-derived aromatic compounds. Appl Microbiol Biotechnol, 2009. [100] Harwood, C.S. and R.E. Parales, The beta-ketoadipate pathway and the biology of self-identity. Annu Rev Microbiol, 1996. 50 : p. 553-90. [101] Degrassi, G., P. Polverino De Laureto, and C.V. Bruschi, Purification and characterization of ferulate and p-coumarate decarboxylase from Bacillus pumilus. Appl Environ Microbiol, 1995. 61 (1): p. 326-32. [102] Husain, Q., Potential applications of the oxidoreductive enzymes in the decolorization and detoxification of textile and other synthetic dyes from polluted water: a review. Crit Rev Biotechnol, 2006. 26 (4): p. 201-21. [103] Gudelj, M., et al., A catalase-peroxidase from a newly isolated thermoalkalphilic Bacillus sp. with potential for the treatment of textile bleaching effluents. . Extremophiles, 2002. 5: p. 423-429. [104] Itoh, K., C. Yatome, and T. Ogawa, Biodegradation of anthraquinone dyes by Bacillus subtilis. Bull Environ Contam Toxicol, 1993. 50 (4): p. 522-7. [105] Raj, A., et al., Biodegradation of kraft-lignin by Bacillus sp. isolated from sludge of pulp and paper mill. Biodegradation, 2007. 18 (6): p. 783-92. [106] Suzuki, Y., et al., Molecular cloning and characterization of the gene coding for azoreductase from Bacillus sp. OY1-2 isolated from soil. J Biol Chem, 2001. 276 (12): p. 9059-65. 31 Chapter 2

32 Putrescine catabolism in Pseudomonas putida S12

Chapter 2

Redundancy in putrescine catabolism in solvent tolerant Pseudomonas putida S12.

This chapter was published as:

Bandounas L, Ballerstedt H, de Winde JH, Ruijssenaars HJ. Redundancy in putrescine catabolism in solvent tolerant Pseudomonas putida S12. Journal of Biotechnology , 2011, 154 (1) :1-10 .

33 Chapter 2

Abstract

Pseudomonas putida S12 is a promising platform organism for the biological production of substituted aromatic compounds due to its extreme tolerance towards toxic chemicals. Solvent or aromatic stress tolerance may be due to membrane modifications and efflux pumps; however in general, polyamines have also been implicated in stressed cells. Previous transcriptomics results of P. putida strains producing an aromatic compound, or being exposed to the solvent toluene, indicated differentially expressed genes involved in polyamine transport and metabolism. Therefore, the metabolism of the polyamine, putrescine was investigated in P. putida S12, as no putrescine degradation pathways have been described for this strain. Via transcriptome analysis various, often redundant, putrescine-induced genes were identified as being potentially involved in putrescine catabolism via oxidative deamination and transamination. A series of knockout mutants were constructed in which up to six of these genes were sequentially deleted, and although putrescine degradation was affected in some of these mutants, complete elimination of putrescine degradation in P. putida S12 was not achieved. Evidence was found for the presence of an alternative pathway for putrescine degradation involving γ- glutamylation. The occurrence of multiple putrescine degradation routes in the solvent- tolerant P. putida S12 is indicative of the importance of controlling polyamine homeostasis, as well as of the high metabolic flexibility exhibited by this microorganism.

34 Putrescine catabolism in Pseudomonas putida S12

Introduction

Pseudomonas putida S12 displays exceptional tolerance towards a range of toxic organic solvents at concentrations which are usually lethal to other microorganisms [1-3]. Solvent- tolerant P. putida strains dispose of special mechanisms to cope with hydrophobic toxic compounds, which include modifications to the inner and outer membrane, as well as active extrusion of solvents through membrane-associated efflux pumps [4, 5]. Due to the robust nature and extreme tolerance towards various types of chemicals, including aromatic compounds [6-8], P. putida S12 was considered a suitable host for high-level aromatics production. For this reason, P. putida S12 was engineered for the production of substituted aromatic compounds such as phenol, t-cinnamate, p-coumarate, p- hydroxybenzoate and p-hydroxystyrene from renewable feedstock [9-14].

Polyamines like putrescine are found in many organisms [15], acting as signalling and regulatory molecules in response to stress conditions caused by reactive oxygen species (ROS), heat, UV, acid and osmotic pressure [16, 17]. They enhance the expression of genes encoding transcription factors such as Cya, FecI, Fis [18], as well as RpoS which has been implicated in stress tolerance of Pseudomonas fluorescens [19]. Polyamines also play a critical role in outer membrane function, affecting both the production and the functionality of outer membrane porins such as OmpF and OmpC in E. coli . By blocking such porins, polyamines decrease membrane permeability and protect the cell, e.g ., against acidic stress [20, 21].

The membrane damage brought about by solvent exposure is typically associated with the formation of ROS [22-24]. In view of the involvement of polyamines in oxidative stress and outer membrane functioning, these compounds may be expected to contribute to solvent tolerance, in addition to established mechanisms such as efflux pumps and membrane modification systems. Recently, novel indications for such a role of polyamines, particularly putrescine, were found in a transcriptome analysis of a p-hydroxybenzoate producing P. putida S12 strain. This study revealed a number of genes presumably involved in putrescine transport and metabolism that were differentially expressed in this 35 Chapter 2 strain compared to a non-producing control strain [25]. Also in other studies, polyamine- associated genes were found to be differentially expressed in solvent-exposed P. putida strains [3, 26].

In view of the apparent relationship between putrescine and solvent stress response, information on putrescine metabolism in P. putida S12 is of importance to gain a better understanding and control of the solvent tolerance mechanisms. Although several pathways have been described in literature [15, 16, 27, 28], it is unknown which of these are present or relevant in P. putida S12. The present study describes genes and pathways involved in putrescine degradation by P. putida S12, identified via transcriptome analysis.

Materials and Methods

Media and cultivation The plasmids and strains used are listed in Table 1a and 1b. The growth media used were Luria broth (LB) [29] or phosphate buffered mineral salts medium (MM) [30]. Unless otherwise stated, 20 mM glucose was used as the carbon source in the mineral salts media

(MMG). If putrescine was added as nitrogen source, (NH 4)2SO 4 was omitted from MMG. Cultures were grown in 20 ml liquid medium in 100-ml Erlenmeyer shake flasks at 180 rpm in a rotary shaker. Antibiotics were added to solid or liquid media as required in the following concentrations: for E. coli 10 μg/L tetracycline or 10 μg/L gentamicin; for P. putida S12 30 μg/L tetracycline (Tc) or 25 μg/L gentamicin (Gm). P. putida was cultivated at 30 °C and E. coli at 37 °C. Cultures were inoculated with cells from an overnight culture, to a starting OD 600 of 0,05 - 0,1 .

DNA techniques Genomic DNA was isolated using the FastDNA kit (Q-Biogene) in combination with a FastPrep FP120 Homogenizer (Thermo Scientific). The QIAprep spin miniprep kit (QIAGEN) was used to isolate plasmid DNA. A Gene Pulser electroporation device (BioRad) was used

36 Putrescine catabolism in Pseudomonas putida S12 to introduce plasmid DNA into electrocompetent cells. DNA concentrations were determined using a ND-1000 spectrophotometer (Nanodrop). The QIAEXII gel extraction kit (QIAGEN) was used to isolate agarose-trapped DNA fragments. Polymerase Chain Reactions (PCRs) were performed using Accuprime Pfx polymerase (Invitrogen) as specified by the manufacturer. Enzymes from Fermentas GmbH were used for DNA digestions and ligations according to the manufacturer’s instructions. Oligonucleotide synthesis and DNA sequence analysis were performed by Eurofins MWG Operon. Nucleotide sequences were analysed using the Basic Local Alignment Search Tool (BLAST) [31].

Table 1a: Plasmids used in this study.

Plasmids Characteristics Reference

pJQ200SK P15A ori sacB RP4 Gmr (pBluescriptSK); suicide vector [34]

pJN under control of nagR-pNagAa, for expression of pJNNcre(t) the Cre-recombinase used in the Cre-loxP site specific Unpublished recombination system [35]

pGEM-T Easy Ap r, used for cloning PCR fragments Promega

Used as source for the tetA - marker flanked by loxP pJQ∆hpd-lox-tetA [36] sites ( tetA-lox )

pJQ200SK containing aminotransferase (PP5182) pJQspuC-lox-tetA This study interrupted by tetA-lox

pJQ200SK containing β-alanine pyruvate pJQaptA-lox-tetA This study aminotransferase (PP0596) interrupted by tetA-lox

pJQ200SK containing oxidoreductase (PP3146) pJQox1-lox-tetA This study interrupted by tetA-lox

pJQ200SK containing 4-aminobutyrate pJQgabT-lox-tetA This study aminotransferase (PP0214) interrupted by tetA-lox

pJQ200SK containing oxidoreductase (PP4548) pJQox2-lox-tetA This study interrupted by tetA-lox

pJQ200SK containing glutamate—putrescine ligase pJQpuuA-lox-tetA This study (PP5299) interrupted by tetA-lox

37

Table 1b: Strains used in this study. Bacterial strain Alternate Properties # Reference name * P. putida S12 - Wild-type (ATCC 700801) [30] P. putida S12∆spuC KO-1 Gene encoding class III aminotransferase (PP5182) deleted This study P. putida S12∆aptA - Gene encoding β-alanine pyruvate aminotransferase (PP0596) This study deleted P. putida S12∆gabT - Gene encoding putative 4-aminobutyrate aminotransferase This study (PP0214) deleted P. putida S12∆ox1 - Gene encoding oxidoreductase (PP3146) deleted This study P. putida S12∆spuC∆gabT KO-2 Gene encoding putative 4-aminobutyrate aminotransferase This study (PP0214) deleted in S12∆spuC P. putida S12∆spuC∆aptA - Gene encoding β-alanine pyruvate aminotransferase (PP0596) This study deleted in S12∆spuC P. putida S12∆ox1 ∆ox2 - Gene encoding oxidoreductase (PP4548) deleted in S12∆ox1 This study P. putida S12∆spuC∆gabT∆aptA KO-3 Gene encoding β-alanine pyruvate aminotransferase (PP0596) This study deleted in S12∆spuC∆gabT P. putida S12∆spuC∆gabT∆aptA∆ox1 KO-4 Gene encoding oxidoreductase (PP3146) deleted in This study S12∆spuC∆gabT∆aptA P. putida KO-5 Gene encoding oxidoreductase (PP4548) deleted in This study S12∆spuC∆gabT∆aptA∆ox1 ∆ox2 S12∆spuC∆gabT∆aptA∆ox1 P. putida KO-6 Gene encoding glutamate—putrescine ligase (PP5299) deleted in This study S12∆spuC∆gabT∆aptA∆ox1 ∆ox2∆puuA S12∆spuC∆gabT∆aptA∆ox1 ∆ox2 E. coli DH5α - General cloning strain Invitrogen * Simplified strain name used throughout text. # PP locus tags taken from the P. putida KT2440 genome database [32, 33] were used to denote orthologous genes in P. putida S12.

Putrescine catabolism in Pseudomonas putida S12

Construction of knockout mutants The suicide vector pJQ200SK [34] was used as the backbone for constructing knockout plasmids for targeted gene disruption [37]. The target genes were amplified in two fragments, with a minimum size of 600 basepairs (bp) by polymerase chain reaction (PCR) using the primers listed in Table 2. The primers were designed such that 50 -100 bp were omitted from the centre of the gene. The suicide vector and the gene fragments were digested with the appropriate restriction enzymes, ligated and transformed into E. coli . The plasmids were isolated and linearized at the unique restriction site between the cloned target gene fragments. Subsequently, a tetracycline resistance gene ( tetA ) flanked by 2 loxP sites, isolated from pJQ∆hpd-lox-tetA [36], was ligated into the linearized plasmid after treatment with bacterial alkaline phosphatase (BAP, Invitrogen).

After transformation to E. coli , the knockout plasmids were isolated, verified by restriction analysis and transformed to P. putida S12 for targeted gene disruption as previously described [36, 38]. P. putida S12 disruption mutants were plated on LB + Tc agar plates. For double cross-over recombinant selection, P. putida S12 transformants were scored for tetracycline resistant (Tc R) and gentamicin sensitive (Gm S) phenotype. Successful replacement of the original gene with the tet-loxP disrupted copy was confirmed by PCR and/or sequence analysis. The tetA marker was cured from selected double cross-over knockout mutants to enable a subsequent round of targeted gene disruption. For this purpose, plasmid pJNNcre(t) (Table 1a) was transformed to selected knockout mutants which were selected on LB + Gm agar plates. These Gm R colonies were incubated in LB + Gm until growth was visible, after which 0,1 mM salicylic acid was added for Cre induction. After overnight incubation, dilutions were plated onto LB agar and the following day, loss of the Tc antibiotic marker was screened for on LB agar and LB + Tc agar plates. To cure the tetracycline sensitive (Tc S) colonies from pJNNcre(t), these were grown overnight in liquid LB medium. Loss of pJNNcre(t) was confirmed by screening for gentamicin sensitive (Gm S) phenotype and loss of the tetA marker was confirmed by PCR.

39 Chapter 2

Microarray analysis Three 1-L Erlenmeyer shake flasks containing 150 ml MM + glucose (10 mM), were inoculated with P. putida S12 from an overnight culture to a starting OD 600 of 0.06. At mid- exponential phase, 1 mM putrescine (Putrescine dihydrochloride, Sigma-Aldrich) was added to each flask. Samples (1 ml) were drawn after 0, 15 and 30 minutes of putrescine addition. Immediately after sampling, the samples were quenched and handled as previously described [36]. The cDNA samples were analysed on high-density, custom- made P. putida KT2440-genome based microarrays (Affymetrix) with additional probe sets of known sequences of P. putida S12 and related strains [39, 40]. The arrays were hybridized and scanned according to modified manufacturer’s protocols [36, 41, 42]. GeneSpring GX Software (version 7.3.1) and the GC RMA algorithm was used for data analysis [42]. After normalization, a 1-way ANOVA ( P value cut-off of 0.05) was used to select genes which changed significantly in conditions t = 15 min vs. t = 0, t = 30 min vs. t = 0 and t = 30 min vs. t = 15 min. Genes which were more than 2-fold up-regulated were considered for further analysis.

Analytical methods Putrescine was analysed by ion-exchange chromatography (DIONEX ICS-3000). A CSRS ultra II suppressor was used in combination with an IonPac CS17 column (2 x 250 mm) and an IonPac CG17 (2 x 50 mm) guard column all at 30 ºC. Methane sulphonic acid (MSA, 10 mM) was used as the eluent, at a flow rate of 0,4 ml/min. The cell density was measured at 600 nm (OD 600 ) with a µQuant MQX200 universal microplate spectrophotometer (Bio- tek), using flat-bottom 96-well microplates (Greiner). Cell dry weight (CDW) was calculated from the OD 600 value, assuming that an OD 600 of 1 is equivalent to 0,47 g/L CDW [43].

40 Putrescine catabolism in Pseudomonas putida S12

Table 2: Primers used for PCR amplification of genes targeted for disruption.

Primer Sequence (5’ → 3’) # Details MW1 5’-gcg ggatcc ttattgaatcgcctcaagggtcag-3’ Upstream spuC1 (PP5182) , BamH I MW2 5’-gcg tctaga aaggtgaaggagatcctcgcc-3’ Downstream spuC1 (PP5182) , Xba I MW3 5’-gcg tctaga tcggcgatgaaggccgcgac-3’ Upstream spuC2 (PP5182) , Xba I MW4 5’-gcg gcggccgc atgagcgtcaacaacccgcaaacc-3’ Downstream spuC2 (PP5182) , Not I MW5 5’-gcg ggatcc gatgtagtccgaccagttatag-3’ Upstream spuC1 (PP5182) , BamH I MW6 5’-gcg gcggccgc gccgccagcctactgtgtgg-3’ Downstream spuC2 (PP5182) , Not I LB09 5’-gcg ggatcc tgagaatgcccttcgg-3’ Upstream aptA 1 (PP0596) , BamH I LB10 5’-gcg tctaga ccgaagggttacctgaagcg-3’ Downstream aptA 1 (PP0596) , Xba I LB11 5’-gcg tctaga agcgcgataccgccc-3’ Upstream aptA 2 (PP0596), Xba I LB12 5’-gcg gcggccgc cgtctatctgctgg-3’ Downstream aptA 2 (PP0596), Not I LBA13 5’-gcg ggatcc gcccagaagctggccgctgc-3’ Upstream gabT 1 (PP0214) , BamH I LBA14 5’-gcg tctaga tcggcgtcgttcttgaaaatgcgc-3’ Downstream gabT 1 (PP0214) , Xba I LBA15 5’-gcg tctaga gaccagcacggcatcctgc-3’ Upstream gabT 2 (PP0214), Xba I LBA16-2 5’-gcg gcggccgc atcagcaccgccgtgacttgc-3’ Downstream gabT 2 (PP0214), Not I LB17 5’-gcg ggatcc ccgacatccacatcctggt-3’ Upstream ox1 -1 (PP3146) , BamH I LB18 5’-gcg tctaga gcagctatgtggtggcca-3’ Downstream ox1 -1 (PP3146) , Xba I LB19 5’-gcg tctaga gtgcagcgtcaacc-3’ Upstream ox1 -2 (PP3146), Xba I LB20 5’-gcg gcgccgc gaccgtatcgactcggg-3’ Downstream ox1 -2 (PP3146) , Not I LBA33 5’-gcg ggatcc ggccggcaccggcatgc-3’ Upstream ox 2-1 (PP4548) , BamH I LB34 5’-gcg tctaga tgaggtcgcaatcgatgccatag-3’ Downstream ox 2-1 (PP4548) , Xba I LB27 5’-gcg tctaga cagcgccgcggcatgggc-3’ Upstream ox 2-2 (PP4548), Xba I LB36 5’-gcg gcggccgc gagacgcggccctgggattt-3’ Downstream ox 2-2 (PP4548) , Not I MW67 5'-gcgtcagcgcagcttctcgagca-3' Downstream ox 2 gene (PP4548) LB132F 5’-gcg actagt acatgctgctgcactggtag -3’ Upstream puuA 1 (PP5299), Bcu I LB132R 5’-gcg tctaga aggtacacctgggtcgactg- 3’ Downstream puuA 1 (PP5299) , Xba I LB133F 5’-gcg tctaga gtcgcctacgaccatgaaat- 3’ Upstream puuA 2 (PP5299), Xba I LB133R 5’-gcg gcggccgc atgcggctcgatatttgaag- 3’ Downstream puuA 2 (PP5299) , Not I # Restriction enzyme sites included into the primer design indicated in bold.

41 Chapter 2

Results

Putrescine degradation by P. putida S12

Putrescine was found to be efficiently utilized by P. putida S12, either as a single source of carbon and nitrogen, as a carbon source in the presence of ammonium, or as a nitrogen source in the presence of glucose. In order to establish whether putrescine degradation in P. putida S12 was an inducible trait, cultures of P. putida S12 were cultivated with MMG (10 mM). At mid-exponential phase, 1 mM of putrescine was added to one set of cultures. After 30 min of incubation, 100 µg/ml chloramphenicol (CAM) was added to all cultures to inhibit further protein synthesis. After 20 min, 1 mM of putrescine was added both to the pre-induced and the non-induced cultures and putrescine degradation was monitored. It was observed that putrescine was only degraded in the putrescine-pre-induced culture (Fig. 1), demonstrating that the key enzymes involved in putrescine degradation were not constitutively expressed.

Transcriptome analysis of P. putida S12 cultivated in the presence of putrescine

To identify the genes involved in putrescine degradation in P. putida S12, a transcriptomics approach was followed. Since putrescine degradation was shown to be an inducible trait, the key genes involved in putrescine metabolism were expected to respond to the presence of putrescine. Thus, P. putida S12 was cultivated in MMG and 1 mM putrescine was added to the cultures at mid-exponential phase. Samples for transcriptome analysis were drawn at t = 0, 15 and 30 min after putrescine addition. The transcriptomes were analyzed using custom P. putida KT2440-based microarrays as previously described [41].

42 Putrescine catabolism in Pseudomonas putida S12

Fig. 1: Degradation of putrescine in P. putida S12 after arresting protein synthesis with chloramphenicol. Non-induced MMG grown culture ( □); putrescine pre-induced MMG- grown culture ( ■). 100 % corresponds to 1 mM putrescine for the non-induced culture, and 2 mM putrescine for the pre-induced culture (1 mM putrescine added for pre- induction). Data are the averages of triplicate experiments; the maximum deviation of the mean was less than 10 %.

A relatively small set of genes, 79 out of 5338 genes in total (1.5 %), were differentially expressed in response to putrescine at a fold change of 2 or more. For most genes, the expression difference was most prominent 15 min after the addition of putrescine, although some genes were expressed to a higher level after 30 min of putrescine exposure (Table S1, supplemental data). The differentially expressed genes were further investigated for their potential role in putrescine degradation, specifically related to the conventional transamination/deamination route of putrescine to succinate [44, 45]. Table 3 presents the differentially expressed genes that have an apparent connection to nitrogen metabolism. Table S1 (supplemental data) presents the complete overview of putrescine-responsive genes.

Transport genes

As expected, many genes putatively involved in amine (polyamine and amino acid) transport were up-regulated in response to the addition of putrescine, such as the putrescine and polyamine ABC transporters represented by PP5177 - PP5180 and PP0411 - PP0414 respectively. Other up-regulated transporters were annotated as a major facilitator family transporter (PP0702), a cation efflux family protein (PP1227), a gamma- aminobutyrate transporter (PP2543) and an amino acid transporter (PP1229). Also a

43 Chapter 2 putative major facilitator family transporter (PP5277) with a predicted sugar transporter motif was highly up-regulated in response to putrescine.

Putrescine metabolic genes

Putrescine is synthesized from arginine by arginine decarboxylase (PP0567) and from ornithine by ornithine decarboxylase (PP0864). In the presence of putrescine these genes were down-regulated as expected (Table 3). Conversely, the genes expected to be involved in putrescine degradation via the oxidative transamination / deamination pathway [27, 46] were all up-regulated. Fig. 2 presents an overview of this pathway and all putrescine-responsive genes from P. putida potentially involved.

Table 4 lists these genes as well as the similarity of the encoded enzymes to established putrescine catabolic enzymes from public databases. Two potential orthologues of the spuC gene, which encodes the first enzyme of the oxidative transamination pathway, were up-regulated in P. putida S12: PP5182 annotated as a class III aminotransferase, and PP0596 annotated as a beta-alanine—pyruvate transaminase.

Instead of an oxidative transamination, the first step in putrescine degradation may also be an oxidative deamination, catalyzed by putrescine oxidase (PuO) in Micrococcus rubens, Rhodococcus erythropolis, including Aeromonas and Pseudomonas species [47-49]. In the presence of putrescine, three genes encoding putative oxidoreductases (PP3146, PP4548, PP2448) were up-regulated. PP3146 and PP4548 showed low similarity to PuO from M. rubens (Table 4), whereas the (highly up-regulated) PP2448 showed no significant similarity to PuO at all (less than 33 % over 16 amino acids). However, PFAM prediction (http://pfam.sanger.ac.uk) classified PP3146, PP4548, PP2448 as FAD-dependent D-amino acid oxidases (DAO’s), which may indicate a role in polyamine oxidation analogous to PuO. Remarkably, the P. putida gene encoding the closest PuO homologue, PP4983 (22.5 % identity in a 484-amino acid stretch), was not differentially expressed upon putrescine exposure.

44 Putrescine catabolism in Pseudomonas putida S12

Table 3: Differentially expressed genes of P. putida S12 upon putrescine exposure.

Primary annotation from KT2440 Fold KT2440 Functionality database [32, 33] * change # locus tag UP-REGULATED GENES Polyamine Succinate-semialdehyde dehydrogenase 2,8 PP0213 metabolism (gabD ) 4-aminobutyrate aminotransferase 2,6 PP0214 (gabT) Polyamine transport Polyamine ABC transporter, ATP-binding 22,3 PP0411 protein Polyamine ABC transporter, periplasmic 18,4 PP0412 polyamine binding protein Polyamine ABC transporter, permease 31,0 PP0413 protein Polyamine ABC transporter, permease 10,6 PP0414 protein Nitrogen metabolism/ Beta-alanine—pyruvate transaminase 8 PP0596 transamination Methylmalonate-semialdehyde 6,8 PP0597 dehydrogenase (mmsA-1) Amino acid transport Major facilitator family transporter 3,5 PP0702 Hypothetical protein 9,9 PP1227 Signal transduction Methyl-accepting chemotaxis transducer 27,7 PP1228 Amino acid transport Amino acid transporter, putative 56,2 PP1229 Nitrogen metabolism Hypothetical protein 21,5 PP2448 Amino acid transport Gamma-aminobutyrate transporter, 48,5 PP2543 putative Nitrogen metabolism Pyridoxalphosphate dependent 2,4 PP2782 / deamination aminotransferase, class III L-serine dehydratase, putative 2,6 PP2930 Unknown function Conserved hypothetical protein 2,6 PP2931 TIGR00148 Nitrogen metabolism Amidase family protein 12,4 PP2932 / deamination Nitrogen metabolism Hypothetical protein 2,6 PP3145 Oxidoreductase, putative 4 PP3146 Periplasmic polyamine-binding protein, 8,4 PP3147 putative Glutamine synthetase, putative 11,6 PP3148 Hypothetical protein 71,2 PP3598 Glutamine synthetase, putative 3,3 PP4547 Oxidoreductase, putative 3,7 PP4548

Polyamine transport Hypothetical protein 10,3 PP5176 Putrescine ABC transporter, permease protein ( potI) 8,2 PP5177 Putrescine ABC transporter, permease 6,7 PP5178 protein ( potH ) Putrescine ABC transporter, ATP-binding 7 PP5179 protein ( potG ) Putrescine ABC transporter, periplasmic 8,3 PP5180 putrescine-binding protein ( potF-1) Putrescine ABC transporter, periplasmic 5,6 PP5181 putrescine-binding protein (potF-2)

45 Chapter 2

Nitrogen metabolism Putative aminotransferase 19,5 PP5182 / transamination Glutamine synthetase, putative 7,3 PP5183 Glutamine synthetase, putative 10,2 PP5184 Amino acid transport Major facilitator family transporter 38,9 PP5277 Nitrogen metabolism Aldehyde dehydrogenase family protein 18,9 PP5278 Amino acid transport Amino acid transporter, putative 21,7 PP5297 Nitrogen metabolism Hypothetical protein 23,9 PP5298 Glutamine synthetase, putative 50,2 PP5299 DOWN-REGULATED GENES Polyamine Biosynthetic arginine decarboxylase 0,2 PP0567 biosynthesis (speA ) Ornithine decarboxylase, putative ( speC ) 0,1 PP0864 *These genes are presented based on involvement in nitrogen metabolism or polyamine transport. Genes selected for inactivation are indicated in bold. # Fold change after 15 min exposure to putrescine (t=15 min), compared to the condition prior to putrescine addition (t=0).

The next step in the oxidative transamination/deamination of putrescine is catalyzed by aminobutyraldehyde dehydrogenase, encoded by ydcW in E. coli K-12. Also for this gene, an orthologue was up-regulated in P. putida S12: PP5278. Similarly, orthologues for the genes responsible for the next two steps in the putrescine degradation pathway were found among the putrescine-induced genes in P. putida S12: PP0214 encoding a 4- aminobutyrate aminotransferase (GabT-orthologue) and PP0213 encoding a succinate- semialdehyde dehydrogenase (GabD-orthologue). Thus, several orthologues for all genes involved in oxidative transamination as well as deamination of putrescine to succinate appeared to be up-regulated in P. putida S12 in response to putrescine.

46 Putrescine catabolism in Pseudomonas putida S12

Fig. 2: Proposed transamination/deamination catabolic pathway for putrescine in P. putida S12 (adapted from Chou et al. 2008). Genes that were up-regulated in P. putida S12 in

47 Chapter 2 response to putrescine are indicated in red, while down-regulated genes are indicated in green; fold change in brackets. Genes targeted for inactivation in P. putida S12 are underlined.

Inactivation of putrescine responsive genes

From the transcriptome analysis of putrescine-exposed P. putida S12, a number of genes were selected for targeted deletion in order to confirm their role in putrescine catabolism. Since putrescine degradation was shown to be an inducible trait, non-putrescine responsive genes were not considered in the selection of knockout targets. To assess the effect of the gene deletions on putrescine catabolism, deletion mutants were cultured in minimal medium to which putrescine was added as a single source of either nitrogen or carbon, as different regulatory effects were anticipated. When putrescine was added as the nitrogen source, (NH4) 2SO 4 was omitted from the minimal medium.

The putative spuC orthologue PP5182 was deleted resulting in the single knockout mutant KO-1 (see Table 1b for full strain descriptions). As shown in Table 5, putrescine degradation behaviour of strain KO-1 was very similar to the wild-type strain. No major differences were observed whether putrescine was offered as sole source of nitrogen or carbon. Therefore, the aminobutyrate aminotransferase encoding gene PP0214 ( gabT homologue) was selected as an additional target for inactivation.

GabT is primarily involved in the transamination of 4-aminobutyrate (GABA) in the lower part of the putrescine degradation pathway (Fig. 2). However, GabT has also been reported to accept putrescine as a substrate [53]. A single-knockout of this gene showed no effect on putrescine degradation (not shown). However, when combined with the PP5182-knockout of strain KO-1, yielding strain KO-2, a 2-fold reduction of the maximum specific growth rate was observed (Table 5, fig. 3B). Furthermore, the lag phase increased from 2 h to 13 h. Remarkably, this effect was only observed when putrescine was provided as the sole source of carbon; the effect was much less pronounced with putrescine as the sole nitrogen source (Table 5, Fig 3A). Despite the lower growth rate and increased lag phase, putrescine was eventually completely metabolized by strain KO-2. Unexpectedly,

48 Putrescine catabolism in Pseudomonas putida S12 no difference was observed in the ability of strain KO-2 to utilize GABA as sole carbon source (Table 5), suggesting a redundancy in GabT functionality.

In addition to PP5182, also the second up-regulated SpuC-orthologue, PP0596, was selected as a target for disruption. The single-knockout strain showed no difference in putrescine degradative capability over the wild-type strain (not shown). Deletion of PP0596 in strain KO-2 yielded strain KO-3, that exhibited a significantly lower growth rate when putrescine was provided as the sole nitrogen source (Table 5, fig. 3A). Another prominent effect was a dramatically extended lag phase (> 30 h) when putrescine, or GABA, was offered as the sole carbon source (fig. 3B, Table 5). However, in contrast to strain KO-2, the growth rate on putrescine as carbon source was not significantly different from that of the wild-type or KO-1 strain once growth had commenced. On the other hand, the growth rate on GABA was significantly lower in strain KO-3.

Fig. 3a) Growth of wild-type P. putida and

selected mutants on MMG (without NH 4) with 15 mM putrescine as sole nitrogen source. b) Growth of wild-type P. putida and selected mutants on MM with

(NH 4)2SO 4 as the primary nitrogen source and 30 mM putrescine as sole carbon source. Symbols: P. putida S12 (■), KO-2 (□), KO-3 (
), KO-4 (), KO-5 (O), KO-6 (×). Refer to table 1b for complete strain descriptions. Data are averages of triplicate experiments; the maximum deviation from the average was less than 10%.

49 Table 4: Up-regulated putrescine-responsive genes in P. putida S12 and similarity to established putrescine catabolic genes of the transamination/deamination pathway. Annotated gene in putrescine Orthologous genes in x % identity / y Organism References metabolic pathway KT2440 (locus tag) # amino acids Putrescine aminotransferase ( spuC ) P. aeruginosa PAO1 PP5182 76,9 / 450 [27, 33, 48, 50] (accession: Q9I6J2) PP0596 36,7 / 442 Putrescine oxidase ( puO ) (accession: Micrococcus rubens PP4983 22,5 / 484 [33, 48, 50-52] P40974) PP3146 32,3 / 98 PP4548 30,8 / 78 Aminobutyraldehyde dehydrogenase E. coli K-12 PP5278 35,3 / 482 [33, 48, 50] (ydcW ) (accession: P77674) 4-aminobutyrate aminotransferase E. coli K-12 PP0214 73,4 / 421 [33, 48, 50] (gabT ) (accession: P22256) Succinate-semialdehyde E. coli K-12 PP0213 82,4 / 482 [33, 48, 50] dehydrogenase ( gabD ) (accession: P25526) # Locus tags indicated in bold were targeted for inactivation.

Table 5: Growth characteristics and putrescine degradation rates of P. putida S12 and the knockout strains Growth characteristics Putrescine as sole N- Putrescine as sole C- GABA as sole C-source source source Strain lag lag lag -1 -1 -1 µmax (h ) phase µmax (h ) phase µmax (h ) phase (h) (h) (h) P. putida S12 S12 0,71 2 0,63 2 0,83 2 P. putida S12∆spuC KO-1 0,71 2 0,62 2 0,98 2 P. putida S12∆spuC∆gabT KO-2 0,56 3 0,30 13 0,83 4 P. putida S12∆spuC∆gabT∆aptA KO-3 0,31 4 0,57 38 0,37 40 P. putida S12∆spuC∆gabT∆aptA∆ox1 KO-4 0,63 3 0,56 7 0,45 10 P. putida KO-5 0,79 2 0,61 2 0,74 2 S12∆spuC∆gabT∆aptA∆ox1 ∆ox2 P. putida KO-6 0,67 3 0,58 2 0,79 2 S12∆spuC∆gabT∆aptA∆ox1 ∆ox2∆puuA Data are averages of triplicate experiments; the maximum deviation from the average was less than 10 %

Chapter 2

The transcriptomics results suggested a role for oxidative deamination in addition to oxidative transamination. Therefore, also the puO orthologues PP3146 and PP4548 were inactivated in strain KO-3, resulting in strain KO-4 and KO-5, respectively. Remarkably, when putrescine was offered as the sole nitrogen source, the growth behaviour of strain KO-4 appeared to revert to a level between that of strain KO-1 and KO-2. When putrescine was offered as the sole carbon source, a similar growth rate was observed as for strain KO- 3, but the extended lag phase was considerably shortened from > 30 h to 7 h. Surprisingly, strain KO-5 showed more or less wild-type P. putida S12 behaviour: lag phases were very similar to the wild type strain whether putrescine was provided as sole source of nitrogen or carbon (Table 5).

Putrescine metabolism via γ-glutamylation: an alternative pathway in P. putida S12?

The disruption of multiple putrescine catabolic genes of the transamination / deamination pathway, all of which were up-regulated in the presence of putrescine in wild-type P. putida S12, had an erratic effect on putrescine degradation. More importantly, putrescine degradation was not eliminated in any of the multiple gene-deletion mutants. This observation strongly suggested the presence of an alternative route for putrescine utilization, which was supported by the up-regulation of several genes in response to putrescine that could not be linked to the oxidative transamination / deamination pathway.

Among the up-regulated putrescine responsive genes, five genes were annotated as putative glutamine synthetases: PP3148, PP4547, PP5183, PP5184 and PP5299 (Table 3). The up-regulation of these genes was initially interpreted as a response to the high nitrogen load associated with the addition of putrescine, in agreement with the key role of glutamine as nitrogen-sink [54]. However, upon manual curation the encoded enzymes were found to display some similarity to the PuuA protein (Table 6). This enzyme is involved in an alternative putrescine utilization pathway via γ-glutamylation (Fig. 4), which is encoded by the puu -cluster of E. coli [15, 16]. Also in Pseudomonas aeruginosa several

52 Putrescine catabolism in Pseudomonas putida S12 orthologues of puu -genes have recently been identified, but their role in polyamine utilization has not been clarified [16].

A re-examination of the annotation of putrescine-responsive genes of P. putida S12 suggested the presence of other potential puu -cluster orthologues in addition to puuA . The putative D-amino acid oxidase encoded by PP2448 displayed a relatively high identity to PuuB (Table 6), of which the enzyme catalyzes the oxidative deamination of glutamyl- putrescine. Also PP3146 and PP4548, which were initially identified as PuO orthologues, actually exhibited somewhat higher similarity to PuuB (Table 4, 6). The third reaction in the γ-glutamylation pathway is catalyzed by PuuC ( γ-glu-γ-aminobutyraldehyde dehydrogenase). PP5278, initially annotated as a ydcW homologue and among the up- regulated putrescine responsive genes in P. putida S12, may also be considered a puuC orthologue (Table 4, 6). The final step in the γ-glutamylation pathway, from γ-glu-γ- aminobutyrate to GABA, is catalyzed by PuuD ( γ-glu-γ-aminobutyrate hydrolase). From this point onwards, the oxidative transamination / deamination and γ-glutamylation pathways converge (Fig. 2, 4). Two putative PuuD homologues were up-regulated in response to putrescine: PP3598 and PP5298 (Table 6).

The level of identity between the putative Puu-proteins of P. putida and their E. coli counterparts ranged between 36 and 65 % (Table 6). Thus, although the similarities were rather low, it could not be excluded that putrescine is degraded in P. putida S12 via the γ- glutamylation pathway, in addition to the oxidative deamination / transamination pathway. In order to shed more light onto this matter the puuA homologue with the highest fold-change (PP5299; fold change 50.2 (Table 3)) was inactivated in strain KO-5. In this strain also the putative puuB and puuE orthologues PP3146, PP4548 and PP0214 had been inactivated. The resulting strain, KO-6, exhibited a similar growth rate on putrescine as sole carbon source, but a slight decrease in growth rate was observed with putrescine as sole nitrogen-source compared to strain KO-5 (Table 5, Fig. 3). Although putrescine utilization was not eliminated and the growth rate was only slightly affected by this additional gene inactivation, this result is indicative of the existence of a γ-glutamylation

53 Chapter 2 pathway in P. putida S12. However, the importance and role of this pathway requires further investigation.

Fig. 4: Proposed γ-glutamylation catabolic pathway for putrescine in P. putida S12. Genes that were up-regulated in putrescine-exposed P. putida S12 are indicated in red; fold change in brackets. Genes targeted for inactivation in P. putida S12 are underlined.

54

Table 6: Up-regulated putrescine-responsive genes in P. putida S12 and similarity to established putrescine catabolic genes of the γ- glutamylation pathway.

Annotated gene in Orthologous genes in x % identity / y amino putrescine metabolic Organism References KT2440 (locus tag) # acids pathway Glutamate-putrescine ligase E. coli K-12 PP5299 44,4 / 448 [15, 33, 48, 50] (puuA ) (accession: P78061) PP5184 44,7 / 445 PP4547 37,3 / 450 PP5183 36 / 441 PP3148 32,8 / 451 γ-glutamylputrescine oxidase E. coli K-12 PP2448 65,4 / 408 [15, 33, 48, 50] (puuB ) (accession: P37906) PP3146 45,3 / 428 PP4548 39,2 / 411 γ-glu-γ-aminobutyraldehyde E. coli K-12 PP5278 58,7 / 491 [15, 33, 48, 50] dehydrogenase ( puuC ) (accession: P23883)

γ-glu-γ-aminobutyrate E. coli K-12 PP5298 46,6 / 249 [15, 33, 48, 50] hydrolase ( puuD ) (accession: PP3598 37,7 / 252 P76038) 4-aminobutyrate E. coli K-12 PP0214 55 / 420 [15, 33, 48, 50] aminotransferase ( puuE) (accession: P50457) # Locus tags indicated in bold were targeted for inactivation Chapter 2

Discussion

The metabolism of putrescine in Pseudomonads is a largely unexplored area, which has been studied in more detail only for P. aeruginosa [16]. We demonstrated that putrescine degradation is an inducible trait in P. putida S12, and a transcriptome analysis approach was taken to identify the genes involved. A number of the identified putrescine-responsive genes could be linked to the oxidative putrescine deamination/transamination pathway commonly associated with putrescine degradation [16, 27, 46]. Unexpectedly, the sequential deletion of several of these genes did not lead to the elimination of putrescine catabolism. This observation suggested either a large redundancy in the deleted functionalities, or the existence of an alternative pathway for putrescine degradation.

The observation that multiple homologous enzymes were present for each putative step of putrescine degradation (Fig. 2, 4) agrees with the proposed redundancy, although only a limited number of the encoding genes responded to putrescine exposure. Since the transcriptomes of knockout mutants were not investigated, it cannot be excluded that some of the initially non-responding genes were induced after other genes had been eliminated. In addition, the microarrays employed were based on P. putida KT2440. Therefore, strain-specific genes involved in putrescine degradation may have escaped detection despite the high level of similarity between strain KT2440 and P. putida S12 [41].

Manual curation of the putrescine-responsive genes furthermore provided indications for a degradative pathway via γ-glutamylation as described for E. coli . Although the degrees of homology between the putative γ-glutamylation pathway enzymes of P. putida and E. coli was rather low, several of the genes initially assigned to the oxidative transamination / deamination pathway may actually encode a putrescine degradative pathway via γ- glutamylation. In E. coli , the γ-glutamylation pathway has been shown to be more critical for putrescine degradation, than genes of the oxidative transamination / deamination route ( ygjG and ydcW) [55]. Nevertheless, the first step of the γ-glutamylation pathway requires ATP and is therefore strictly regulated. Induction of the pathway occurs only when putrescine accumulates or under starvation conditions [15]. Presumably, the γ- 56 Putrescine catabolism in Pseudomonas putida S12 glutamylation pathway is favoured when the transamination/deamination pathway is not, or insufficiently, active [16]. Possibly, the deletion of multiple genes of the transamination/deamination pathway provoked the activation of the γ-glutamylation pathway, explaining the sudden reversal to wild-type putrescine degradation behaviour.

The oxidative transamination/deamination and the γ-glutamylation pathways converge at the level of 4-aminobutyrate (GABA). Remarkably, GABA metabolism could not be eliminated by deleting gabT orthologue PP0214. Additional elimination of spuC homologue PP0596 in strain KO3 and KO4, resulted in impaired growth on GABA. PP0596 also shows homology to the probable pyridoxal-dependent aminotransferase PA5313 (50.2 % identity over 432 amino acids) of P. aeruginosa PAO1, which is responsible for GABA utilization in conjunction with gabT [16]. In P. putida S12, PP0596 may likewise have a dual role in putrescine and GABA degradation. The observation that GABA utilization was not completely eliminated in the PP0214-PP0596 knockout strains however, suggests the involvement of at least one other redundant transaminase. A possible candidate may be PP4108, which was not up-regulated in response to putrescine. PP4108 showed 44 % identity (over 397 amino acids) to PuuE from E. coli which presumably is responsible for residual GABA degradation in an E. coli gabT mutant [15].

The partially impeded putrescine degradation observed in some of the multiple-knockout strains strongly depended on the culture conditions. Putrescine degradation was most affected when it was supplied as nitrogen source in the KO-3 strain, or as carbon source in the KO-2 strain. Since putrescine can serve as a source of both carbon and nitrogen for P. putida S12, its metabolism is presumably controlled by regulatory mechanisms associated with nitrogen as well as carbon metabolism [56, 57]. No regulator genes were observed among the putrescine-responsive genes, although it is possible that regulator genes would respond only to below the set fold-change threshold ( ≥ 2). The presumed regulatory complexity of polyamine metabolism in P. putida S12, in addition to the observed genetic and metabolic redundancy, severely complicates the investigation of the role of polyamines in (solvent) stress tolerance. At the same time, it reflects the essentiality of

57 Chapter 2 maintaining polyamine homeostasis, and presents an intriguing example of the metabolic plasticity that is characteristic of stress tolerant Pseudomonas species.

Acknowledgments

This project is financially supported by the Netherlands Ministry of Economic Affairs and the B-Basic partner organizations ( www.b-basic.nl ) through B-Basic, a public-private NWO- ACTS programme (ACTS = Advanced Chemical Technologies for Sustainability). This project was carried out within the research programme of the Kluyver Centre for Genomics of Industrial Fermentation which is part of the Netherlands Genomics Initiative / Netherlands Organization for Scientific Research. We would like to acknowledge Maaike Westerhof, Lars Wilms and Ismael Ahidar for their practical assistance during the construction of the knockout mutants; and Jean-Paul Meijnen for his assistance with the ion-exchange chromatography.

Supplemental data

Table S1: page 132

References

[1] Isken, S., de Bont, J. A., Bacteria tolerant to organic solvents. Extremophiles, 1998. 2(3): p. 229-38. [2] Ramos, J.L., Duque, E., Gallegos, M. T., Godoy, P., Ramos-Gonzalez, M. I., Rojas, A., Teran, W., Segura, A., Mechanisms of solvent tolerance in gram-negative bacteria. Annu Rev Microbiol, 2002. 56 : p. 743-68. [3] Volkers, R.J., Ballerstedt, H., Ruijssenaars, H., de Bont, J. A., de Winde, J. H., Wery, J., TrgI, toluene repressed gene I, a novel gene involved in toluene-tolerance in Pseudomonas putida S12. Extremophiles, 2009. 13 (2): p. 283-97. [4] Kieboom, J., Dennis, J. J., de Bont, J. A., Zylstra, G. J., Identification and molecular characterization of an efflux pump involved in Pseudomonas putida S12 solvent tolerance. J Biol Chem, 1998. 273 (1): p. 85-91. [5] Ramos, J.L., Duque, E., Godoy, P., Segura, A., Efflux pumps involved in toluene tolerance in Pseudomonas putida DOT-T1E. J Bacteriol, 1998. 180 (13): p. 3323-9. [6] de Bont, J.A.M., Solvent-tolerant bacteria in biocatalysis Trends Biotechnol, 1998. 16 (12): p. 493-499. [7] Heipieper, H.J., Meulenbeld, G., van Oirschot, Q., de Bont, J., Effect of Environmental Factors on the trans/cis Ratio of Unsaturated Fatty Acids in Pseudomonas putida S12. Appl Environ Microbiol, 1996. 62 (8): p. 2773-2777. [8] Ramos, J.L., Gallegos, M. T., Marques, S., Ramos-Gonzalez, M. I., Espinosa-Urgel, M., Segura, A., Responses of Gram-negative bacteria to certain environmental stressors. Curr Opin Microbiol, 2001. 4(2): p. 166-71. [9] Nijkamp, K., van Luijk, N., de Bont, J. A., Wery, J., The solvent-tolerant Pseudomonas putida S12 as host for the production of cinnamic acid from glucose. Appl Microbiol Biotechnol, 2005. 69 (2): p. 170-7. 58 Putrescine catabolism in Pseudomonas putida S12

[10] Nijkamp, K., Westerhof, R. G., Ballerstedt, H., de Bont, J. A., Wery, J., Optimization of the solvent-tolerant Pseudomonas putida S12 as host for the production of p-coumarate from glucose. Appl Microbiol Biotechnol, 2007. 74 (3): p. 617-24. [11] Verhoef, S., Ruijssenaars, H. J., de Bont, J. A., Wery, J., Bioproduction of p-hydroxybenzoate from renewable feedstock by solvent-tolerant Pseudomonas putida S12. J Biotechnol, 2007. 132 (1): p. 49-56. [12] Verhoef, S., Wierckx, N., Westerhof, R. G., de Winde, J. H., Ruijssenaars, H. J., Bioproduction of p- hydroxystyrene from glucose by the solvent-tolerant bacterium Pseudomonas putida S12 in a two-phase water-decanol fermentation. Appl Environ Microbiol, 2009. 75 (4): p. 931-6. [13] Volkers, R.J., de Jong, A. L., Hulst, A. G., van Baar, B. L., de Bont, J. A., Wery, J., Chemostat-based proteomic analysis of toluene-affected Pseudomonas putida S12. Environ Microbiol, 2006. 8(9): p. 1674-9. [14] Wierckx, N.J., Ballerstedt, H., de Bont, J. A., and J. Wery, Engineering of solvent-tolerant Pseudomonas putida S12 for bioproduction of phenol from glucose. Appl Environ Microbiol, 2005. 71 (12): p. 8221-7. [15] Kurihara, S., et al., A novel putrescine utilization pathway involves gamma-glutamylated intermediates of Escherichia coli K-12. J Biol Chem, 2005. 280 (6): p. 4602-8. [16] Chou, H.T., et al., Transcriptome analysis of agmatine and putrescine catabolism in Pseudomonas aeruginosa PAO1. J Bacteriol, 2008. 190 (6): p. 1966-75. [17] Rhee, H.J., E.J. Kim, and J.K. Lee, Physiological polyamines: simple primordial stress molecules. J Cell Mol Med, 2007. 11 (4): p. 685-03. [18] Yoshida, M., et al., A unifying model for the role of polyamines in bacterial cell growth, the polyamine modulon. J Biol Chem, 2004. 279 (44): p. 46008-13. [19] Stockwell, V.O. and J.E. Loper, The sigma factor RpoS is required for stress tolerance and environmental fitness of Pseudomonas fluorescens Pf-5. Microbiology, 2005. 151 (Pt 9): p. 3001-9. [20] Shah, P., Swiatlo, E., A multifaceted role for polyamines in bacterial pathogens. Mol Microbiol, 2008. 68 (1): p. 4-16. [21] Yohannes, E., Thurber, A. E., Wilks, J. C., Tate, D. P., Slonczewski, J. L., Polyamine stress at high pH in Escherichia coli K-12. BMC Microbiol, 2005. 5: p. 59. [22] Dominguez-Cuevas, P., et al., Transcriptional tradeoff between metabolic and stress-response programs in Pseudomonas putida KT2440 cells exposed to toluene. J Biol Chem, 2006. 281 (17): p. 11981-91. [23] Sikkema, J., de Bont, J. A., Poolman, B., Mechanisms of membrane toxicity of hydrocarbons. Microbiol Rev, 1995. 59 (2): p. 201-22. [24] Weber, F.J., de Bont, J. A., Adaptation mechanisms of microorganisms to the toxic effects of organic solvents on membranes. Biochim Biophys Acta, 1996. 1286 (3): p. 225-45. [25] Verhoef, S., Ballerstedt, H., Volkers, R. J., de Winde, J. H., Ruijssenaars, H. J., Comparative transcriptomics and proteomics of p-hydroxybenzoate producing Pseudomonas putida S12: novel responses and implications for strain improvement. Appl Microbiol Biotechnol, 2010. 87 (2): p. 679-90. [26] Duque, E., Rodriguez-Herva, J. J., de la Torre, J., Dominguez-Cuevas, P., Munoz-Rojas, J., Ramos, J. L., The RpoT regulon of Pseudomonas putida DOT-T1E and its role in stress endurance against solvents. J Bacteriol, 2007. 189 (1): p. 207-19. [27] Lu, C.D., et al., Functional analysis and regulation of the divergent spuABCDEFGH-spuI operons for polyamine uptake and utilization in Pseudomonas aeruginosa PAO1. J Bacteriol, 2002. 184 (14): p. 3765-73. [28] Qian, Z.G., X.X. Xia, and S.Y. Lee, Metabolic engineering of Escherichia coli for the production of putrescine: a four carbon diamine. Biotechnol Bioeng, 2009. 104 (4): p. 651-62. [29] Sambrook, J., Russell, D.W., Molecular cloning: a laboratory manual. Third ed, ed. D.W. Sambrook J. ; Russell. Vol. 3. 2001, Cold Spring Harbor, N.Y.: Cold Spring Harbor Press. Appendix A2.2. [30] Hartmans, S., et al., Metabolism of Styrene Oxide and 2-Phenylethanol in the Styrene-Degrading Xanthobacter Strain 124X. Appl Environ Microbiol, 1989. 55 (11): p. 2850-55. [31] Altschul, S.F., et al., Basic local alignment search tool. J Mol Biol, 1990. 215 (3): p. 403-10. [32] Quandt, J. and M.F. Hynes, Versatile suicide vectors which allow direct selection for gene replacement in gram-negative bacteria. Gene, 1993. 127 (1): p. 15-21. [33] Sauer, B., Functional expression of the cre-lox site-specific recombination system in the yeast Saccharomyces cerevisiae. Mol Cell Biol, 1987. 7(6): p. 2087-96. [34] Wierckx, N.J., et al., Transcriptome analysis of a phenol-producing Pseudomonas putida S12 construct: genetic and physiological basis for improved production. J Bacteriol, 2008. 190 (8): p. 2822-30. [35] Winsor, G.L., et al., Pseudomonas Genome Database: facilitating user-friendly, comprehensive comparisons of microbial genomes. Nucleic Acids Res, 2009. 37 (Database issue): p. D483-8. [36] Nelson, K.E., et al., Complete genome sequence and comparative analysis of the metabolically versatile Pseudomonas putida KT2440. Environ Microbiol, 2002. 4(12): p. 799-08.

59 Chapter 2

[37] Meijnen, J.P., J.H. de Winde, and H.J. Ruijssenaars, Engineering Pseudomonas putida S12 for efficient utilization of D-xylose and L-arabinose. Appl Environ Microbiol, 2008. 74 (16): p. 5031-7. [38] Verhoef, S., et al., Bioproduction of p-hydroxybenzoate from renewable feedstock by solvent-tolerant Pseudomonas putida S12. J Biotechnol, 2007. 132 (1): p. 49-56. [39] Ballerstedt, H., Volkers, R. J., Mars, A. E., Hallsworth, J. E., Santos, V. A., Puchalka, J., van Duuren, J., Eggink, G., Timmis, K. N., de Bont, J. A., Wery, J., Genomotyping of Pseudomonas putida strains using P. putida KT2440-based high-density DNA microarrays: implications for transcriptomics studies. Appl Microbiol Biotechnol, 2007. 75 : p. 1133-42. [40] Wierckx, N.J., Ballerstedt, H., de Bont, J. A., de Winde, J. H., Ruijssenaars, H. J., Wery, J., Transcriptome analysis of a phenol-producing Pseudomonas putida S12 construct: genetic and physiological basis for improved production. J Bacteriol, 2008. 190 (8): p. 2822-30. [41] Ballerstedt, H., et al., Genomotyping of Pseudomonas putida strains using P. putida KT2440-based high- density DNA microarrays: implications for transcriptomics studies. Appl Microbiol Biotechnol, 2007. 75 : p. 1133-42. [42] Volkers, R.J., et al., TrgI, toluene repressed gene I, a novel gene involved in toluene-tolerance in Pseudomonas putida S12. Extremophiles, 2009. 13 (2): p. 283-97. [43] Koopman, F.W., J.H. de Winde, and H.J. Ruijssenaars, C(1) compounds as auxiliary substrate for engineered Pseudomonas putida S12. Appl Microbiol Biotechnol, 2009. 83 (4): p. 705-13. [44] Chou, H.T., Kwon, D. H., Hegazy, M., Lu, C. D., Transcriptome analysis of agmatine and putrescine catabolism in Pseudomonas aeruginosa PAO1. J Bacteriol, 2008. 190 (6): p. 1966-75. [45] Lu, C.D., Itoh, Y., Nakada, Y., Jiang, Y., Functional analysis and regulation of the divergent spuABCDEFGH- spuI operons for polyamine uptake and utilization in Pseudomonas aeruginosa PAO1. J Bacteriol, 2002. 184 (14): p. 3765-73. [46] Yang, Z. and C.D. Lu, Functional genomics enables identification of genes of the arginine transaminase pathway in Pseudomonas aeruginosa. J Bacteriol, 2007. 189 (11): p. 3945-53. [47] Cunin, R., et al., Biosynthesis and metabolism of arginine in bacteria. Microbiol Rev, 1986. 50 (3): p. 314-52. [48] Kanehisa, M., et al., KEGG for linking genomes to life and the environment. Nucleic Acids Res, 2008. 36 (Database issue): p. D480-4. [49] van Hellemond, E.W., et al., Discovery and characterization of a putrescine oxidase from Rhodococcus erythropolis NCIMB 11540. Appl Microbiol Biotechnol, 2008. 78 (3): p. 455-63. [50] Voellym, R. and T. Leisinger, Role of 4-aminobutyrate aminotransferase in the arginine metabolism of Pseudomonas aeruginosa. J Bacteriol, 1976. 128 (3): p. 722-9. [51] Gasteiger, E., et al., ExPASy: The proteomics server for in-depth protein knowledge and analysis. Nucleic Acids Res, 2003. 31 (13): p. 3784-8. [52] Shimizu, E., Tabata, Y., Hayakawa, R. and Yorifugi, T., Specific measurement of putrescine with putrescine oxidase and aminobutyraldehyde dehydrogenase. Agric. Biol. Chem., 1988. 52 (11): p. 2865-2871. [53] Swain, W.F. and R.J. Desa, Mechanism of action of putrescine oxidase. Binding characteristics of the active site of putrescine oxidase from Micrococcus rubens. Biochim Biophys Acta, 1976. 429 (2): p. 331-41. [54] Reitzer, L., Nitrogen assimilation and global regulation in Escherichia coli. Annu Rev Microbiol, 2003. 57 : p. 155-76. [55] Kurihara, S., et al., gamma-Glutamylputrescine synthetase in the putrescine utilization pathway of Escherichia coli K-12. J Biol Chem, 2008. 283 (29): p. 19981-90. [56] Commichau, F.M., K. Forchhammer, and J. Stulke, Regulatory links between carbon and nitrogen metabolism. Curr Opin Microbiol, 2006. 9(2): p. 167-72. [57] Shaibe, E., E. Metzer, and Y.S. Halpern, Control of utilization of L-arginine, L-ornithine, agmatine, and putrescine as nitrogen sources in Escherichia coli K-12. J Bacteriol, 1985. 163 (3): p. 938-42.

60 Putrescine catabolism in Pseudomonas putida S12

61 Chapter 3

62 Bacterial strains exhibiting ligninolytic potential

Chapter 3

Isolation and characterization of novel bacterial strains exhibiting ligninolytic potential.

This chapter was accepted as: Bandounas L, Wierckx N, de Winde JH, Ruijssenaars HJ. Isolation and characterization of novel bacterial strains exhibiting ligninolytic potential. BMC Biotechnology, 2011, 11 :94 .

63 Chapter 3

Abstract

Three soil bacteria were isolated by enrichment on Kraft lignin and evaluated for their ligninolytic potential as a source of novel enzymes for waste lignin valorization. Based on 16S rRNA gene sequencing and phenotypic characterization, the organisms were identified as Pandoraea norimbergensis LD001 , Pseudomonas sp LD002 and Bacillus sp LD003. The ligninolytic capability of each of these isolates was assessed by growth on high-molecular weight and low-molecular weight lignin fractions, utilization of lignin-associated aromatic monomers and degradation of ligninolytic indicator dyes. Pandoraea norimbergensis LD001 and Pseudomonas sp. LD002 exhibited best growth on lignin fractions, but limited dye-decolourizing capacity. Bacillus sp. LD003, however, showed least efficient growth on lignin fractions but extensive dye-decolourizing capacity, with a particular preference for the recalcitrant phenothiazine dye class (Azure B, Methylene Blue and Toluidene Blue O). Based on this dye-decolourizing capacity, Bacillus sp. LD003 was selected as a promising source of novel types of ligninolytic enzymes. Our observations suggested that lignin mineralization and depolymerization are separate events which place additional challenges on the screening of ligninolytic microorganisms for specific ligninolytic enzymes.

64 Bacterial strains exhibiting ligninolytic potential

Introduction

Lignin is a complex, three-dimensional aromatic polymer consisting of dimethoxylated, monomethoxylated and non-methoxylated phenylpropanoid subunits [1]. It is found in the secondary cell wall of plants, where it fills the spaces between the cellulose, hemicellulose and pectin components, making the cell wall more rigid and hydrophobic. Lignin provides plants with compressive strength and protection from pathogens [2, 3]. Presently, millions of tons of lignin and lignin-related compounds are produced as waste effluent from the pulping and paper industries [4]. These amounts are expected to further increase in the near future as a result of the recent developments aimed at replacing fossil feedstocks with lignocellulosic biomass for the production of fuels and chemicals. Generally, biorefinery processes only employ the (hemi-) cellulosic part; the lignin component remains as a low-value waste stream [5] that is commonly incinerated to generate heat and power [6-8]. To date, less than 100 000 t a -1 of lignin obtained from the Kraft pulping process is commercially exploited [9].

Much more value may be obtained if lignin could be utilized as feedstock for value-added chemicals such as substituted aromatics [7, 8]. Such valorization would require controlled depolymerization of lignin, which is hampered by its high resistance towards chemical and biological degradation [1]. Lignin can be depolymerized by thermochemical methods such as pyrolysis (thermolysis), chemical oxidation, hydrogenolysis, gasification, and hydrolysis under supercritical conditions [10]. However, many of these processes are environmentally harsh and occur under severe conditions requiring large amounts of energy [11], therefore these processes are not adequate for efficient lignin valorization .

Enzymes could provide a more specific and effective alternative for lignin depolymerization. Furthermore, biocatalytic processes generally take place under mild conditions, which lowers the energy input and reduces the environmental impact [12, 13]. A complicating factor for biocatalytic lignin degradation, however, are the structural modifications that lignin undergoes during lignocellulose processing [14, 15]. Thus, “industrial” waste lignin may differ considerably from natural lignin, and “natural”

65 Chapter 3 ligninolytic enzyme systems may not be the most effective for controlled depolymerization of industrial lignin waste.

The white rot basidiomycetes are the most extensively studied natural lignin degrading microorganisms. These fungi produce an array of powerful ligninolytic enzymes such as laccases, lignin peroxidases (LiP’s) and manganese peroxidases (MnP’s) [16, 17]. These oxidative enzyme systems commonly require low-molecular weight co-factors and mediators, such as manganese, organic acids, veratryl alcohol and substituted aromatics (e.g . 4-hydroxybenzyl alcohol, aniline, 4-hydroxybenzioc acid) [12, 18]. These mediators are the actual oxidants responsible for lignin degradation, and can penetrate deeply into the lignocellulosic matrix thanks to their limited size. Fungal lignin depolymerization usually results in a variety of low molecular weight aromatic compounds such as guaiacol, coniferyl alcohol, p-coumarate, ferulate, protocatechuate, p-hydroxybenzoate and vanillate [19, 20].

Ligninolytic bacteria are less well studied, but several examples have been found among α- proteobacteria ( e.g ., Sphingomonas sp. [21-23]), γ-proteobacteria ( e.g ., Pseudomonas sp. [24]) and actinomycetes ( Rhodococcus , Nocardia and Streptomyces sp. [25]). The enzymes reported to be involved in bacterial lignin degradation are laccases, glutathione S- transferases, ring cleaving dioxygenases [22, 26], monooxygenases and phenol oxidases [27]. Such enzymes are also involved in degradation of polycyclic aromatic hydrocarbons (PAHs), which show similar structural properties and resistance to microbial degradation as lignin [26, 28].

Thus, the bacterial ligninolytic potential is still largely unexplored and many novel ligninolytic enzymes may await discovery. These bacterial enzymes may be superior to their fungal counterparts with regard to specificity, thermostability and mediator dependency [2, 20, 29]. They may also have specific advantages for the depolymerization of the modified lignin residues typically encountered in waste streams from the pulping or 2nd generation biofuel / biobased chemicals industry. In the present study, we describe the isolation and identification of three novel ligninolytic bacterial strains. As a model

66 Bacterial strains exhibiting ligninolytic potential industrial lignin residue we employed lignin from the Kraft process, which at present is the predominant process in the pulping industry.

Materials and methods

Lignin preparation Commercially available Kraft lignin (Sigma) was used throughout this study. According to the suppliers’ specifications, the lignin was water-soluble, contained 4 % sulfur impurities, and had an average M w of 10.000 Da. Sterile stock solutions of 50 g/L in 15 mM potassium phosphate (KP i) buffer (pH 7.6) were prepared by autoclaving.

HPLC analysis of diluted lignin solution (0.5 g/L) showed that low-molecular weight (LMW) aromatic compounds were present (not shown). In order to remove the LMW compounds, a 50-ml aliquot of the lignin stock solution was dialyzed overnight against 500 ml of KP i buffer at 2 ºC using benzoylated dialysis tubing with a 2000-Da MW cut-off (Sigma- Aldrich). The dialysis buffer contained approximately 90 % of the LMW lignin fraction, and was stored at -20 ºC for further testing. Three additional buffer changes (5 L/change) were performed over a 96-h period until no further release of LMW compounds was observed (See Fig. 1). The retentate containing the HMW lignin fraction was stored at 2 ºC until further use.

Isolation and identification of lignin degrading bacteria A phosphate buffered mineral salts medium, pH 7, (MM) [30] supplemented with 5 g/L of non-dialysed Kraft lignin, 0.5 mg/L copper sulfate and 0.1 g/L yeast extract (MML), was used for enrichment of lignin-degrading bacteria. As inoculum material, soil collected from beneath decomposing wood logs in a forest near Austerlitz (The Netherlands) was used. The inoculum was prepared by suspending 5 g of soil in 100 ml of sterile 0.9 % (w/v) NaCl. After incubating for 1 h at 30 ºC with shaking at 200 rpm, 5-ml aliquots were used to inoculate four 500–ml Erlenmeyer flasks containing 100 ml of MML. The cultures were grown at 30 ºC in a shaking incubator and after 48 h, 1-ml aliquots were transferred to

67 Chapter 3 fresh MML. Over a period of 24 d, seven successive transfers were performed after which the cultures were streaked onto Luria Broth (LB) agar to obtain pure cultures.

Fig. 1 a) HPLC analysis of the LMW lignin fraction. Vanillin is indicated as a representative for the aromatic lignin monomers. Other peaks were not identified, but the absorption spectra spectra (not shown) suggested an aromatic structure; b) HPLC analysis of the high molecular weight lignin (HMW) fraction (diluted 10 times prior to HPLC measurement). Less LMW aromatic peaks were observed and vanillin was absent; c) Absorbance spectra of the dialysis buffer between 200 - 400 nm (absorption range for aromatic compounds). The absorption decreased with consecutive buffer changes, indicating that no further LMW aromatic compound were released from the HMW fraction.

Total DNA was isolated from the pure cultures using a Fast-DNA kit (Q-Biogene). Partial 16S rRNA gene sequences were amplified by polymerase chain reaction (PCR) using primers FD1/2: AGAGTTTGATCMTGGCTCAG and RP1/2: ACGGYTACCTTGTTACGACTT [31] using Pfu DNA polymerase (Fermentas). The resulting PCR products were sequenced by

68 Bacterial strains exhibiting ligninolytic potential

MWG Biotech AG and a Basic Local Alignment Search Tool (BLAST) analysis was performed on these sequences to determine the identity of the bacterial isolates [32]. The isolates were further characterized at the German Resource Centre for Biological Material (DSMZ, Braunschweig, Germany), including cellular fatty acid analysis, API, BIOLOG and classical physiological tests.

Cultivation of lignin degrading bacteria The bacteria isolated from the enrichment cultures (Pandoraea norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003) were routinely cultured on Luria broth (LB; 5 g/L NaCl, 10 g/L tryptone, 5 g/L yeast extract, pH 7). Alternatively, the strains were cultured on mineral salts medium, i.e ., MM supplemented with 40 mM of glycerol, 20 mM of glucose, or 20 mM of citrate. For monitoring growth on lignin, MM was supplemented with either the LMW or HMW lignin preparations, to a final concentration of 5 g/L. For Bacillus sp. LD003, MM was furthermore supplemented with yeast extract (0.1 g/L) and

CuSO 4 (0.5 mg/L). The strains were precultured overnight on mineral salts medium and centrifuged. The cell pellet was washed (in 0.9 % NaCl), resuspended, and used to inoculate amber Boston bottles (250 ml) containing 10 ml of lignin medium. The cultures were incubated at 30 ºC, with shaking, and samples were drawn at regular intervals to monitor growth. The lignin concentrations employed prevented accurate optical density measurements. Therefore, growth was monitored by colony-forming unit (CFU) counts. For CFU-counts, the cultures were serially diluted in 0.9 % (w/v) NaCl and plated on LB agar plates. Daily CFU determinations were made until no further CFU increase occurred. Subsequently, cultures were transferred to fresh media, followed by daily CFU determinations over a period of 5 d (120 h) to exclude that growth was due to the presence of residual carbon from the precultures.

Dye decolourization assays Decolourization of lignin-mimicking dyes was assessed both in liquid and in solid-phase assays. The following dyes were selected: Azure B (AB), Indigo Carmine (IC), Malachite Green (MG), Congo Red (CR), Xylidine ponceau (XP), Methylene Blue (MB), Toluidene Blue O (TB) and Remazol Brilliant Blue R (RBBR) (refer to Table S3 for dye structures). For liquid

69 Chapter 3 assays, the individual strains were grown in LB to an OD 600 of approximately 0.7 – 0.9 (mid- exponential growth). Dyes were added to 25 mg/L (RBBR: 50 mg/L) and cultivation was continued for another 48 h in 250-ml amber Boston bottles at 30 ºC with shaking at 200 rpm. Cultures without inoculum were included as controls for spontaneous dye decolourization. Samples were drawn at various time intervals and centrifuged for 2.5 min at 15,000 x g. The d ecolourization of a specific dye was calculated as a percentage of the initial absorbance at λmax [33]. The colour of the pellet was also visually inspected to establish whether the dye had adsorbed to the cells rather than being degraded. For s olid phase dye decolourization assays, bacteria were streaked onto dye-containing agar plates with various media. Either LB or mineral salts medium (MM) was used, supplemented with a carbon source (glycerol (40 mM) or citrate (20 mM)) and YE (0.5 g/L). Dyes were added to 50 mg/L (IC, MG, CR, XP and RBBR) or 25 mg/L (AB, MB and TB). The plates were monitored daily over a period of 120 h for growth and the development of decolourization zones.

Analytical methods For spectrophotometric analysis of the various dyes, a µQuant MQX200 universal microplate spectrophotometer (Bio-tek) was used. The absorbance spectra of the dyes between 200 nm to 800 nm were measured to establish λmax for each dye (AB, 650 nm; MB, 665 nm; TB, 635 nm; IC, 615 nm; XP, 505 nm; MG, 615 nm; CR, 470 nm; RBBR, 595 nm). Cell density was measured at 600 nm (OD 600 ) using flat-bottom 96-well microplates

(Greiner). Cell dry weight (CDW) was calculated from the OD 600 value. It was established that an OD 600 of 1 corresponded to 0.42 g/L CDW for the Bacillus sp . LB003 and P. norimbergensis LB001, respectively, 0.39 g/L CDW for Pseudomonas sp LB002. Aromatic compounds were analyzed on an Agilent 1100 HPLC system equipped with a diode array detector set at 254 nm and a 3.5 µM Zorbax SB-C18 column (4.6 x 50 mm) . 20 mM KH 2PO 4 (pH 2) was used as the eluent at a flow of 1.5 ml/min, with a 0 – 20 % acetonitrile gradient developing between 0 – 6 min, followed by 20 % acetonitrile for a further minute, after which the gradient was decreased to 0 % within the final minute.

70 Bacterial strains exhibiting ligninolytic potential

Results

Enrichment and identification of novel lignin-utilizing bacterial strains

Lignin-degrading microorganisms were enriched in liquid medium with undialysed Kraft lignin (see Materials and Methods) as the sole carbon source, using a suspension of soil from beneath rotting logs as the inoculum. After seven successive transfers to fresh lignin medium, individual colonies were obtained on LB agar plates. Seven individual colonies were selected and tentatively identified by partial 16S rRNA gene sequencing. The isolated colonies represented three different species: Pandoraea norimbergensis , Pseudomonas sp. and Bacillus sp. (Table 1). In addition, various standard biochemical tests and cellular fatty acids analysis was performed by the German Resource Centre for Biological Material (DSMZ, Braunschweig, Germany; www.dsmz.de/index.htm ). Table 2 presents general phenotypic properties of the three isolates and their ability to utilize a selection of substrates.

Isolate LD001 - Pandoraea norimbergensis

Isolate LD001 was putatively identified as Pandoraea norimbergensis by 16S rRNA gene sequencing. Similarities to other members of this genus were lower. The identity of this strain was confirmed by cellular fatty acid profiles (Table 3) which were typical for the genus Burkholderia and related genera such as Pandoraea. The general strain characteristics are stated in Table 2. Interestingly, this strain was unable to metabolize a variety of sugars such as glucose, mannose, arabinose, ribose, maltose, trehalose or cellobiose. It was, however, capable of utilizing fructose as well as a variety of organic acids (citrate, malate, gluconate, mesaconate).

Isolate LD002 - Pseudomonas sp.

The partial 16S rRNA gene sequence of isolate LD002 showed highest similarity to Pseudomonas sp. NZ099 and Pseudomonas jessenii PS06 (Table 1). The cellular fatty acid

71 Chapter 3 profile was typical for the genus Pseudomonas (see Table 3) . Due to the high similarities between different species in this group, both genetically and physiologically, it was not possible to identify this isolate LD002 to the species level. The general characteristics of isolate LD002, designated as Pseudomonas sp. LD002, are stated in Table 2.

Isolate LD003 - Bacillus sp.

The partial 16S rRNA gene sequence of isolate LD003 showed 99 % similarity to various members of the Bacillus genus, such as B. cereus and B. thuringiensis . The fatty acids profile (Table 3) of isolate LD003 was typical for that of the Bacillus cereus group, which includes B. cereus, B. thuringiensis and B. anthracis. Due to the high degree of biochemical and morphological similarity between these species, positive species differentiation is difficult [34]. However, B. anthracis could be excluded since isolate LD003 exhibited motility, hemolysis and growth on penicillin, which is uncharacteristic of B. anthracis [35]. B. thuringiensis has commonly been differentiated from B. cereus through the presence of plasmid-associated genes that encode insecticidal toxins. These are usually visible as parasporal crystals. However, as such plasmids may be lost or transferred horizontally between the two species, B. thuringiensis and B. cereus are basically indistinguishable [36]. Thus, although no parasporal crystals were observed with isolate LD003, no definitive identification as either B. thuringiensis or B. cereus was made, and the isolate was designated Bacillus sp. LD003.

72 Bacterial strains exhibiting ligninolytic potential

Table 1: Preliminary identification of isolated strains from enrichment cultures by 16S rRNA gene sequencing.

Accession number % Number of Most probable BLAST a Isolate name (16S rRNA Sequence colonies hits with 16S rRNA gene gene identity isolated sequence) Pandoraea HQ713574 Pandoraea 99 % 1 norimbergensis norimbergensis LD001 (AF139171.1) [37]

Pseudomonas sp. HQ713573 Pseudomonas sp. NZ099 99 % 4 LD002 (AF388207.1) [38]

Pseudomonas jessinii 99% PS06 (AY206685.1) [39] Bacillus sp. LD003 HQ713575 Bacillus thuringiensis 99 % 2 CMG 861 (EU697392.1) [40]

Bacillus sp. NS-4 99 % (EU622630.1) [41]

Bacillus cereus C10-1 99 % (AB244465.1) [42] a Reference: [32]

73 Table 2: General characteristics of isolated strains Strain properties Pandoraea norimbergensis LD001 Pseudomonas sp. LD002 Bacillus sp . LD003 Shape of cells Rods Rods Rods Width of cells (μm) 0,7 – 0,8 0,7 – 0,8 0,9 – 1,1 Length of cells (μm) 2,0 – 3,5 2,5 -> 5,0 2,0 – 4,5 Gram reaction Gram negative Gram negative Gram positive Motility + + + Oxidase + + + Catalase + + + ADH + + ND Urease - - ND Lysis by 3% KOH + + ND Aminopeptidase + + ND Lecithinase ND + + Levan from sucrose ND + ND Indole reaction ND ND - Phenylalaninedesaminase ND ND - Arginine dihydrolase ND ND -

Denitrification - - NO 2 from NO 3 Hemolysis ND ND + Fluorescent pigments ND + ND VP reaction ND ND + pH in VP broth ND ND 4.9 Anaerobic growth ND ND + Negative growth 42ºC 41ºC 50ºC Growth on Pseudomonas Confluent growth Light growth ND Isolation Agar (PIA) Growth in medium pH 5.7 ND ND + Growth in lysozyme broth ND ND + Growth in 2 – 7% NaCl ND ND +

Table 2 continued: Strain properties Pandoraea norimbergensis LD001 Pseudomonas sp. LD002 Bacillus sp . LD003 Growth on LB agar + antibiotic after 1 day: 10 µg/ml tetracycline + - - 30 µg/ml tetracycline - - - 10 µg/ml gentamycin + - - 25 µg/ml gentamycin + - - 100 µg/ml ampicillin + + + 5 µg/ml kanamycin + - + 50 µg/ml kanamycin + - - 5 µg/ml erythromycin + + - 10 µg/ml kanamycin + + - 50 µg/ml kanamycin + + - 10 µg/ml chloramphenicol + + - 100 µg/ml spectinomycin ND ND + 200 µg/ml spectinomycin ND ND - Hydrolysis of: gelatine + + + esculin - - + casein - ND + starch ND ND + Tween 80 ND ND - Utilization of: glucose - + + phenylacetate + - ND citrate + + + malate + + ND mannose - + ND mannitol - + - for acid from D- mannitol

Table 2 continued: Strain properties Pandoraea norimbergensis LD001 Pseudomonas sp. LD002 Bacillus sp . LD003 gluconate + + ND maltose - - ND trehalose - - ND citraconate - - ND adonitol - ND ND m-hydroxybenzoate + ND ND arabinose - + - for acid from L- arabinose mesaconate + ND ND ribose - ND ND N-acetylglucosamine - ND ND cellobiose - ND ND m-inositol ND - ND sorbitol ND - ND erythrite ND - ND hippurate ND - ND D-mandelate ND - ND 2-ketogluconate ND + ND propionate ND ND - fructose + + + for acid from D- fructose D-xylose ND ND - for acid from D-xylose +,positive; - ,negative; ND, not determined

Bacterial strains exhibiting ligninolytic potential

Table 3: Fatty acid composition of strains studied.

Pandoraea Pseudomonas sp. Bacillus sp . Fatty acids norimbergensis LD002 LD003 LD001 10:0 trace ND ND 10:0 3OH ND 4.00 ND 12:0 ISO ND ND trace 12:0 2.54 1.24 trace 13:0 ISO ND ND 8.47 13:0 ANTEISO ND ND 1.53 13:1 AT 12-13 1.48 ND ND 12:0 2OH trace 6.20 ND 12:1 3OH ND trace ND 12:0 3OH ND 4.46 ND 14:0 ISO ND ND 5.81 14:1 w5c trace ND ND 14:0 trace trace 2.37 Unknown 14.502 trace ND ND 15:0 ISO ND ND 28.23 15:0 ANTEISO ND ND 6.13 15:1 w6c trace ND ND 15:0 trace ND trace 16:1 w7c alcohol ND ND 1.2 16:0 ISO ND ND 8.4 16:1 w11c ND ND trace 16:1 w5c trace ND ND 16:0 20.64 32.66 5.54 15:0 2OH ND ND trace ISO 17:1 w10c ND ND 3.08 ISO 17:1 w5c ND ND 3.84 ANTEISO 17:1 w9c trace ND ND 17:1 ANTEISO A ND ND 1.07 17:0 ISO ND ND 7.88 17:0 ANTEISO ND ND 1.66 17:0 CYCLO 14.7 12.22 ND 17:0 trace ND ND 16:1 2OH trace ND ND 16:0 2OH trace ND ND 16:0 3OH 3.14 ND ND 18:1 w7c 26.14 11.72 ND 18:1 w5c trace ND ND 18:0 1.01 trace ND 19:0 CYCLO w8c 7.74 ND ND 18:1 2OH 1.07 ND ND Summed feature 2: Trace 12:0 ALDE, ND 3.21 4.51 16:1 ISO I/14:0 3OH Summed feature 3: 16:1 w7c/15 iso 13.06 25.96 9.14 2OH Results of fatty acids expressed as percentages. ND: Not detected. Trace: trace amount detected (less than 1 %).

77 Chapter 3

Lignin degrading capacities of the bacterial isolates

The ligninolytic potential of the isolated strains was assessed by their ability to utilize the LMW and HMW fractions of Kraft lignin. The ability to utilize aromatic lignin-monomers as sole carbon source was assessed. In addition, their capacity to decolourize ligninolytic indicator dyes was determined, which is a common method to demonstrate lignin- degrading ability in fungi [43, 44].

Growth on lignin

Mineral salts medium (MM) containing either the HMW or LMW lignin fraction and additional supplements (see Materials and Methods) were inoculated with washed cells from overnight cultures of the three bacterial isolates. The inoculum density was kept between 10,000 and 100,000 CFUs/ml and growth was monitored daily by CFU counts (see Materials and Methods). If growth was observed, 1 % (v/v) of the culture was transferred to fresh medium to verify that growth occurred on the lignin fractions and not as result of carry-over of medium components that may have escaped the washing step.

All the bacterial isolates showed an obvious increase of the number of CFUs in the lignin media (Fig. 2). Pseudomonas sp. LD002 and P. norimbergensis LD001 showed most extensive growth, whereas the CFU increase for Bacillus sp. LD003 was relatively modest.

Furthermore, Bacillus sp. LD003 required supplementation of CuSO 4 and yeast extract for growth on lignin; without these components, no growth at all was observed (not shown). For all three isolates, the observed growth was similar on the LMW and HMW lignin fractions.

78 Bacterial strains exhibiting ligninolytic potential

Fig. 2: Growth of bacterial isolates P. norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003 on a) HMW lignin fraction and b) LMW lignin fraction. Experiments were performed in 4-fold; mean values of CFUs are shown with error bars indicating the maximum deviation from the mean. * Last CFU count for Bacillus sp. LD003 was performed at 96 h instead of 120 h.

Utilization of aromatic monomers

Lignin degrading capacity does not necessarily correlate with efficient growth on lignin, as the released lignin degradation products may not be efficiently metabolized. Particularly actinomycetes have been reported to solubilize and modify lignin, despite exhibiting a limited ability to mineralize lignin [45]. Similarly, the relatively inefficient growth on lignin by Bacillus sp. LD003 may indicate a low capacity to depolymerize lignin, but alternatively a low capacity to utilize lignin degradation products. Therefore, the isolated strains were assessed for the capability to utilize monomeric aromatic compounds that are typically associated with lignin degradation.

The spectrum of lignin monomers that could be utilized for growth was relatively limited for all isolates, although P. norimbergensis LD001 appeared to have a slightly broader substrate range (Table 4). Remarkably, the alcoholic forms of the aromatic monomers (4-

79 Chapter 3 hydroxybenzyl alcohol, vanillyl alcohol, veratryl alcohol, syringol, guaiacol) were not metabolized by any of the isolates. Also the aromatic aldehydes were utilized to a limited extent: vanillin was not degraded by Pseudomonas sp. LD002 and syringaldehyde was not utilized by Bacillus sp. LD003. In contrast, each isolate consumed all the aromatic acids tested (4-hydroxybenzoic acid, syringic acid and vanillic acid) within 1-2 days. This observation suggests that the isolated strains have a fairly extensive capability for aromatics degradation. However, they appear to lack the alcohol and aldehyde dehydrogenase activities required to oxidize the aromatic alcohols and aldehydes to the carboxylic acid form.

Table 4: Growth of bacterial isolates in MM with 5 mM lignin monomer as sole carbon and energy source. Growth was considered positive if observed after a successive transfer to fresh medium. Experiments were performed in duplicate. Pandoraea Pseudomonas sp. Bacillus sp. Aromatic compound norimbergensis LD002 LD003 LD001 phenol + + + 4-hydroxybenzylalcohol - - - 4-hydroxybenzaldehyde + + + 4-hydroxybenzoic acid + + + guaiacol - - - vanillyl alcohol - - - vanillin + - + vanillic acid + + + syringol - - - syringaldehyde + + - syringic acid + + + veratryl alcohol - - - anisole - - - +, Growth; -, no growth

80 Bacterial strains exhibiting ligninolytic potential

Decolourization of ligninolytic indicator dyes

As indicated above, growth on polymeric lignin or lignin monomers is not necessarily a good measure of the ligninolytic potential of a bacterial isolate. In order to study ligninolytic potential independently from lignin utilization, the decolourization of synthetic lignin-like dyes may be monitored [46]. This approach was followed for the three isolates, employing a range of lignin-mimicking dyes (see Table 5 for dye structures). Dye decolourization was assessed in liquid assays with growing cultures (Fig. 3), as well as in solid phase plate assays (see Fig. 4 for an example). Cell pellets and colonies were inspected for dye adsorption and cell-free incubations were assayed as control for abiotic dye decolourization.

Fig. 3: Decolourization (% of initial value) of ligninolytic indicator dyes in LB medium, 25 h after dye addition to exponentially growing cultures of P. norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003. Error bars indicate the maximum deviation from the mean of duplicate experiments. * indicates that dye adsorption to the cell pellet was observed after centrifugation. Dyes: Azure B (AB), Methylene blue (MB), Toluidene Blue O (TB), Malachite Green (MG), Congo red (CR), Xylidine ponceau (XP), Indigo Carmine (IC) and Remazol Brilliant Blue R (RBBR).

P. norimbergensis LD001 appeared to decolourize a broad range of dyes in the liquid assays, among which the triarylmethane dye Malachite green (MG) and the indigoid dye Indigo Carmine (IC). However, also the abiotic controls of these dyes showed significant

81 Chapter 3 decolourization, indicating that MG and IC were not stable under the conditions tested. Still, P. norimbergensis LD001 was found to decolourize IC as well as MG to a larger extent than the abiotic controls (25 %, respectively, 24 %), suggesting that at least part of the decolourization was biogenic. Upon inspection of the cell pellets Congo Red (CR), Toluidine Blue (TB), Azure B (AB) and Methylene Blue (MB) were found to adsorb to the cells rather than being degraded. Also in plate assays, the decolourization zones for TB and MB were very small and always accompanied by dye adsorption to the colonies (not shown). In contrast to the liquid assays, no decolourization of AB or CR was observed at all in the plate assays. It was therefore concluded that the actual capacity of P. norimbergensis LD001 to degrade lignin-mimicking dyes was rather limited, and that the decolourization observed should be attributed mostly to dye adsorption.

Pseudomonas sp. LD002 decolourized AB, TB, MB and the azo dye Xylidine Ponceau (XP) by 7 -15 % in the liquid assays (Fig. 3). Of these dyes, only XP was found to adsorb to the cell pellet. Also in the plate assays AB, MB, TB, and XP were decolourized, however, these dyes were also found to adsorb to the colonies (not shown). Therefore, it is unclear to which extent Pseudomonas sp. LD002 was truly capable of degrading these dyes, or to which extent decolourization is affected by the growth conditions. IC was not found to be decolourized more than in the abiotic control, whereas MG was decolourized to 21 % over the abiotic control.

Bacillus sp. LD003 decolourized the thiazine dyes Methylene blue (MB), Azure B (AB) and Toluidene Blue O (TB) to 53, 47 and 8 % of their initial value within 25 h in the liquid assays (Fig. 3). Similarly, distinct decolourization zones were visible on the solid phase assays within 24 h (Fig. 4). Also Remazol Brilliant Blue R (RBBR) appeared to be decolourized by 15 % and Congo red (CR) by 52 %. However, in solid phase assays, RBBR and CR appeared to adsorb to the cells rather than being degraded. Like Pseudomonas sp. LD002, Bacillus sp. LD003 did not decolourize IC more than the abiotic control, but MG was decolourized to 18 % over the abiotic control. Bacillus sp. LD003 did not grow on plates containing MG (not shown), which can probably be attributed to its antimicrobial properties that are particularly effective against Gram-positive microorganisms [47]. As observed for lignin

82 Bacterial strains exhibiting ligninolytic potential utilization, addition of YE was necessary for dye decolourization by Bacillus sp. LD003; however, supplementation with CuSO 4 could be omitted.

Thus, several lignin-mimicking dyes were decolourized to varying extents by the different bacterial isolates. Especially the dye-decolourizing potential of Bacillus sp. LD003 appeared to be substantial, in contrast to its relatively poor growth on lignin (-monomers). This strain decolourized more dyes, and decolourized the dyes more completely, under various conditions, than the other isolates. In order to assess whether the dye-decolourizing activities were extracellular or rather cell-associated, the dyes were also incubated with culture supernatants of P. norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003. These cultures were performed in the presence of lignin or dye to induce the lignin / dye-degrading enzymes. However, none of the culture supernatants showed any dye- degrading capacity, suggesting that the dye-decolourizing activities were cell-associated (not shown).

Fig. 4: Decolourization zones in dye-containing plates after 72 h of incubation. a) LB with 25 mg/L Methylene Blue (MB); b) LB with 25 mg/L Azure B (AB); c) LB with 25 mg/L Toluidine Blue O (TB); d) MM + 20 mM citrate + 0,5 g/L YE with 25 mg/L Methylene Blue; e) MM + 20 mM citrate + 0,5 g/L YE with 25 mg/L Azure B; f) MM + 20 mM citrate + 0,5 g/L YE with 25 mg/L Toluidine Blue; 1 - P. norimbergensis LD001, 2 - Pseudomonas sp. LD002,3 - Bacillus sp. LD003. Experiment performed in duplicate.

83 Chapter 3

Table 5: Dyes used in this study .

84 Bacterial strains exhibiting ligninolytic potential

Discussion

Three microbial soil inhabitants identified as Pandoraea norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003 were isolated as potential lignin depolymerizing bacteria. The isolated strains showed growth on both high and low- molecular weight lignin fractions, although growth of Bacillus sp. LD003 was relatively poor. Typical lignin-associated monomers were utilized to a rather limited extent by all three isolates. Remarkably, the isolated strains appeared to lack the ability to oxidize aromatic alcohols or aldehydes to their corresponding carboxylic acid form.

The ligninolytic potential of the isolates was furthermore assessed by establishing their ability to decolourize synthetic, lignin-like dyes. The recalcitrant thiazine dye Azure B (AB) is particularly suited for this purpose. This dye is decolourized by high redox potential agents, specifically LiP’s [16, 48, 49], whereas it cannot be oxidized by nonperoxidase alcohol oxidases, MnP’s or laccases alone [48, 50]. In contrast to the other two isolates, Bacillus sp. LD003 readily decolourized AB as well as most of the other lignin-mimicking dyes tested. Also other Bacillus species as well as members of the Streptomyces genus have been reported to degrade AB within 4 - 6 days. These bacteria were isolated from wooden objects, and decolourization of AB was measured to demonstrate lignin peroxidase activity [51]. AB closely resembles methylene blue (MB) and toluidine blue O that were also readily degraded by Bacillus sp. LD003. MB has previously been found to be oxidized by lignin peroxidase [52, 53].

The seemingly contradictory finding that the highest ligninolytic potential appeared to be associated with the strain that showed poorest growth on lignin may be understood from an ecological perspective. Often, recalcitrant compounds such as lignin are degraded by microbial consortia in which the individual strains have specialized roles: some attack the complex substrate whereas others provide essential nutrients [33]. Such ligninolytic bacterial consortia can be found, e.g ., in the gut of wood-feeding termites[54]. Also bacteria like Rhodococcus erythropolis, Burkholderia sp., Citrobacter sp. and Pseudomonas sp. have been isolated from the guts of wood-feeding termites and beetles. These bacteria

85 Chapter 3 typically degrade aromatic compounds [24, 55, 56], which suggests that they feed on the aromatic compounds liberated by the lignin degrading species of the gut microflora. In a ligninolytic consortium, Bacillus sp. LD003 may fulfill the role of lignin degrader that has to rely on other microbes for specific nutrients, as suggested by its requirement for yeast extract. The other isolates in this study, Pseudomonas sp. LD002 and P. norimbergensis LD001, showed lesser ligninolytic capacities, but utilized a somewhat wider range of aromatics and did not depend on additional nutrients. Thus, such strains may fulfill the role of nutrient-provider.

The present study confirmed that ligninolytic microorganisms can be found outside the fungal kingdom. Furthermore, the bacterial isolates in this study appear to have an alternative type of ligninolytic system. The enzymes are presumably cell-surface associated, in view of the large size of lignin, whereas fungal lignin degradation occurs via extracellular enzymes and secreted secondary metabolites [57-60]. Thus, a new and presumably vast source may be tapped for novel ligninolytic enzyme activities. A few considerations, however, must be taken into account when hunting for novel ligninolytic activities for lignin valorization. First, the type of lignin to be valorized is a key factor, since the process by which it is obtained will result in structural modifications [15, 61, 62]. Thus, “natural” ligninolytic systems like those associated with white-rot fungi may not be the most efficient to valorize “industrial” lignin streams such as the Kraft lignin employed in this study. Furthermore, the most efficient lignin mineralizing strains may not be the most efficient lignin depolymerizers. Therefore, lignin degradation should be monitored as directly as possible. Ideally, the actual substrate should be used in degradation assays, but the heterogeneous nature of lignin severely complicates the analytics. Alternatively, synthetic dyes may be used to mimic lignin as we did in the present study. However, the ligninolytic activities obtained by this approach should be evaluated for their utility on the proper type of lignin. The ligninolytic enzymes of Bacillus sp. LD003, and their potential for biocatalytic Kraft lignin depolymerization, are currently under investigation.

86 Bacterial strains exhibiting ligninolytic potential

Acknowledgments

This project is financially supported by the Netherlands Ministry of Economic Affairs and the B-Basic partner organizations (www.b-basic.nl) through B-Basic, a public-private NWO- ACTS programme (ACTS = Advanced Chemical Technologies for Sustainability). This project was carried out within the research programme of the Kluyver Centre for Genomics of Industrial Fermentation which is part of the Netherlands Genomics Initiative / Netherlands Organization for Scientific Research. We would like to thank Rita Volkers for her advice and practical assistance.

References

[1] Martinez, A.T., et al., Biodegradation of lignocellulosics: microbial, chemical, and enzymatic aspects of the fungal attack of lignin. Int Microbiol, 2005. 8(3): p. 195-204. [2] Kumar, R., S. Singh, and O.V. Singh, Bioconversion of lignocellulosic biomass: biochemical and molecular perspectives. J Ind Microbiol Biotechnol, 2008. 35 (5): p. 377-391. [3] Rubin, E.M., Genomics of cellulosic biofuels. Nature, 2008. 454 (7206): p. 841-5. [4] De los Santos Ramos, W., et al., Remediation of lignin and its derivatives from pulp and paper industry wastewater by the combination of chemical precipitation and ozonation. J Hazard Mater, 2009. 169 (1-3): p. 428-34. [5] Stewart, D., Lignin as a base material for materials applications: Chemistry, application and economics. Ind. Crops Prod, 2008. 27 : p. 202-207. [6] Himmel, M.E., et al., Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science, 2007. 315 (5813): p. 804-7. [7] Ragauskas, A.J., et al., The path forward for biofuels and biomaterials. Science, 2006. 311 (5760): p. 484-9. [8] Zaldivar, J., J. Nielsen, and L. Olsson, Fuel ethanol production from lignocellulose: a challenge for metabolic engineering and process integration. Appl Microbiol Biotechnol, 2001. 56 (1-2): p. 17-34. [9] Gosselink, R.J.A., et al., Co-ordination network for lignin—standardisation, production and applications adapted to market requirements (EUROLIGNIN) Ind. Crops Prod, 2004. 20 : p. 121- 129. [10] Pandey, M.P. and C.S. Kim, Lignin Depolymerization and Conversion: A Review of Thermochemical Methods. Chem Engin Technol, 2011. 34 (1): p. 29-41. [11] Ward, P.O. and A. Singh, Bioethanol technology: developments and perspectives. Advan Appl Microbiol, 2002. 51 : p. 53-80. [12] Perez, J., et al., Biodegradation and biological treatments of cellulose, hemicellulose and lignin: an overview. Int Microbiol, 2002. 5(2): p. 53-63. [13] Sun, Y. and J. Cheng, Hydrolysis of lignocellulosic materials for ethanol production: a review. Bioresour Technol, 2002. 83 (1): p. 1-11.

87 Chapter 3

[14] Emmel, A., et al., Fractionation of Eucalyptus grandis chips by dilute acid-catalysed steam explosion. Bioresour Technol, 2003. 86 (2): p. 105-15. [15] Crawford, R.L., Lignin Biodegradation and Transformation. , ed. J.W.a. Sons. 1981: Wiley- Interscience. 154. [16] Arantes, V. and A.M. Ferreira Milagres, The synergistic action of ligninolytic enzymes (MnP and Laccase) and Fe3+-reducing activity from white-rot fungi for degradation of Azure B Enzyme Microb Technol, 2007. 42 : p. 17-22. [17] Shary, S., et al., Differential expression in Phanerochaete chrysosporium of membrane- associated proteins relevant to lignin degradation. Appl Environ Microbiol, 2008. 74 (23): p. 7252-7. [18] Kunamneni, A., et al., Engineering and Applications of fungal laccases for organic synthesis. Microb Cell Fact, 2008. 7: p. 32. [19] Harwood, C.S. and R.E. Parales, The beta-ketoadipate pathway and the biology of self-identity. Annu Rev Microbiol, 1996. 50 : p. 553-90. [20] Masai, E., Y. Katayama, and M. Fukuda, Genetic and biochemical investigations on bacterial catabolic pathways for lignin-derived aromatic compounds. Biosci Biotechnol Biochem, 2007. 71 (1): p. 1-15. [21] Wenzel, M., et al., Aerobic and facultatively anaerobic cellulolytic bacteria from the gut of the termite Zootermopsis angusticollis. J Appl Microbiol, 2002. 92 (1): p. 32-40. [22] Masai, E., et al., Roles of the enantioselective glutathione S-transferases in cleavage of beta- aryl ether. J Bacteriol, 2003. 185 (6): p. 1768-75. [23] Masai, E., et al., Characterization of Sphingomonas paucimobilis SYK-6 genes involved in degradation of lignin-related compounds. J Ind Microbiol Biotechnol, 1999. 23 (4-5): p. 364- 373. [24] Delalibera, I., et al., Composition of the bacterial community in the gut of the pine engraver, Ips pini (Say) (Coleoptera) colonizing red pine. Symbiosis, 2007. 43 : p. 97-104. [25] Zimmermann, W., Degradation of lignin by bacteria. J Biotechnol, 1990. 13 : p. 119-130. [26] Allocati, N., et al., Glutathione transferases in bacteria. Febs J, 2009. 276 (1): p. 58-75. [27] Perestelo, F., et al., Bioalteration of Kraft Pine Lignin by Bacillus rnegaterium Isolated from Compost Piles. J Fermen Bioeng, 1989. 68 (2): p. 151-153. [28] Canas, A.I., et al., Transformation of polycyclic aromatic hydrocarbons by laccase is strongly enhanced by phenolic compounds present in soil. Environ Sci Technol, 2007. 41 (8): p. 2964-71. [29] Ruijssenaars, H.J. and S. Hartmans, A cloned Bacillus halodurans multicopper oxidase exhibiting alkaline laccase activity. Appl Microbiol Biotechnol, 2004. 65 (2): p. 177-82. [30] Hartmans, S., Smits, J. P., van der Werf, M. J., Volkering, F., de Bont, J. A., Metabolism of Styrene Oxide and 2-Phenylethanol in the Styrene-Degrading Xanthobacter Strain 124X. Appl Environ Microbiol, 1989. 55 (11): p. 2850-2855. [31] Weisberg, W.G., et al., 16S Ribosomal DNA Amplification for Phylogenetic Study. Journal of Bacteriology, 1991. 173 (2): p. 697-703. [32] Altschul, S.F., Gish, W., Miller, W., Myers, E. W., Lipman, D. J., Basic local alignment search tool. J Mol Biol, 1990. 215 (3): p. 403-10. [33] Saratale, R.G., et al., Enhanced decolorization and biodegradation of textile azo dye Scarlet R by using developed microbial consortium-GR. Bioresour Technol, 2009. 100 (9): p. 2493-500. [34] Zhong, W., et al., Differentiation of Bacillus anthracis, B. cereus, and B. thuringiensis by using pulsed-field gel electrophoresis. Appl Environ Microbiol, 2007. 73 (10): p. 3446-9. [35] Henderson, I., C.J. Duggleby, and P.C. Turnbull, Differentiation of Bacillus anthracis from other Bacillus cereus group bacteria with the PCR. Int J Syst Bacteriol, 1994. 44 (1): p. 99-105. [36] Helgason, E., et al., Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis--one species on the basis of genetic evidence. Appl Environ Microbiol, 2000. 66 (6): p. 2627-30.

88 Bacterial strains exhibiting ligninolytic potential

[37] Coenye, T., et al., Description of Pandoraea gen. nov. with Pandoraea apista sp. nov., Pandoraea pulmonicola sp. nov., Pandoraea pnomenusa sp. nov., Pandoraea sputorum sp. nov. and Pandoraea norimbergensis comb. nov. Int J Syst Evol Microbiol, 2000. 50 Pt 2 : p. 887-99. [38] Godfrey, S.A., et al., Characterization by 16S rRNA sequence analysis of pseudomonads causing blotch disease of cultivated Agaricus bisporus. Appl Environ Microbiol, 2001. 67 (9): p. 4316-23. [39] Valverde, A., et al., Differential effects of coinoculations with Pseudomonas jessenii PS06 (a phosphate-solubilizing bacterium) and Mesorhizobium ciceri C-2/2 strains on the growth and seed yield of chickpea under greenhouse and field conditions Plant and soil 2006. 287 : p. 43-50. [40] Banerjee, A. and A.K. Ghoshal, Isolation and characterization of hyper phenol tolerant Bacillus sp. from oil refinery and exploration sites. J Hazard Mater, 2010. 176 (1-3): p. 85-91. [41] Prakash, N.T., et al., Removal of selenium from Se enriched natural soils by a consortium of Bacillus isolates. Bull Environ Contam Toxicol, 2010. 85 (2): p. 214-8. [42] Nishiwaki, H., et al., Insecticidal bacteria isolated from predatory larvae of the antlion species Myrmeleon bore (Neuroptera: Myrmeleontidae). J Invertebr Pathol, 2007. 96 (1): p. 80-8. [43] Kiiskinen, L.L., M. Ratto, and K. Kruus, Screening for novel laccase-producing microbes. J Appl Microbiol, 2004. 97 (3): p. 640-6. [44] Field, J.A., et al., Screening for ligninolytic fungi applicable to the biodegradation of xenobiotics. Trends Biotechnol, 1993. 11 : p. 44-49. [45] Vikman, M., et al., The influence of lignin content and temperature on the biodegradation of lignocellulose in composting conditions. Appl Microbiol Biotechnol, 2002. 59 (4-5): p. 591-8. [46] Husain, Q., Potential applications of the oxidoreductive enzymes in the decolorization and detoxification of textile and other synthetic dyes from polluted water: a review. Crit Rev Biotechnol, 2006. 26 (4): p. 201-21. [47] van Schothorst, M. and A.M. Renaud, Malachite green pre-enrichment medium for improved salmonella isolation from heavily contaminated samples. J Appl Bacteriol, 1985. 59 (3): p. 223- 30. [48] Archibald, F.S., A new assay for lignin-type peroxidases employing the dye azure B. Appl Environ Microbiol, 1992. 58 (9): p. 3110-6. [49] Aguiar, A. and A. Ferraz, Fe( 3+ )- and Cu( 2+ )-reduction by phenol derivatives associated with Azure B degradation in Fenton-like reactions. Chemosphere, 2007. 66 (5): p. 947-54. [50] Arora, D.S. and P.K. Gill, Comparison of two assay procedures for lignin peroxidase. Enzyme Microb Technol, 2001. 28 (7-8): p. 602-605. [51] Pangallo, D., et al., Investigation of microbial community isolated from indoor artworks and air environment: identification, biodegradative abilities, and DNA typing. Can J Microbiol, 2009. 55 (3): p. 277-87. [52] Demidova, T.N. and M.R. Hamblin, Photodynamic inactivation of Bacillus spores, mediated by phenothiazinium dyes. Appl Environ Microbiol, 2005. 71 (11): p. 6918-25. [53] Ferreira-Leitao, V.S., J. Godinho da Silva, and E.P.S. Bon, Methylene blue and azure B oxidation by horseradish peroxidase:a comparative evaluation of class II and class III peroxidases. Applied Catalysis B: Environmental, 2003. 42 : p. 213-221. [54] Geib, S.M., et al., Lignin degradation in wood-feeding insects. Proc Natl Acad Sci U S A, 2008. 105 (35): p. 12932-7. [55] Chung, S.Y., et al., A Gram-positive Polychlorinated Biphenyl-degrading Bacterium, Rhodococcus erythropolis Strain TA421, Isolated from a Termite Ecosystem. Biosci Biotechnol Biochem, 1994. 58 : p. 2111-2113. [56] Harazono, K., et al., Isolation and Characterization of Aromatics-degrading Microorganisms from the Gut of the Lower Termite Coptotermes formosanus. Biosci Biotechnol Biochem, 2003. 67 (4): p. 889-892. [57] Eriksson, K.E., R.A. Blanchette, and P. Ander, Microbial and Enzymatic Degradation of Wood and Wood Components , ed. Springer. 1990, Berlin, Germany.

89 Chapter 3

[58] Hammel, K.E. and D. Cullen, Role of fungal peroxidases in biological ligninolysis. Curr Opin Plant Biol, 2008. 11 (3): p. 349-55. [59] Hammel, K.E., et al., Reactive oxygen species as agents of wood decay by fungi. Enzyme Microb Technol, 2002. 30 : p. 445-453. [60] Kersten, P. and D. Cullen, Extracellular oxidative systems of the lignin-degrading Basidiomycete Phanerochaete chrysosporium. Fungal Genet Biol, 2007. 44 (2): p. 77-87. [61] Biermann, C.J., Essentials of pulping and papermaking ed. C.J. Biermann. 1993, New york: Academic Press. [62] Gierer, J., The Reactions of Lignin during Pulping. A Description and Comparison of Conventional Pulping Processes Sven. Papperstidn., 1970. 73 : p. 571-596.

90 Bacterial strains exhibiting ligninolytic potential

91 Chapter 4

92 Azure B decolourization by a ligninolytic Bacillus sp.

Chapter 4

Decolourization of the lignin-model dye, Azure B by a ligninolytic Bacillus sp. and initial identification of enzymes involved

Manuscript submitted for publication as: Bandounas L, Pinkse M, de Winde JH, Ruijssenaars HJ. Decolourization of the lignin-model dye, Azure B by a ligninolytic Bacillus sp. and initial identification of enzymes involved. Submitted.

93 Chapter 4

Abstract

In this study we have investigated the molecular background of the ligninolytic potential of Bacillus sp. LD003. Strain LD003 was previously isolated on Kraft lignin and selected for its ligninolytic potential based on its ability to decolourize various lignin model dyes. Specifically Azure B (AB), a recalcitrant dye frequently utilised to assess ligninolytic activity, was decolourized efficiently. Proteins possibly involved in AB decolourization were partially purified, fractionated by gel electrophoresis and identified via mass spectrometry. Five candidate enzymes were selected and expressed in Escherichia coli. Of these only a quinone dehydrogenase was shown to decolourize AB. Thus, the quinone dehydrogenase was identified as an AB decolourizing enzyme of Bacillus sp. LD003. Although the ligninolytic potential of this enzyme remains to be established, it may be of industrial relevance for waste dye degradation and, possibly, lignin depolymerization.

94 Azure B decolourization by a ligninolytic Bacillus sp.

Introduction

Lignin is a complex aromatic polymer, which forms three-dimensional structures in the secondary cell walls of plants. The polymer is resistant to microbial degradation due to its high molecular weight and fairly stable ether and C-C bonds [1, 2]. Nevertheless, lignin can be degraded by specialized organisms such as white rot fungi, that attack the polymer with an array of oxidative enzymes and low-molecular weight mediators [3]. In addition to fungi, also bacteria have been reported to degrade lignin [4, 5]. Such bacteria are typically found in soils containing plant material, in the gut of wood-feeding insects, or in the rumen of herbivores [5, 6]. Besides fungal enzymes, also bacterial enzymes might be of special biotechnological interest, particularly for applications that require more specificity, thermostability, alkaline pH and are less dependent on mediators [2, 5, 7, 8].

Lignin degradation can be assessed by means of decolourization of synthetic dyes that share structural similarities with lignin. As the irregular structure and variable composition of lignin greatly complicates analysis, the decolourization of such dyes allows easy visual or spectrophotometric detection of ligninolytic activity [9]. The dye-decolourizing / ligninolytic capacity of these enzymes may furthermore involve a capacity to degrade recalcitrant xenobiotics such as polycyclic aromatic hydrocarbons (PAH’s), organochlorides, petroleum processing wastes, pesticides, and herbicides [9, 10]. It should be noted that also the lignin-mimicking synthetic dyes themselves may classify as recalcitrant xenobiotics, the wide-spread application of which in cosmetics, food, paper, printing, leather and textile is a major environmental concern [11].

A dye frequently employed to assess lignin degrading activity is Azure B. This recalcitrant thiazine dye is decolourized by high redox potential agents such as lignin peroxidase, but not by laccases or Mn peroxidases alone [12-14]. However, partial decolourization by laccases may occur in the presence of certain mediators, such as acetosyringone or p- coumaric acid [13, 15]. Azure B closely resembles Methylene Blue and Toluidene Blue O, of which all have a planar aromatic molecular structure with an intrinsic cationic charge (see Fig.1) [10, 16]. We have previously demonstrated that Bacillus sp. LD003, isolated from

95 Chapter 4 enrichment cultures on Kraft lignin, efficiently decolourized all of the thiazine dyes mentioned above. For this reason Bacillus sp. LD003 was considered to be a promising source of novel ligninolytic enzymes [17]. The present study aims to identify the enzyme(s) of Bacillus sp. LD003 responsible for the decolourization of Azure B. Insight into the ligninolytic potential of this strain may be exploited in industrial applications like controlled depolymerization of waste lignin or degradation of recalcitrant dyes.

Fig. 1 : Dye structures A) Methylene blue (MB), B) Azure B (AB), C) Toluidene Blue O (TB) [18] .

Materials and Methods

Strains and culture conditions The strains used in this study are listed in table 1. Bacillus sp. LD003 was isolated from enrichment cultures on industrial lignin as previously described (Bandounas et al. submitted). Other Bacillus strains ( B. subtilis , B. thuringiensis serovar konkukian and B. thuringiensis sv israelensis ) were obtained from the Bacillus Genetic Stock Center (BGSC). The growth media used were Luria broth (LB) [19] or a phosphate buffered mineral salts medium (MM) [20] with 20 mM glucose and 1 g/L yeast extract. For shake flask experiments, the Bacillus strains and Escherichia coli TOP10 were cultivated at 30 °C and 37 °C respectively, in a rotary shaker at 180 rpm and inoculated with cells from overnight cultures to a starting OD 600 of 0.05 – 0.1.

For experiments with Bacillus sp. LD003 spores, the culture was incubated at 30°C and 180 rpm for 5 days, after which the presence of spores was confirmed by microscopic inspection. Spores were harvested from stationary-phase culture of Bacillus sp. LD003 on

LB (OD 600 of 5.2). As vegetative control, a logarithmic-phase LB-culture of Bacillus sp.

96 Azure B decolourization by a ligninolytic Bacillus sp.

LD003 was used (OD 600 of 2.3). As inactivated controls, cells or spore suspensions were used that had been boiled for 2 h.

Construction of expression plasmids and expression testing The codon usage of the target genes possibly involved in Azure B decolourization was optimized for expression in Escherichia coli and the codon-optimized genes were synthesized by Eurofins MWG Operon. The genes were cloned into pBAD/gIIIA as Nco I- Xho I or Bsp HI-Xho I fragments and transformed to E. coli TOP10 cells (Table 1). For expression testing, quadruple cultures of each construct were cultivated in LB + 100 µg/ml ampicillin at 37 ºC to an OD 600 of approximately 0.5 – 0.8. Gene expression was induced by the addition of 0.005, 0.05 or 0.5 % (w/v) L-arabinose and incubation was continued for another 4 h at 37 ºC. Protein expression was assessed by SDS-PAGE and by Azure B decolourization assays.

Preparation of cell fractions A 50-ml overnight culture of Bacillus sp. LD003 was used to inoculate 1.5 L of LB in a 3-L fermentor vessel (New Brunswick Scientific). The working volume of the culture was set at 1.5-L by continuous overflow. A dilution rate ( D) of 0.2 h -1 was applied and the air flow was set at 1 L/min. After 18 h, 1 L of culture was removed for harvesting. The fermentor was allowed to refill to 1.5 L for further culturing and harvesting the following day. Cells were separated from the culture liquid by centrifugation at 3000 x g. The supernatant was stored at -80 °C and the cell pellet (40 g/L wet weight) was washed once with 50 mM Tris- HCl buffer (pH 7.6). The pellet was resuspended in an equivalent volume of 50 mM Tris-

HCl buffer (pH 7.6), 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 mM MgSO 4 and a few grains of solid DNAse. The cell suspension was immediately passed twice through a French Pressure Cell at 1.8 kbar, followed by the addition of 10 mM EDTA. The broken cells were centrifuged at 3000 x g for 10 min, after which the pellet was stored on ice and the supernatant was centrifuged at 50 000 x g for 60 min to collect the membrane fraction. All fractions were stored at -80 °C until required.

97 Chapter 4

Table 1 : Bacterial strains and plasmids used in this study Bacterial strains and Description Reference plasmids Bacillus sp. LD003 Isolated from soil [17] Bacillus thuringiensis sv Strain BGSC 4AH1 (Bacillus Genetic Stock - konkukian Center) Bacillus thuringiensis sv Strain BGSC 4Q1 (Bacillus Genetic Stock [21] israelensis Center) Strain BGSC 1A1 (Bacillus Genetic Stock Bacillus subtilis [22] Center) Escherichia coli TOP10 similar to strain DH10B™ Invitrogen TOP10, pBADgIIIa_AhpC, Accession number E. coli _ AhpC This study JF431000 TOP10, pBADgIIIa_AKR, Accession number E. coli _AKR This study JF431002 TOP10, pBADgIIIa_QD , Accession number E. coli_ QD This study JF431001 TOP10, pBADgIIIa_OYE, Accession number E. coli_ OYE This study JF430998 TOP10, pBADgIIIa_NR, Accession number E. coli_ NR This study JF430999 Plasmids Characteristics Reference pBR322 origin, ara BAD promoter, gene III pBADgIIIa secretion signal Invitrogen Alkyl hydroperoxide reductase (AhpC) from pBADgIIIa_AhpC Bacillus thuringiensis sv konkukian , cloned as This study 569-bp Nco I-Xho I fragment Aldo/keto reductase family oxidoreductase (AKR) from Bacillus thuringiensis sv pBADgIIIa_AKR This study konkukian , cloned as 835-bp Nco I-Xho I fragment NAD(P)H dehydrogenase (quinone) (QD) pBADgIIIa_QD from Bacillus thuringiensis sv konkukian , This study cloned as 549-bp Nco I-Xho I fragment OYE family NADH-dependent flavin oxidoreductase (OYE) from Bacillus pBADgIIIa_OYE This study thuringiensis sv konkukian , cloned as 1128- bp Nco I-Xho I fragment. NAD(P)H nitroreductase (NR), from Bacillus pBADgIIIa_NR thuringiensis sv konkukian , cloned as 644-bp This study Bsp HI-Xho I fragment Alternatively, for small-scale cell extract preparations, 150 ml of mid-exponential phase cells on LB were washed as described above and cell extracts were prepared from 30-fold concentrated cells by sonication, followed by centrifugation as described above.

98 Azure B decolourization by a ligninolytic Bacillus sp.

Protein gel electrophoresis Protein expression was assessed by SDS-PAGE. Culture samples (1 ml) were centrifuged and the cell pellet was resuspended in 300 μl of loading dye. After boiling for 10 min the sample was cooled and recentrifuged. A 20-µl aliquot of the supernatant was loaded onto a Criterion Tris-HCl gel (4-15 % acrylamide gradient; Bio-Rad) and run for 1 h at 120 V. Electrophoresis was performed using running buffer as described by Sambrook and Russell [19]. A 10 - 250 kD range prestained protein standard (Bio-Rad) was used as a marker and gels were stained with BioSafe Coomassie stain (Bio-Rad).

For zymogram stains, cell fractions were loaded (25 µl) in duplicate on Criterion Tris-HCl gels (4 -15 % acrylamide gradient; Bio-Rad) without SDS. The gels were run with native running buffer (50 mM TRIS, 380 mM glycine, pH 7) on ice at 100 V for 30 min, followed by 130 V for a further 2-3 h on ice (or until loading dye had migrated to the bottom of the gel). Duplicate lanes were either stained with BioSafe Coomassie or incubated overnight at 30 ºC in MM + 25 mg/L AB. Decolourized bands on the AB zymograms with corresponding stained band on the Coomassie gel were considered a positive reaction.

Decolourization assays The enzymatic decolourization of Azure B was determined spectrophotometrically using a µQuant MQX200 universal microplate spectrophotometer (Bio-tek). Azure B (10-25 mg/L) was incubated at 30 ºC with growing cells, spores, cell extract or cellular fractions and samples were drawn at regular intervals. After centrifuging, the absorption spectrum of the supernatant was measured between 200 and 800 nm and the pellet, if any, was visually inspected for dye adsorption. The decolourization was calculated as a percentage of the initial absorbance at λmax [23].

Analytical methods

Cell densities were measured at 600 nm (OD 600 ) using flat-bottom 96-well microplates (Greiner) in a µQuant MQX200 universal microplate spectrophotometer (Bio-tek). Cell dry weight (CDW) was calculated from the OD 600 value, assuming that an OD 600 of 1 corresponds to 0.42 g/L CDW. Protein concentrations of the cell extracts were determined

99 Chapter 4 spectrophotometrically at 595 nm using Bradford reagent (Sigma-Aldrich), measured on a Cary 50 spectrophotometer (Varian).

Dyes and chemicals The dyes Azure B (AB), Methylene Blue (MB) and Toluidene Blue O (TB) were purchased from Sigma-Aldrich. Stock solutions (0.5 g/L) were prepared in water, filter sterilized and stored at 4 ºC in the dark.

In-gel digestion, mass spectrometry and data analysis Protein bands excised from the zymogram stained gel were subjected to in-gel proteolytic digestion with trypsin overnight at 37 ºC. Prior to digestion, proteins were reduced and alkylated with respectively 5 mM dithiotreitol and 10 mM iodoacetamide, both in 25 mM ammonium bicarbonate, pH 8.0. In-gel digests were acidified by adding formic acid (5% final concentration) and were subjected to nanoscale liquid chromatography tandem mass spectrometry (nanoLC-MS/MS) analysis, using an Agilent 1200 binary HPLC solvent system (Agilent technologies, Waldbronn, Germany) essentially as previously described [24]. The nanoLC was connected to an LTQ/Orbitrap Velos mass spectrometer (Thermo Fisher, Bremen, Germany). For protein identifications, raw MS data were converted to peak lists using Proteome Discoverer 1.1 (Thermo Fisher). Spectra were searched against the NCBInr database (release November 2010; 11759209 sequences), using the Mascot search engine 2.2.0 ( www.matrixscience.com ), using a taxonomy restraint for Bacillus (3321572 sequences). Protein identification was accepted with individual peptide scores >40 and at least 3 peptide identifications per protein were considered significant for further analysis.

100 Azure B decolourization by a ligninolytic Bacillus sp.

Results

Bacillus sp. LD003 was previously isolated from soil by enrichment on Kraft lignin [17]. The strain was shown to efficiently decolourize the thiazine dyes Methylene blue (MB), Azure B (AB) and Toluidene Blue O (TB) in both liquid and solid phase assays [17]. The dye decolourizing activity was not observed in culture supernatants, suggesting that the activity was intracellular or cell-wall associated. No activity was detected in dead (boiled) controls, confirming that the dye decolourization was biogenic. Furthermore, the activity was not lost upon washing or freezing of the cells.

Decolourization of Azure B is oxygen dependent

Although ligninolytic and synthetic-dye decolourizing enzymes commonly are oxidative enzymes, it could not be excluded that the dye-decolourizing enzymes of Bacillus sp. LD003 were active under anaerobic conditions, since Bacillus sp. LD003 is a facultative aerobe. Thus, to determine whether the decolourization of Azure B required molecular oxygen, Bacillus sp. LD003 was grown aerobically and anaerobically on LB + 25 mg/L Azure B. The anaerobic incubations were flushed with nitrogen for 1 h prior to inoculation, followed by flushing for another 20 min after inoculation. AB was hardly decolourized in the anaerobic cultures (Fig. 2B), whereas the dye was decolourized by 90 % after 46 h in the presence of oxygen (Fig. 2A). Thus, the decolourization of AB by Bacillus sp. LD003 was clearly shown to be oxygen-dependent.

Fig. 2 : Decrease in absorbance spectrum of Azure B by Bacillus sp. LD003 in the presence and absence of oxygen. a) LB + 0.025g/L AB + oxygen, b) LB + 0.025g/L AB – oxygen. The

101 Chapter 4 data are averages of duplicate experiments. The maximum deviation between duplicate experiments was less than 10 %.

Bacillus sp. LD003 spores exhibit Azure B decolourizing activity

Potential ligninolytic enzymes such as laccases have been reported to be associated with endospores, such as CotA in B. subtilis [25]. Therefore, we compared the decolourization potential of Bacillus sp. LD003 spores to that of Bacillus sp. LD003 vegetative cells. Cultures and spore suspensions were harvested, washed and resuspended in MM, and incubated with Azure B at 30 °C for a maximum of 88 h.

Fig. 3: Absorbance spectra of incubations of Bacillus sp. LD003 (spores and vegetative cells) with Azure B (25 mg/L) in MM. a) Spores, b) Boiled spores, c) Vegetative cells, d) Boiled cells. Results are averages of duplicate experiments; the maximum deviation between duplicate experiments was less than 10 %.

Decolourization of AB was observed both in the vegetative controls and in the incubations with spores (Fig. 3 A and C). The inactivated controls showed a negligible absorbance decrease (Fig. 3 B, D). In all incubations, dark blue pellets were observed, suggesting that at least part of the dye adsorbed to cells, spores or debris. As dye adsorption was also

102 Azure B decolourization by a ligninolytic Bacillus sp. observed in the inactivated controls, the contribution to overall dye decolourization was considered negligible. Azure B was decolourized faster by vegetative cells than by spores, suggesting that either metabolically active cells were needed for efficient dye decolourization, or that the dye-decolourizing activity was not specifically associated with spores. In view of this observation, further experiments to identify the enzymes involved in AB decolourization were performed with growing cells rather than spore suspensions.

Table 2: Partial purification of the Azure B decolourizing activity of Bacillus sp. LD003. Total Specific Total Purification Cell fraction protein activity (U activity (U a) factor (mg) mg -1) intact cells b 1191 3400 c 0.35 1 cell extract 3928 741 5.3 13.25 cell debris 25 9.3 2.68 7.65 clarified cell 4789 705 6.79 19.4 extract membrane 63 28.5 2.21 6.31 fraction aOne unit (U) is defined as the % of Azure B decolourized in an Azure B decolourization assay at 30 ºC in 1 h. Presented values are averages of triplicate measurements after 23 h of incubation; maximum deviation from the mean was less than 10%. bCulture supernatant showed no decolourization of Azure B (not shown). cCalculated from cell wet weight: 34 g of wet cells correspond to 6.8 g of dry cells; a protein content of 50 % (w/w) in dry cells was assumed.

Partial purification of Azure B decolourizing activity from Bacillus sp. LD003

Enzymes potentially involved in AB decolourization were partially purified by fractionating a Bacillus sp. LD003 culture (see Material and Methods). The results are presented in Table 2. Remarkably, the total AB decolourizing activity of the cell extract exceeded that of whole cells. This observation suggested that the AB decolourizing activity is located intracellularly, and that the rate of AB decolourization with intact cells was determined by

103 Chapter 4 the rate of AB transport over the cell membrane. The low activity of the cell debris fraction supported a cytoplasmic location of the AB decolourizing enzyme(s). Upon clarifying the cell extract by removal of membrane fragments, the total activity increased even further. Apparently, compounds or cell fragments inhibiting the AB decolourizing enzyme were removed through ultracentrifugation.

Electrophoretic fractionation and detection of Azure B decolourizing activity

The Azure B decolourizing enzyme was purified 19-fold by fractionation as described above, and the clarified cell extract appeared to contain most activity. Therefore, this fraction was used as an enzyme source for further fractionation by native PAGE electrophoresis. The clarified cell extract was loaded onto a 4-15 % acrylamide-gradient Tris-HCl gel and run under native conditions. When the electrophoretic separation was completed, the gel was incubated with Azure B in order to identify protein bands capable of decolourizing the dye. The clarified cell extract showed a distinct decolourized band in the AB stained gel (Fig. 4A). The corresponding area on the Coomassie gel clearly stained blue, confirming the presence of protein at the AB decolourized spot (Fig. 4B). The Coomassie-stained band, located between 25 and 37 kDa, was cut from the zymogram and analyzed by mass spectrometry to identify the enzymes contained within the gel slice.

Identification of potential Azure B decolourizing enzymes

A large number of proteins were identified in the zymogram gel slice, but only proteins that were identified based on three or more peptide matches were considered significant (Table S1, Supplemental data). The Bacillus cereus family, which includes B. cereus, B. thuringiensis and B. anthracis, was used as the taxonomy reference. Therefore, several individual peptides showed multiple matches with similar proteins from different Bacillus species. Five proteins were selected as potential AB decolourizing enzymes, based on their

104 Azure B decolourization by a ligninolytic Bacillus sp. annotation as redox enzymes as well as their apparent abundance indicated by the number of peptide matches to a single protein based on the MS analyses.

Fig: 4: Zymogram analysis of Azure B decolourizing cell fractions. A) Azure B stained native PAGE gel; B) Coomassie-stained native PAGE gel. Lanes 1, cell extract; lanes 2, 3, clarified (ultracentifuged) cell extract. The arrows indicate the decolourized band in the Azure B zymogram and the corresponding band in the Coomassie-stained gel that was recovered from the gel for mass spectrometry analysis.

Fig. 5: Various Bacillus strains decolourizing LB + 0.025 g/L Azure B. A) Bacillus sp. LD003; B) B. thuringiensis sv konkukian , C) B. thuringiensis sv israelensis .

The selected enzymes (a peroxiredoxin, an Old Yellow Enzyme (OYE) family oxidoreductase, an aldo/keto reductase (AKR), a quinone dehydrogenase (QD), and a nitroreductase (NR) (Table 3)) were identified based on sequences present in various B. cereus, B. thuringiensis and B. anthracis strains. For all of these enzymes, however, genes encoding highly similar proteins were found in the publicly available genome sequence of B. thuringiensis sv konkukian 97-27 ( http://genome.jgi-psf.org/bacth/bacth.info.html ).

105 Chapter 4

This suggested that this strain may also be capable of decolourizing AB. Indeed, Bacillus thuringiensis sv konkukian 97-27, as well as Bacillus thuringiensis sv israelensis , were shown to decolourize AB (Fig. 5). This observation supported the notion that the dye- decolourizing activity was among the identified enzymes, and furthermore suggested that this property is a rather common trait in B. thuringiensis and related Bacillus strains.

Heterologous expression of potential Azure B decolourizing enzymes

Synthetic, codon-optimized genes of the selected redox proteins identified in the zymogram, based on the available sequences of B. thuringiensis sv konkukian (Table 1), were cloned into the pBAD expression system. The genes were expressed in E. coli by arabinose induction (Fig. 6). The OYE family NADH-dependent flavin oxidoreductase (OYE) and the NAD(P)H nitroreductase (NR) were not visibly expressed (SDS-PAGE) and therefore excluded from further study. For the Alkyl hydroperoxide reductase (AhpC), the Aldo/keto reductase family oxidoreductase (AKR) and the NAD(P)H quinone dehydrogenase (QD) cell extracts were prepared and Azure B decolourization experiments were performed.

Fig. 6: SDS-PAGE gel showing the expression of the various genes, as indicated by the arrow. Lane 1: uninduced E.coli _AhpC, lane 2 – 4: induced E.coli _AhpC at 0.005; 0.05 and 0.5 % arabinose concentrations, Lane 5: uninduced E.coli_ AKR, lane 6 – 8: induced E.coli _AKR. Lane 9: uninduced E.coli _OYE, lane 10 – 12: induced E.coli _OYE (no expression), lane 13: uninduced E.coli _QD, lane 14 – 16: induced E.coli _QD.

106

Table 3 : List of potential Azure B decolorizing proteins identified by MS-analysis of protein bands from the zymogram (Fig. 4).

Protein Accession Similar/identical protein in B. thuringiensis sv Protein size Protein identified using BLAST number of proteins konkukian 97-27 2, (kDa) [26] 1 identified using BLAST (% identity to protein identified with BLAST)

Peroxiredoxin ( Bacillus cereus ZP_03235070.1 Alkyl hydroperoxide reductase (BT9727_0314). 20.6 kDa H3081.97) (100% identity) NADH-dependent flavin ZP_03237104 Oye family NADH-dependent flavin 41.2 kDa oxidoreductase, OYE family oxidoreductase (BT9727_3398). (95% identity) (Bacillus cereus H3081.97) Aldo/keto reductase family NP_976544.1 Aldo/keto reductase family oxidoreductase 31.7 kDa oxidoreductase ( Bacillus cereus (BT9727_0184). ATCC 10987) (97% identity) Quinone family NAD(P)H NP_843355.1 NAD(P)H dehydrogenase (quinone) 21.1 kDa dehydrogenase ( Bacillus (BT9727_0747). anthracis str . Ames ) (100% identity) NAD(P)H nitroreductase ( Bacillus YP_895505.1 NAD(P)H nitroreductase (BT9727_2798). 23.4 kDa thuringiensis str . Al Hakam ) (98% identity) 1 Proteins were identified based on the best BLAST hits of peptides determined by mass spectrometry. 2 Proteins identified in B. thuringiensis sv konkukian 97-27 were 95 – 100 % similar to the proteins identified in the other Bacillus sp. mentioned in the table.

Chapter 4

For initial testing of dye-decolourizing activity, the cell extracts were spotted on MM plates containing 25 mg/L AB. The plates were incubated overnight at 30 °C and cell extract of Bacillus sp. LD003 was used as a positive control; cell extract of an E. coli transformant carrying an empty pBADgIlIa vector was used as negative control. As expected, the positive control (Fig. 7D) showed clear decolourization of AB. Also the cell extract of the QD-expressing E. coli strain showed AB decolourization (Fig. 7C). No AB decolourization was observed, however, by the cell extracts containing AhpC or AKR (Fig. 7A, B). Addition of reduced or oxidized cofactors (NAD(P), glutathione) did not affect the decolourization results in the spot test (not shown).

Fig. 7 : Cell extracts of E. coli expressing potential dye decolourizing enzymes of Bacillus sp. LD003 incubated at 30 °C overnight on MM + 0.025 g/L Azure B. A) E.coli _AhpC B) E.coli _AKR C) E.coli _QD D) Bacillus sp. LD003. Decolourization spot tests were performed in duplicate; single representative result shown here.

To confirm the Azure B decolourizing potential of the quinone dehydrogenase (QD), a liquid assay was performed in MM + 25 mg/L AB. Cell extracts of Bacillus sp. LD003 and the empty-vector E. coli strain were used as positive and negative control. TRIS buffer (pH 7) was furthermore included as an abiotic control. The assays were performed in the presence and absence of 1 mM NADH. Samples were incubated at 30 ºC; samples were drawn after 19 and 72 h and the supernatants were analyzed spectrophotometrically.

In all incubations some degree of Azure B decolourization was observed. However, the abiotic control showed less than 8% decolourization with or without NADH demonstrating that spontaneous or co-factor-associated decolourization was negligible. The negative E. coli control exhibited approximately 25% background AB decolourization after 72 h. E. coli _QD, however, showed considerably more AB decolourization: approximately 50% after 19 h with NADH. Slightly less decolourization was observed in the absence of co- factor (approximately 40%). As expected, the most extensive AB decolourization was observed with cell extract of Bacillus sp. LD003. Possibly, other enzymes and / or other

108 Azure B decolourization by a ligninolytic Bacillus sp. factors (co-factors, mediators) are involved in Azure B decolourization in Bacillus sp. LD003. Nevertheless, the results clearly demonstrated that QD is capable of decolourizing Azure B.

Fig. 8: Decolourization of Azure B (0.02 g/L in MM) by cell extracts in the presence and absence of 1 mM NADH. A) E. coli pBADgIIIa cell extract, B) E.coli _QD cell extract C) Bacillus sp. LD003 cell extract D) TRIS buffer. Experiments were performed in duplicate; error bars represent the maximum deviation from the mean.

Discussion

Five redox proteins were identified as potential AB decolourizing enzymes, based on MS analysis of partially purified AB decolourizing protein fractions. Three of these, an Alkyl hydroperoxide reductase (AhpC), an Aldo/keto reductase family oxidoreductase (AKR) and an NAD(P)H quinone dehydrogenase (QD) could be expressed in E. coli . Based on a literature survey, each of these enzymes was considered to be potentially associated with the Azure B-decolourizing activity of Bacillus sp. LD003.

Alkyl hydroperoxide reductases (Ahps) or peroxiredoxins belong to the class of non-haem peroxidases [27] with antioxidant properties, which can protect the cell against peroxides and reactive oxygen species (ROS) [27, 28]. AhpC, along with two putative peroxide 109 Chapter 4 sensor regulatory gene homologs (AhpX and OxyR), was identified in the lignocellulose degrading actinomycete Streptomyces viridosporus T7A. These stress-related genes were found to be involved in the response to peroxide and through OxyR, peroxides may induce transcription of lignin peroxidase [29, 30].

Aldo/keto reductases (AKRs) are associated with the detoxification of carbonyl compounds [33, 34]. If not present exogenously, carbonyl compounds can be produced in the cell by alcohol and aldehyde dehydrogenases, glutathione S-transferases (GSTs), or cytochrome P450s (CYPs) [35]. The oxidoreductases from the AKR superfamily have a broad substrate specificity ranging from aromatic aldehydes and ketones to aldoses, steroid hormones and monosaccharides [31]. Their precise physiological role is still unclear [32], and AKRs have not previously been associated with lignin degradation. A possible role in Azure B degradation cannot be excluded but since AB does not contain any carbonyl groups, it is not likely to be a direct substrate for AKRs.

Quinone NAD(P)H dehydrogenases (QDs) are ubiquitous flavin-dependent enzymes that have been implicated in, a.o., xenobiotic metabolism and protection from quinone-derived free radical toxicity [37, 38]. Quinones are naturally occurring in plants, fungi and bacteria as central components of electron transport chains. The one-electron reduction of quinones leads to the formation of semiquinones that may react with oxygen to generate ROS [37]. QDs catalyse the two-electron reduction of acceptor quinones to the less reactive hydroquinone species [40]. Thus, by preventing the formation of one electron- reduced semiquinones, QDs protect cells from oxidative stress and associated damage [37].

QDs have been implicated in lignin degradation by white rot fungi and in the first stages of wood decay by brown-rot fungi [48]. During the degradation of polymeric lignin, as well as aromatic and lignin model compounds, substituted quinones may occur as degradation intermediates, or as mediators that generate extracellular Fenton reagents [49-55]. Quinone metabolism is therefore important for lignin degradation, on the one hand to

110 Azure B decolourization by a ligninolytic Bacillus sp. protect the cell against the formation of quinone-derived free-radicals, on the other to produce such radicals to accelerate lignin degradation.

The observation that QD was found to decolourize AB suggested that AhpC and AKR do not play a role in AB decolourization. However, dysfunctional expression of AhpC and AKR cannot be fully excluded. Moreover, these enzymes may require specific mediators, cofactors, other cellular components or even additional enzymes, in order to decolourize AB. The reaction mechanism by which the Bacillus sp. LD003 QD decolourizes AB, however, is not clear. It has been reported that the thiazine dye methylene blue (MB) can bind to the active site of flavin-dependent enzymes [41, 42]. It can act as a tricyclic inhibitor, but also as an electron acceptor for flavin-bound NADH dehydrogenases [41, 43]. In the latter case, the dye is reduced to a colourless (leuco-)form. Considering the structural similarity to MB, AB may also be a suitable electron acceptor for such enzymes. This has indeed been demonstrated in biosensors, in which both MB and AB have been successfully used as electron transfer shuttles for diverse reductases such as horseradish peroxidases, nitrite and nitrate reductases [44-46]. These dyes may therefore play a similar role with QDs, as the catalytic site of various quinone reductases can accommodate a wide range of ring-containing substrates [37], including quinones with one to three fused rings [47].

QD-catalysed reduction of AB to its leuco-form is, however, unlikely as some form of reducing power is needed. By contrast, it was observed that AB decolourization occurred in the absence of reduced cofactors, most notably in the zymogram stains. Furthermore, even if the reducing power required for AB reduction would be from a less conventional source such as light, it would be expected that the leuco-form would be rapidly re-oxidized to the coloured form, resulting in little or no net decolourization. Instead, it was observed that AB was decolourized exclusively in the presence of oxygen. Hence, it appears likely that AB was truly degraded, possibly via the uncharged and more lipophilic quinoneimine form [41, 56] (see Fig. 9). This form results from deprotonation of the primary ring amino group, and several of such quinoneimines have been reported to be substrates for NAD(P)H quinone reductases [57]. Nevertheless, it was reported that reduced cofactors

111 Chapter 4 were also required for the –as of yet unelucidated- metabolism of quinoneimines. Therefore, the exact oxygen-dependent, reduced cofactor-independent mechanism by which QD apparently decolourizes AB remains to be elucidated.

Fig. 9 : Azure B oxidation-reduction and protonation-deprotonation reactions [18].

The Bacillus sp. LD003 investigated in this study utilizes a variety of aromatic compounds and decolourizes certain lignin-model dyes [17]. A quinone dehydrogenase (QD) was identified as (one of) the Azure B decolourizing enzyme(s) and this enzyme may therefore play a role in the ligninolytic potential of this strain. In addition to a potential application in lignin depolymerization, the QD may be employed for the degradation of lignin-derived intermediates, as well as for the detoxification or decolourization of certain dyes in waste streams. In spite of these findings, however, a more in-depth characterisation of the Bacillus sp. LD003 quinone dehydrogenase is necessary to clarify the actual role of this enzyme in lignin degradation as well as to provide a prospect of its application potential.

Acknowledgments

This project is financially supported by the Netherlands Ministry of Economic Affairs and the B-Basic partner organizations (www.b-basic.nl) through B-Basic, a public-private NWO- ACTS programme (ACTS = Advanced Chemical Technologies for Sustainability). This project was carried out within the research programme of the Kluyver Centre for Genomics of

112 Azure B decolourization by a ligninolytic Bacillus sp.

Industrial Fermentation which is part of the Netherlands Genomics Initiative / Netherlands Organization for Scientific Research. We would like to thank Alexandra Shchegoleva and Marc Stampraad for their assistance in the laboratory. The Netherlands Proteomics Centre, embedded in the Netherlands Genomics Initiative, is kindly acknowledged for financial support.

Supplemental data

Table S1, page 148

References

[1] Martinez, A.T., et al., Biodegradation of lignocellulosics: microbial, chemical, and enzymatic aspects of the fungal attack of lignin. Int Microbiol, 2005. 8(3): p. 195-204. [2] Masai, E., Y. Katayama, and M. Fukuda, Genetic and biochemical investigations on bacterial catabolic pathways for lignin-derived aromatic compounds. Biosci Biotechnol Biochem, 2007. 71 (1): p. 1-15. [3] Sanchez, C., Lignocellulosic residues: biodegradation and bioconversion by fungi. Biotechnol Adv, 2009. 27 (2): p. 185-94. [4] Vicuna, R., et al., Ability of natural bacterial isolates to metabolize high and low molecular weight lignin-derived molecules. Journal of Biotechnology, 1993. 30 : p. 9-13. [5] Zimmermann, W., Degradation of lignin by bacteria. J Biotechnol, 1990. 13 : p. 119-130. [6] Geib, S.M., et al., Lignin degradation in wood-feeding insects. Proc Natl Acad Sci U S A, 2008. 105 (35): p. 12932-7. [7] Kumar, R., S. Singh, and O.V. Singh, Bioconversion of lignocellulosic biomass: biochemical and molecular perspectives. J Ind Microbiol Biotechnol, 2008. 35 (5): p. 377-391. [8] Ruijssenaars, H.J. and S. Hartmans, A cloned Bacillus halodurans multicopper oxidase exhibiting alkaline laccase activity. Appl Microbiol Biotechnol, 2004. 65 (2): p. 177-82. [9] Kiiskinen, L.L., M. Ratto, and K. Kruus, Screening for novel laccase-producing microbes. J Appl Microbiol, 2004. 97 (3): p. 640-6. [10] Ferreira-Leitao, V.S., J. Godinho da Silva, and E.P.S. Bon, Methylene blue and azure B oxidation by horseradish peroxidase:a comparative evaluation of class II and class III peroxidases. Applied Catalysis B: Environmental, 2003. 42 : p. 213-221. [11] Zilly, A., et al., Decolorization of industrial dyes by a Brazilian strain of Pleurotus pulmonarius producing laccase as the sole phenol-oxidizing enzyme. Folia Microbiol (Praha), 2002. 47 (3): p. 273-7. [12] Arantes, V. and A.M. Ferreira Milagres, The synergistic action of ligninolytic enzymes (MnP and Laccase) and Fe3+-reducing activity from white-rot fungi for degradation of Azure B Enzyme Microb Technol, 2007. 42 : p. 17-22. [13] Archibald, F.S., A new assay for lignin-type peroxidases employing the dye azure B. Appl Environ Microbiol, 1992. 58 (9): p. 3110-6. [14] Aguiar, A. and A. Ferraz, Fe( 3+ )- and Cu( 2+ )-reduction by phenol derivatives associated with Azure B degradation in Fenton-like reactions. Chemosphere, 2007. 66 (5): p. 947-54. [15] Camarero, S., et al., Lignin-derived compounds as efficient laccase mediators for decolorization of different types of recalcitrant dyes. Appl Environ Microbiol, 2005. 71 (4): p. 1775-84. [16] Demidova, T.N. and M.R. Hamblin, Photodynamic inactivation of Bacillus spores, mediated by phenothiazinium dyes. Appl Environ Microbiol, 2005. 71 (11): p. 6918-25. [17] Bandounas, L., et al., Isolation and characterization of bacterial strains exhibiting ligninolytic potential. . Submitted.

113 Chapter 4

[18] Wainwright, M. and L. Amaral, The phenothiazinium chromophore and the evolution of antimalarial drugs. Trop Med Int Health, 2005. 10 (6): p. 501-11. [19] Sambrook, J. and E.F. Fritsch, Molecular Cloning: A Laboratory Manual ed. T. Maniatis. 1989, USA: Cold spring harbour. [20] Hartmans, S., Smits, J. P., van der Werf, M. J., Volkering, F., de Bont, J. A., Metabolism of Styrene Oxide and 2-Phenylethanol in the Styrene-Degrading Xanthobacter Strain 124X. Appl Environ Microbiol, 1989. 55 (11): p. 2850-2855. [21] Goldberg, L.J. and J. Margalit, A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Culex univitattus, Aedes aegypti and Culex pipiens. . Mosquito News, 1979. 37 : p. 355-358. [22] Kunst, F., et al., The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature, 1997. 390 (6657): p. 249-56. [23] Saratale, R.G., et al., Enhanced decolorization and biodegradation of textile azo dye Scarlet R by using developed microbial consortium-GR. Bioresour Technol, 2009. 100 (9): p. 2493-500. [24] Meiring, H.D., van der Heeft, E., ten Hove, G.J., de Jong, A. P. J. M., Nanoscale LC–MS (n) : technical design and applications to peptide and protein analysis. Journal of Separation Science, 2002. 25 (9): p. 557-568. [25] Hullo, M.F., et al., CotA of Bacillus subtilis is a copper-dependent laccase. J Bacteriol, 2001. 183 (18): p. 5426-30. [26] Altschul, S.F., Gish, W., Miller, W., Myers, E. W., Lipman, D. J., Basic local alignment search tool. J Mol Biol, 1990. 215 (3): p. 403-10. [27] Koua, D., et al., PeroxiBase: a database with new tools for peroxidase family classification. Nucleic Acids Res, 2009. 37 (Database issue): p. D261-6. [28] Rhee, S.G., H.Z. Chae, and K. Kim, Peroxiredoxins: a historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic Biol Med, 2005. 38 (12): p. 1543- 52. [29] Kirby, R., Actinomycetes and lignin degradation. Adv Appl Microbiol, 2006. 58 : p. 125-68. [30] Ramachandran, S., T.S. Magnuson, and D.L. Crawford, Isolation and analysis of three peroxide sensor regulatory gene homologs ahpC, ahpX and oxyR in Streptomyces viridosporus T7A--a lignocellulose degrading actinomycete. DNA Seq, 2000. 11 (1-2): p. 51-60. [31] Xu, D., et al., Methylglyoxal detoxification by an aldo-keto reductase in the cyanobacterium Synechococcus sp. PCC 7002. Microbiology, 2006. 152 (Pt 7): p. 2013-21. [32] Scoble, J., et al., Crystal structure and comparative functional analyses of a Mycobacterium aldo-keto reductase. J Mol Biol. 398 (1): p. 26-39. [33] Hyndman, D., et al., The aldo-keto reductase superfamily homepage. Chem Biol Interact, 2003. 143- 144 : p. 621-31. [34] Jornvall, H., E. Nordling, and B. Persson, Multiplicity of eukaryotic ADH and other MDR forms. Chem Biol Interact, 2003. 143-144 : p. 255-61. [35] Barski, O.A., S.M. Tipparaju, and A. Bhatnagar, The aldo-keto reductase superfamily and its role in drug metabolism and detoxification. Drug Metab Rev, 2008. 40 (4): p. 553-624. [36] Park, J.H., et al., Evidence for the aldo-keto reductase pathway of polycyclic aromatic trans-dihydrodiol activation in human lung A549 cells. Proc Natl Acad Sci U S A, 2008. 105 (19): p. 6846-51. [37] Deller, S., P. Macheroux, and S. Sollner, Flavin-dependent quinone reductases. Cell Mol Life Sci, 2008. 65 (1): p. 141-60. [38] Preusch, P.C., Lapachol inhibition of DT-diaphorase (NAD(P)H:quinone dehydrogenase). Biochem Biophys Res Commun, 1986. 137 (2): p. 781-7. [39] Miura, T., et al., Initial response and cellular protection through the Keap1/Nrf2 system during the exposure of primary mouse hepatocytes to 1,2-naphthoquinone. Chem Res Toxicol. 24 (4): p. 559-67. [40] Chakraborty, S., et al., Cloning and expression of a Clostridium kluyveri gene responsible for diaphorase activity. Biosci Biotechnol Biochem, 2008. 72 (3): p. 735-41. [41] Schirmer, R.H., et al., "Lest we forget you - methylene blue ..." Neurobiol Aging, 2011. [42] Schomburg, D. and I. Schomburg, NAD(P)H Dehydrogenase (quinone). in Springer Handbook of Enzymes . 2005, Springer Berlin Heidelberg. p. 187-206. [43] Argyrou, A., et al., Catalysis of diaphorase reactions by Mycobacterium tuberculosis lipoamide dehydrogenase occurs at the EH4 level. Biochemistry, 2003. 42 (7): p. 2218-28. [44] Brunetti, B., Ugo, P., Moretto, L.M., Martin, C.R. , Electrochemistry of phenothiazine and methylviologen biosensor electron-transfer mediators at nanoelectrode ensembles. J. Electroanal. Chem., 2000. 491 : p. 166. 114 Azure B decolourization by a ligninolytic Bacillus sp.

[45] Petit, C., et al., Horseradish Peroxidase Immobilized Electrode for Phenothiazine Analysis. Electroanalysis, 1998. 10 (18): p. 1241. [46] Strehlitz, B., Gründig, B., Vorlop, K. -D., Bartholmes, P., Kotte, H., Stottmeister, U., Artificial electron donors for nitrate and nitrite reductases usable as mediators in amperometric biosensors Fresenius J Anal Chem, 1994. 349 : p. 676. [47] Zhou, Z., et al., Kinetic and docking studies of the interaction of quinones with the quinone reductase active site. Biochemistry, 2003. 42 (7): p. 1985-94. [48] Brock, B.J., S. Rieble, and M.H. Gold, Purification and Characterization of a 1,4-Benzoquinone Reductase from the Basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol, 1995. 61 (8): p. 3076-81. [49] Joshi, D.K. and M.H. Gold, Degradation of 2,4,5-trichlorophenol by the lignin-degrading basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol, 1993. 59 (6): p. 1779-85. [50] Rieble, S., D.K. Joshi, and M.H. Gold, Purification and characterization of a 1,2,4-trihydroxybenzene 1,2-dioxygenase from the basidiomycete Phanerochaete chrysosporium. J Bacteriol, 1994. 176 (16): p. 4838-44. [51] Tuor, U., et al., Oxidation of phenolic arylglycerol beta-aryl ether lignin model compounds by manganese peroxidase from Phanerochaete chrysosporium: oxidative cleavage of an alpha-carbonyl model compound. Biochemistry, 1992. 31 (21): p. 4986-95. [52] Valli, K., et al., Degradation of 2,4-dinitrotoluene by the lignin-degrading fungus Phanerochaete chrysosporium. Appl Environ Microbiol, 1992. 58 (1): p. 221-8. [53] Valli, K. and M.H. Gold, Degradation of 2,4-dichlorophenol by the lignin-degrading fungus Phanerochaete chrysosporium. J Bacteriol, 1991. 173 (1): p. 345-52. [54] Valli, K., H. Wariishi, and M.H. Gold, Degradation of 2,7-dichlorodibenzo-p-dioxin by the lignin- degrading basidiomycete Phanerochaete chrysosporium. J Bacteriol, 1992. 174 (7): p. 2131-7. [55] Wariishi, H., Valli, K., Gold, M.H., Oxidative cleavage of a phenolic diarylpropane lignin model dimer by manganese peroxidase from Phanerochaete chrysosporium. Biochemistry, 1989. 28 : p. 6017–6023. [56] Wainwright, M., Mohr, H., Walker, W.H., Phenothiazinium derivatives for pathogen inactivation in blood products. Journal of Photochemistry and Photobiology B: Biology, 2007. 86 : p. 45–58. [57] Powis, G., See, K.L., Santone, K.S., Melder, D.C., Hodnett, E.M., Quinoneimines as substrates for quinone reductase (NAD(P)H: (quinone-acceptor)oxidoreductase) and the effect of dicumarol on their cytotoxicity. Biochem Pharmacol, 1987. 36 (15): p. 2473-2479.

115 Chapter 5

116 Discussion

Chapter 5

Discussion

117 Chapter 5

Discussion

Research into the use of lignocellulosic raw materials instead of fossil-based resources is extremely important for today’s society. The amount of available fossil resources is declining at an alarming rate, and concerns about the effects of global warming are rapidly increasing. The use of biological agents such as enzymes or microorganisms for the production of fuels and chemicals from renewable resources such as lignocellulose is important for a cleaner, more sustainable future. The transition from fossil-based to bio- based society obviously presents a huge scientific and technological challenge. Part of this challenge is the development of efficient microbial host systems for the production of (fine) chemicals from biomass.

An example is Pseudomonas putida S12, which has been engineered to constitute a platform organism for the production of chemicals [1-6]. Its solvent tolerance properties make this bacterium perfectly suited for the production and the biotransformation of hydrophobic (aromatic) compounds. Aromatic compounds are commonly produced from petroleum, but may also be produced from sugars or glycerol with the P. putida S12 platform. An alternative sustainable source of aromatics may be lignin. However, the lack of efficient procedures to depolymerize this complex aromatic polymer prevents the value-added exploitation of this ubiquitous renewable feedstock.

In this thesis, the metabolism of putrescine in P. putida S12 has been investigated in addition to enzymatic approaches to depolymerize lignin. Putrescine has been implicated in stress and solvent tolerance, and insight into its metabolism may contribute to the use of P. putida S12 as solvent tolerant production platform. Microorganisms capable of utilizing lignin were isolated and characterized in order to identify novel ligninolytic enzymes which could be employed in industrial processes, or in the depolymerization of lignin to high-value products.

118 Discussion

Putrescine catabolism in solvent-tolerant Pseudomonas putida S12 is multifaceted

Polyamine metabolism, in particular putrescine, was investigated in the solvent-tolerant P. putida S12 (Chapter 2) [7]. Polyamines generally occur in many organisms [8] as signalling and regulatory molecules in response to stress conditions. These range from the presence of reactive oxygen species (ROS), heat, UV and acid to osmotic pressure [9, 10]. Polyamines are also involved in the stabilization of membranes and sub-cellular structures for osmotic adaptation [11, 12]. Indications for a role of polyamines in solvent tolerance have been found in several transcriptomics studies on solvent-exposed or aromatics- producing P. putida [13, 14]. In addition to up-regulation of genes involved in polyamine metabolism and transport, an apparent activation of the arginine deiminase (ADI) pathway was observed. Although commonly associated with generating ATP under anoxic conditions, this pathway may also serve to generate polyamines under solvent-stressed conditions (unpublished data, Volkers, this laboratory).

After establishing that putrescine degradation was an inducible trait in P. putida S12, the putrescine degradative pathway in S12 was investigated by inactivating various genes which were up-regulated in response to putrescine. Unexpectedly, no mutant capable of efficiently degrading this molecule was obtained. Only a few of the knockout mutants exhibited slower growth rates and/or longer lag phases when grown on putrescine as carbon or nitrogen source, but putrescine was always degraded eventually. It was established that a large number of iso-genes and multiple pathways were involved in putrescine degradation. This high level of redundancy adds to the complexity of polyamine metabolism in P. putida S12. Apparently, maintaining polyamine homeostasis is crucial which presumably relates to the role of polyamines as regulatory and ancient stress molecules that affect living organisms at the systems level [10].

Based on our experience, it seems likely that a system-wide approach is necessary for controlling putrescine metabolism in P. putida S12. As putrescine can serve both as a source of carbon and nitrogen, putrescine metabolism likely is subject to regulation associated with both N- and C-metabolism. In Escherichia coli, nitrogen regulated (Ntr)

119 Chapter 5 proteins serve to assimilate ammonia, integrate ammonia assimilation with polyamine metabolism and glutamate synthesis, and scavenge nitrogen-containing compounds [15]. In Pseudomonas aeruginosa PAO1, a novel two-component system, CbrA-CbrB, which belongs to the NtrB-NtrC family, is responsible for controlling various pathways, as well as modulating the catabolism of various substrates in response to varying carbon:nitrogen ratios [16]. These genes however, were not differentially expressed in the wildtype P. putida S12 upon putrescine exposure. The σ54 (RpoN) transcription factor, which plays a key role in nitrogen metabolism [17, 18], may also play an important role in maintaining polyamine levels in P. putida S12. A P. aeruginosa rpoN mutant exhibited a growth defect on various substrates such as arginine, orthinine, putrescine or agmatine when used as sole carbon sources [16]. However this gene was also not amongst the genes in P. putida S12 which were differentially expressed in response to putrescine. In P. aeruginosa PAO1, the inactivation of the spuC gene encoding putrescine aminotransferase sufficed to cause the mutant to lose the ability to utilize putrescine as sole carbon or nitrogen source [19]. The deletion of this gene in P. putida S12, however, did not nearly have as great an impact on putrescine catabolism, as no major difference was observed whether putrescine was utilized as either sole carbon or nitrogen source. It appears that putrescine catabolism is multifaceted in P. putida S12, as indicated by the number of redundant genes and parallel pathways involved in putrescine catabolism. Many of these redundant genes were not differentially expressed after putrescine exposure, which seems to indicate a high level of control and stringent regulation in maintaining putrescine homeostasis in P. putida S12. Possibly, polyamines play a more important role in solvent tolerance than previously thought, which would provide an explanation for the complex nature of the polyamine metabolism and its regulation in solvent tolerant P. putida S12. Apparently, other less solvent tolerant organisms, such as P. aeruginosa and E.coli, do not have a strong necessity for maintaining polyamine homeostasis since the inactivation of only one, or a few genes was sufficient to disrupt putrescine catabolism.

120 Discussion

Lignin as a renewable feedstock for the production of bio-based chemicals and fuels

The key challenge for the bioconversion of lignin into high-value products usually lies in the recalcitrant and complex nature of the lignin polymer. Currently, Kraft lignin is the main type of “industrial” lignin as it is produced at large scale as waste stream of the pulping and paper industry [20]. Kraft lignin differs from natural lignin in many ways; it has a lower molecular weight and is highly modified. Some of the structural changes include side chain modifications, aryl-alkyl cleavages and condensation reactions resulting in a polymer which can fragment into smaller soluble moieties [21]. It is usually water- insoluble, but many commercial Kraft lignin preparations are available in a sulfomethylated, water-soluble form [22]. In view of its abundance, Kraft lignin was selected as the substrate to isolate the ligninolytic soil bacteria Pandoraea norimbergensis LD001, Pseudomonas sp. LD002 and Bacillus sp. LD003 described in Chapter 3 . All three strains demonstrated growth on both high molecular weight and low-molecular weight lignin fractions, although growth was generally slow and rather poor. The ability to utilize lignin monomers was also relatively limited for all three isolates. P. norimbergensis was furthermore unable to utilize a variety of conventional, easily accessible and fermentable sugars that are commonly found in lignocellulosic material.

The various treatment methods to improve the accessibility of lignocellulosic biomass result in numerous different, structural modifications of the lignin fraction. Thus, many different lignin preparations are available, all of which differ from the natural substrate [23]. The quantitative and qualitative chemical characterization of lignin is furthermore troublesome: for an unambiguous characterization of lignin, the monomeric composition, total lignin content, and the types of intermonomeric linkages should all be determined. Measuring lignin degradation is even more difficult, as the structure and properties of the polymer change during biodegradation [24]. The study of lignin degradation is not only hampered by the types of lignin preparations available, but also by the lack of convenient assay methods [25]. Organisms that are capable of degrading a specific type of lignin may not be able to degrade a different lignin preparation with the same efficiency. In contrast to lignin-mimicking dyes, radioactively labelled lignin is a “real” substrate that may be

121 Chapter 5 used to identify lignin degrading organisms. However, as this method is extremely time- consuming it is not suitable for large-scale screening experiments [26, 27]. Ahmad et al . (2010) have developed two spectrophotometric assays to monitor lignin degradation. The first method involves a fluorescently-labelled lignin; degradation leads to a change in the environment of the fluorophore, causing a measurable change in fluorescence. The second method is a UV-Vis assay based on chemically nitrated lignin. Degradation of this substrate leads to the release of nitrated phenols and therefore an increase in UV-Vis absorbance. A disadvantage of these methods is that chemically modified lignins are used that may give rise to biased or false (positive or negative) results.

Besides utilizing synthetic lignin to elicit lignolytic enzymes, much research into ligninolytic enzymes has been done based on coloured indicator compounds [28]. Many lignin- degrading fungi as well as numerous bacteria are able to decolourize such synthetic dyes, indicating their lignolytic enzymatic potential [29-33]. Lignin degrading capacity, as indicated by the decolourization of lignin model dyes, does not necessarily match the ability to utilize lignin (or lignin degradation products). Some microorganisms may be able to degrade lignin, but unable to utilize the degradation products released. This was observed in our study as described in Chapter 3 . P. norimbergensis achieved somewhat more efficient growth on high molecular weight lignin and the low molecular weight aromatic compounds than the other two isolates. However, it did not perform as well in the dye-decolourizing experiments. On the other hand, Bacillus sp. LD003 appeared to be better equipped to degrade lignin, in view of the variety of dyes it was able to decolourize, than to utilize the resulting degradation products. Possibly, such microorganisms constitute ligninolytic consortia, relying on each other for ligninolytic capacities and specific nutrients in a symbiotic relationship. The lignin-degraders may release aromatic compounds from lignin that are utilized by the nutrient providers that in their turn provide the nutrients on which the lignin-degraders can grow. Such microbial consortia may be more efficient in lignin degradation than a single species due to the combined efforts of certain organisms to degrade the complex lignin structure. In that case, the Bacillus sp. could be a more efficient lignin degrader that is not able to efficiently utilize lignin-derived compounds without additional nutrients. P. norimbergensis and Pseudomonas sp. may be

122 Discussion the less efficient lignin degraders that utilize the lignin-derived compounds and provide nutrients to other members of the ligninolytic consortium.

The use of lignin model dyes to identify ligninolytic enzymes has caused speculation regarding the relevance of the results obtained. Some reports show a limited correlation between dye decolourization and lignin degradation ability in soil- or wood-inhabiting fungi [34, 35]. However, many studies involving the white-rot basidiomycete, Phanerochaete chrysosporium, have reported a positive correlation between dye decolourization and lignin degradation activity. Here, the best lignocellulose degraders decolourized a specific dye more rapidly and to a greater extent than poor, or mediocre lignin degraders [34, 35]. This variation in the correlation between the extent of lignin degradation and dye decolourization, may be due to different dye decolourization mechanisms present in various species [34]. Alternatively, the microorganism’s growth phase may play an important role. For P. chrysosporium , lignin degradation and dye decolourization are secondary metabolic events [36-38], while lignin degradation occurs during the primary growth of certain soil inhabiting fungi and actinomycetes [39, 40]. Hence, lignin degradation and dye decolourization can be correlated to a specific growth phase or metabolic event, with variation between different organisms. Microorganisms that decolourize lignin-model dyes are generally assumed to exhibit ligninolytic activity, but obviously also lignin should be tested as a substrate to confirm the actual ligninolytic abilities.

The Bacillus sp. LD003 isolated from soil by enrichment on Kraft lignin efficiently decolourized thiazine dyes such as Methylene Blue (MB), Azure B (AB) and Toluidene Blue O (TB) (Chapter 3) . MB and AB have previously been used for the identification of lignin peroxidase activity [41, 42], however these are not the only enzymes decolourizing these compounds. MB has been reported to be decolourized and degraded by Arthrobacter globiformis via reduction, N-demethylation and deamination reactions [43]. Furthermore, a number of different targets have been identified for MB and/or AB, such as acetylcholine esterases, nitric oxide synthases, monoamine oxidase A, guanylate cyclase, methohemoglobin and disulfide reductases e.g. glutathione reductase or

123 Chapter 5 dihydrolipoamide dehydrogenase [44-46]. Specific chemical reactions have also been reported to be involved in lignin or dye degradation, such as Fenton reactions. Fenton reactions are thought to be involved in the initial decay of wood by brown-rot fungi when wood-degrading enzymes are not able to penetrate wood cell walls. Brown rot basidiomycetes cause wood decay via Fenton chemistry by producing reactive oxygen species (ROS), which typically attack the cellulose and hemicellulose in wood and thereby circumvent the lignin barrier [47-50]. In wood, free iron and molecular oxygen are readily available and brown rot fungi require mechanisms to reduce these compounds to generate the Fenton reaction [50]. The brown rot fungus Gloeophyllum trabeum uses a quinone redox cycle to generate extracellular Fenton reagent [50]. Various phenol derivatives found in wood extracts and in lignin degradation products [48, 51, 52], can also 3+ reduce Fe and in the presence of H 2O2 they are active in Fenton-like reactions [53]. AB was also reported to be degraded by a conventional Fenton reaction [53].

It may be clear from the above that Azure B decolourization may indicate, directly or indirectly, a variety of enzymatic activities including ligninases. In Chapter 4 , the Azure B decolourizing activity of Bacillus sp. LD003 was investigated in more detail. The activity was located intracellularly and several candidate enzymes were selected via a zymogram approach followed by MS analysis. It was found that an NAD(P)H quinone dehydrogenase (QD) was involved in the decolourization of AB. Such enzymes have been reported to utilize MB and TB as redox cyclers [54] and due to the structural relatedness of AB, it is very likely that this is also possible for AB. MB can either act as a substrate, or a noncompetitive inhibitor by interfering with redox cycling involving the flavin co-factor [55], when it interacts with flavin-dependent disulphide reductases or flavoenzymes [56]. The decolourization of AB by QD may also occur in a similar manner. QD may serve to protect the cell from oxidative stress caused by quinones [57], which have also been identified as degradation intermediates of lignin [58-61]. QD is involved in thiazine dye decolourization in Bacillus sp. LD003. Unfortunately, the mechanism of Azure B decolourization by the isolated QD has not been clarified. It could therefore not be excluded that AB was simply reduced to the (colourless) leuco-form. Further research will

124 Discussion be required to establish whether the QD has an actual role in the metabolism of lignin or lignin-related compounds.

Most of the lignin produced by the paper industry is presently incinerated in an energy- recovery step, and very little is exploited in another way [62-65]. In view of its rich and complex aromatic structure, this ubiquitous renewable resource has great potential as a feedstock for chemicals production. Some examples have been published, like the use of certain lignin preparations to replace phenol in phenolic resins. This allowed the production of modified polymers which are environmentally friendly, and lignin is less expensive than other binders used in the wood-composite industry [66, 67]. Another example is the production of a variety of low molecular weight aromatic compounds from lignin under high temperatures and pressures, e.g ., the fine chemical vanillin [68]. The presently developed lignin depolymerization processes are commonly energy intensive, environmentally unfriendly and vary in the yield and composition of the end products [69, 70]. Novel and biological lignin depolymerization reactions are therefore required to tap in the full potential of lignin as a renewable feedstock for the production of industrially relevant chemicals such as quinones, phenol, toluene, benzene, substituted coniferols, aromatic polyols and xylene [71].

Microbial Biodiversity: A source of novel lignin degraders?

As shown in Chapters 3 and 4, microorganisms sampled from the environment should be investigated for their ligninolytic abilities, with the aim of isolating and identifying the enzymes possibly involved in lignin degradation. An example of the possible diversity of microorganisms available in the environment, can be observed in the soil, where one gram of soil may contain up to 10 billion microorganisms of thousands of different species [72]. However, it is estimated that less than 1 % of the natural bacterial population can be isolated and/or cultured in the laboratory. Therefore, a vast majority of the population remains unknown in traditional enrichment approaches. This “unculturable” part of the microbial community may be explored via metagenomics approaches [73].

125 Chapter 5

Bacteria in the environment often live together in consortia with other microorganisms where they can adapt to microhabitats [74]. In a composting environment, bacteria can co-exist with fungi and through a synergistic relationship, complex and recalcitrant compounds such as lignocellulose are degraded once the more easily degradable substrates have been utilized [75]. The gut microbiota of termites and wood-feeding insects present a prime example of a microbial consortium that is able to digest lignocellulose [76]. Depending on the host’s feeding habits, and the location in the gut, microbiota of these wood-feeding insects present a huge source of functionally distinct communities with different specialisations, which may eventually play a role in our understanding of complex symbiotic relationships [76, 77] and their possible role in lignin degradation.

Concluding remarks

The need for cleaner and renewable feedstocks for chemical, fuel and energy production is evident, as fossil-based resources are limited and many current processes are environmentally damaging. Lignocellulosic biomass is a promising alternative for fuels and chemical production. More efficient treatment methods would result in a cleaner and higher yield of the feedstock. Furthermore, efficient use of the various feedstock components would serve to increase the value and utility of lignocellulosic biomass. The complex lignin polymer could rather be converted into aromatic-based chemicals or fuel additives, instead of being incinerated for heat and power generation. Research on lignin types, lignin degradation and conversion into products of interest such as described in this thesis is required to pave the way for efficient exploitation of this renewable resource. Once the genes or enzymes involved in lignin degradation have been identified and isolated, they can either be introduced into production strains like the solvent tolerant P. putida S12, or the enzymes may be utilized directly in industrial applications. Furthermore, the work in this thesis has provided additional indications of the involvement of polyamines in solvent tolerance of P. putida S12. These insights should be explored into more detail and may be exploited to improve the production of aromatic compounds, e.g .,

126 Discussion substituted aromatics, via biotransformation of lignin or via bioconversion of lignocellulosic hydrolysate.

References

[1] Meijnen, J.P., de Winde, J. H., Ruijssenaars, H. J., Engineering Pseudomonas putida S12 for efficient utilization of D-xylose and L-arabinose. Appl Environ Microbiol, 2008. 74 (16): p. 5031-7. [2] Nijkamp, K., van Luijk, N., de Bont, J. A., Wery, J., The solvent-tolerant Pseudomonas putida S12 as host for the production of cinnamic acid from glucose. Appl Microbiol Biotechnol, 2005. 69 (2): p. 170- 7. [3] Nijkamp, K., Westerhof, R. G., Ballerstedt, H., de Bont, J. A., Wery, J., Optimization of the solvent- tolerant Pseudomonas putida S12 as host for the production of p-coumarate from glucose. Appl Microbiol Biotechnol, 2007. 74 (3): p. 617-24. [4] Verhoef, S., Ruijssenaars, H. J., de Bont, J. A., Wery, J., Bioproduction of p-hydroxybenzoate from renewable feedstock by solvent-tolerant Pseudomonas putida S12. J Biotechnol, 2007. 132 (1): p. 49- 56. [5] Verhoef, S., Wierckx, N., Westerhof, R. G., de Winde, J. H., Ruijssenaars, H. J., Bioproduction of p- hydroxystyrene from glucose by the solvent-tolerant bacterium Pseudomonas putida S12 in a two- phase water-decanol fermentation. Appl Environ Microbiol, 2009. 75 (4): p. 931-6. [6] Wierckx, N.J., Ballerstedt, H., de Bont, J. A., and J. Wery, Engineering of solvent-tolerant Pseudomonas putida S12 for bioproduction of phenol from glucose. Appl Environ Microbiol, 2005. 71 (12): p. 8221-7. [7] Wierckx, N.J., et al., Transcriptome analysis of a phenol-producing Pseudomonas putida S12 construct: genetic and physiological basis for improved production. J Bacteriol, 2008. 190 (8): p. 2822-30. [8] Kurihara, S., et al., A novel putrescine utilization pathway involves gamma-glutamylated intermediates of Escherichia coli K-12. J Biol Chem, 2005. 280 (6): p. 4602-8. [9] Chou, H.T., et al., Transcriptome analysis of agmatine and putrescine catabolism in Pseudomonas aeruginosa PAO1. J Bacteriol, 2008. 190 (6): p. 1966-75. [10] Rhee, H.J., E.J. Kim, and J.K. Lee, Physiological polyamines: simple primordial stress molecules. J Cell Mol Med, 2007. 11 (4): p. 685-03. [11] Cohen, S.S., Introduction to the Polyamines. , ed. E. Cliffs. 1971, NJ: USA Prentice-Hall. [12] Joshi, N.R., A.D. Agate, and K.M. Paknikar, Polyamine patterns in iron- and sulphur-oxidizing bacteria isolated from an Indian copper mine indicate requirement of spermidine for growth under acid conditions World Journal of Microbiology & Biotechnology, 2000. 16 : p. 631-634. [13] Verhoef, S., Ballerstedt, H., Volkers, R. J., de Winde, J. H., Ruijssenaars, H. J., Comparative transcriptomics and proteomics of p-hydroxybenzoate producing Pseudomonas putida S12: novel responses and implications for strain improvement. Appl Microbiol Biotechnol, 2010. 87 (2): p. 679-90. [14] Volkers, R.J., Ballerstedt, H., Ruijssenaars, H., de Bont, J. A., de Winde, J. H., Wery, J., TrgI, toluene repressed gene I, a novel gene involved in toluene-tolerance in Pseudomonas putida S12. Extremophiles, 2009. 13 (2): p. 283-97. [15] Reitzer, L., Nitrogen assimilation and global regulation in Escherichia coli. Annu Rev Microbiol, 2003. 57 : p. 155-76. [16] Nishijyo, T., D. Haas, and Y. Itoh, The CbrA-CbrB two-component regulatory system controls the utilization of multiple carbon and nitrogen sources in Pseudomonas aeruginosa. Mol Microbiol, 2001. 40 (4): p. 917-31. [17] Cases, I., D.W. Ussery, and V. de Lorenzo, The sigma54 regulon (sigmulon) of Pseudomonas putida. Environ Microbiol, 2003. 5(12): p. 1281-93. [18] Reitzer, L. and B.L. Schneider, Metabolic context and possible physiological themes of sigma(54)- dependent genes in Escherichia coli. Microbiol Mol Biol Rev, 2001. 65 (3): p. 422-44, table of contents. [19] Lu, C.D., Itoh, Y., Nakada, Y., Jiang, Y., Functional analysis and regulation of the divergent spuABCDEFGH-spuI operons for polyamine uptake and utilization in Pseudomonas aeruginosa PAO1. J Bacteriol, 2002. 184 (14): p. 3765-73. [20] De los Santos Ramos, W., et al., Remediation of lignin and its derivatives from pulp and paper industry wastewater by the combination of chemical precipitation and ozonation. J Hazard Mater, 2009. 169 (1- 3): p. 428-34.

127 Chapter 5

[21] Raj, A., et al., Biodegradation of kraft lignin by a newly isolated bacterial strain, Aneurinibacillus aneurinilyticus from the sludge of a pulp paper mill. World J Microbial Biotechnol, 2007. 23 : p. 793- 799. [22] Adler, E., M. Hagglund, and E. Karl, Method of producing water-soluble products from black liquor lignin U.P. Publication, Editor. 1954, SVENSKA CELLULOSAFORENINGENS C [23] Crawford, R.L., Lignin Biodegradation and Transformation. , ed. J.W.a. Sons. 1981: Wiley-Interscience. 154. [24] Monties, B., Chemical assesment of lignin biodegradation some qualitative and quantitative aspects. FEMS Microbiol Rev, 1994. 13 : p. 277-284. [25] Kern, H.W. and T.K. Kirk, Influence of Molecular Size and Ligninase Pretreatment on Degradation of Lignins by Xanthomonas sp. Strain 99. Appl Environ Microbiol, 1987. 53 (9): p. 2242-2246. [26] Ahmad, M., et al., Development of novel assays for lignin degradation: comparative analysis of bacterial and fungal lignin degraders. Mol. BioSyst, 2010. 6: p. 815-821. [27] Zimmermann, W., Degradation of lignin by bacteria. J Biotechnol, 1990. 13 : p. 119-130. [28] Kiiskinen, L.L., M. Ratto, and K. Kruus, Screening for novel laccase-producing microbes. J Appl Microbiol, 2004. 97 (3): p. 640-6. [29] Dube, E., et al., Decolourization of recalcitrant dyes with a laccase from Streptomyces coelicolor under alkaline conditions. J Ind Microbiol Biotechnol, 2008. 35 (10): p. 1123-9. [30] Itoh, K., C. Yatome, and T. Ogawa, Biodegradation of anthraquinone dyes by Bacillus subtilis. Bull Environ Contam Toxicol, 1993. 50 (4): p. 522-7. [31] Kalyani, D.C., et al., Ecofriendly biodegradation and detoxification of Reactive Red 2 textile dye by newly isolated Pseudomonas sp. SUK1. J Hazard Mater, 2009. 163 (2-3): p. 735-42. [32] Zilly, A., et al., Decolorization of industrial dyes by a Brazilian strain of Pleurotus pulmonarius producing laccase as the sole phenol-oxidizing enzyme. Folia Microbiol (Praha), 2002. 47 (3): p. 273-7. [33] Kaushik, P. and A. Malik, Fungal dye decolourization: recent advances and future potential. Environ Int, 2009. 35 (1): p. 127-41. [34] Falcón, M.A., et al., Isolation of microorganisms with lignin transformation potential from soil of Tenerife island. . Soil Biol Biochem, 1995. 27 : p. 121-126. [35] Pasti, M.B. and D.L. Crawford, Relationships between the abilities of streptomycetes to decolorize three anthron-type dyes and to degrade lignocellulose. . Can J Microbiol 1991. 37 : p. 902-907. [36] Eriksson, K.E., R.A. Blanchette, and P. Ander, Microbial and Enzymatic Degradation of Wood and Wood Components , ed. Springer. 1990, Berlin, Germany. [37] Keyser, P., T.K. Kirk, and J.G. Zeikus, Ligninolytic enzyme system of Phanaerochaete chrysosporium: synthesized in the absence of lignin in response to nitrogen starvation. J Bacteriol, 1978. 135 (3): p. 790-7. [38] Kirk, T.K. and R.L. Farrell, Enzymatic "combustion": the microbial degradation of lignin. Annu Rev Microbiol, 1987. 41 : p. 465-505. [39] Ball, A.S., W.B. Betts, and A.J. McCarthy, Degradation of Lignin-Related Compounds by Actinomycetes. Appl Environ Microbiol, 1989. 55 (6): p. 1642-1644. [40] McCarthy, A.J., Broda P., Screening for lignin degrading actinomycetes and characterization of their activity against (‘*C) lignin-labeled wheat lignocellulose. . Journal of General Microbiology, 1984. 130 : p. 2905-2913. [41] Archibald, F.S., A new assay for lignin-type peroxidases employing the dye azure B. Appl Environ Microbiol, 1992. 58 (9): p. 3110-6. [42] Ferreira-Leitao, V.S., J. Godinho da Silva, and E.P.S. Bon, Methylene blue and azure B oxidation by horseradish peroxidase:a comparative evaluation of class II and class III peroxidases. Applied Catalysis B: Environmental, 2003. 42 : p. 213-221. [43] Itoh, K., Decolorization and degradation of methylene blue by Arthrobacter globiformis. Bull Environ Contam Toxicol, 2005. 75 (6): p. 1131-6. [44] Buchholz, K., et al., Interactions of methylene blue with human disulfide reductases and their orthologues from Plasmodium falciparum. Antimicrob Agents Chemother, 2008. 52 (1): p. 183-91. [45] Harvey, B.H., et al., Role of monoamine oxidase, nitric oxide synthase and regional brain monoamines in the antidepressant-like effects of methylene blue and selected structural analogues. Biochem Pharmacol, 2010. 80 (10): p. 1580-91. [46] Juffermans, N.P., et al., A dose-finding study of methylene blue to inhibit nitric oxide actions in the hemodynamics of human septic shock. Nitric Oxide, 2010. 22 (4): p. 275-80.

128 Discussion

[47] Cohen, R., M.R. Suzuki, and K.E. Hammel, Differential stress-induced regulation of two quinone reductases in the brown rot basidiomycete Gloeophyllum trabeum. Appl Environ Microbiol, 2004. 70 (1): p. 324-31. [48] Hammel, K.E., et al., Reactive oxygen species as agents of wood decay by fungi. Enzyme Microb Technol, 2002. 30 : p. 445-453. [49] Kerem, Z., hammel, and K.E. Hammel, Biodegradative mechanism of the brown rot basidiomycete Gloeophyllum trabeum: evidence for an extracellular hydroquinone-driven fenton reaction. FEBS Lett, 1999. 446 (1): p. 49-54. [50] Prousek, J., Fenton chemistry in biology and medicine. Pure Appl. Chem., 2007. 79 (12): p. 2325–2338. [51] Aguiar, A., P. Souza-Cruz, and A. Ferraz, Oxalic acid, Fe3+-reduction activity and oxidative enzymes detected in culture extracts recovered from Pinus taeda wood chips biotreated by Ceriporiopsis subvermispora. Enzyme Microb Technol, 2006. 38 : p. 873-878. [52] Ferraz, A., et al., Occurrence of iron-reducing compounds in biodelignified ‘‘palo podrido’’ wood samples. Int. Biodeterior. Biodegrad., 2001. 47 : p. 203–208. [53] Aguiar, A. and A. Ferraz, Fe( 3+ )- and Cu( 2+ )-reduction by phenol derivatives associated with Azure B degradation in Fenton-like reactions. Chemosphere, 2007. 66 (5): p. 947-54. [54] Wondrak, G.T., NQO1-activated phenothiazinium redox cyclers for the targeted bioreductive induction of cancer cell apoptosis. Free Radic Biol Med, 2007. 43 (2): p. 178-90. [55] Haynes, R.K., et al., Reactions of antimalarial peroxides with each of leucomethylene blue and dihydroflavins: flavin reductase and the cofactor model exemplified. ChemMedChem, 2010. 6(2): p. 279-91. [56] Schirmer, R.H., et al., "Lest we forget you - methylene blue ..." Neurobiol Aging, 2011. [57] Brock, B.J., S. Rieble, and M.H. Gold, Purification and Characterization of a 1,4-Benzoquinone Reductase from the Basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol, 1995. 61 (8): p. 3076-81. [58] Joshi, D.K. and M.H. Gold, Degradation of 2,4,5-trichlorophenol by the lignin-degrading basidiomycete Phanerochaete chrysosporium. Appl Environ Microbiol, 1993. 59 (6): p. 1779-85. [59] Rieble, S., D.K. Joshi, and M.H. Gold, Purification and characterization of a 1,2,4-trihydroxybenzene 1,2-dioxygenase from the basidiomycete Phanerochaete chrysosporium. J Bacteriol, 1994. 176 (16): p. 4838-44. [60] Tuor, U., et al., Oxidation of phenolic arylglycerol beta-aryl ether lignin model compounds by manganese peroxidase from Phanerochaete chrysosporium: oxidative cleavage of an alpha-carbonyl model compound. Biochemistry, 1992. 31 (21): p. 4986-95. [61] Valli, K., et al., Degradation of 2,4-dinitrotoluene by the lignin-degrading fungus Phanerochaete chrysosporium. Appl Environ Microbiol, 1992. 58 (1): p. 221-8. [62] Gosselink, R.J.A., et al., Co-ordination network for lignin—standardisation, production and applications adapted to market requirements (EUROLIGNIN) Ind. Crops Prod, 2004. 20 : p. 121-129. [63] Himmel, M.E., et al., Biomass recalcitrance: engineering plants and enzymes for biofuels production. Science, 2007. 315 (5813): p. 804-7. [64] Ragauskas, A.J., et al., The path forward for biofuels and biomaterials. Science, 2006. 311 (5760): p. 484-9. [65] Zaldivar, J., J. Nielsen, and L. Olsson, Fuel ethanol production from lignocellulose: a challenge for metabolic engineering and process integration. Appl Microbiol Biotechnol, 2001. 56 (1-2): p. 17-34. [66] Stewart, D., Lignin as a base material for materials applications: Chemistry, application and economics. Ind. Crops Prod, 2008. 27 : p. 202-207. [67] Tejado, A., et al., Lignins for phenol replacement in novolac-type phenolic formulations. II. Flexural and compressive mechanical properties. J. Appl. Polym.Sci, 2008. 107 : p. 159-165. [68] Bjørsvik, H.R. and L. Liguori, Organic Processes to Pharmaceutical Chemicals Based on Fine Chemicals from Lignosulfonates. . Organic Process Research & Developement, 2002. 6: p. 279-290. [69] Pandey, M.P. and C.S. Kim, Lignin Depolymerization and Conversion: A Review of Thermochemical Methods. Chem Engin Technol, 2011. 34 (1): p. 29-41. [70] Ward, P.O. and A. Singh, Bioethanol technology: developments and perspectives. Advan Appl Microbiol, 2002. 51 : p. 53-80. [71] Calvo-Flores, F.G. and J.A. Dobado, Lignin as renewable raw material. ChemSusChem, 2010. 3(11): p. 1227-35. [72] Rossello-Mora, R. and R. Amann, The species concept for prokaryotes. FEMS Microbiol Rev, 2001. 25 (1): p. 39-67.

129 Chapter 5

[73] Trigo, A., A. Valencia, and I. Cases, Systemic approaches to biodegradation. FEMS Microbiol Rev, 2009. 33 (1): p. 98-108. [74] Torsvik, V. and L. Ovreas, Microbial diversity and function in soil: from genes to ecosystems. Curr Opin Microbiol, 2002. 5(3): p. 240-5. [75] Perez, J., et al., Biodegradation and biological treatments of cellulose, hemicellulose and lignin: an overview. Int Microbiol, 2002. 5(2): p. 53-63. [76] Brune, A., Microbiology: woodworker's digest. Nature, 2007. 450 (7169): p. 487-8. [77] Brune, A., Symbiotic Associations, Biotechnology, Applied Microbiology. in The Prokaryotes , M. Dworkin, et al., Editors. 2006, Springer: New York. p. 439-474.

130 Discussion

131 Supplemental data

Supplemental data

Supplemental data for Chapter 2 Supplemental data for Chapter 4

132 Supplemental data

Supplemental data for Chapter 2 Table S1: All differentially expressed genes with fold changes (expression values) after 15 minutes (t=15) and 30 minutes (t=30) following putrescine exposure, compared to the condition prior to putrescine addition (t=0).

Primary annotation (from KT2440 Fold change Fold change S12 KT2440 database) (t=15 min) (t=30 min) locus tag glucose-inhibited division protein A 1.2 1.4 PP0004 conserved hypothetical protein TIGR00278 1.1 1.5 PP0007 ribonuclease P protein component 1.1 1.4 PP0008 ribosomal protein L34 1.2 1.4 PP0009 DNA replication and repair protein RecF 1.2 1.4 PP0012 DNA gyrase, subunit B 1.2 1.5 PP0013 transposase, putative 1.2 1.5 PP0014 hypothetical protein 1.0 0.6 PP0023 DNA-binding response regulator CzrR 0.5 0.7 PP0029 sensor histidine kinase 0.5 0.7 PP0030 GtrA family protein 0.5 0.6 PP0035 porin P 0.7 0.7 PP0037 hypothetical protein 0.6 0.6 PP0042 porin, putative 0.6 0.5 PP0046 DNA-binding heavy metal response regulator, putative 0.5 0.5 PP0047 hypothetical protein 0.6 0.6 PP0049 histidinol-phosphatase, putatitve 1.1 1.4 PP0059 glycyl-tRNA synthetase, tetrameric type, beta subunit 1.1 1.5 PP0060 sun protein 1.1 1.5 PP0066 coproporphyrinogen III oxidase, aerobic 1.4 1.4 PP0073 shikimate 5-dehydrogenase 1.1 1.4 PP0074 hypothetical protein 0.7 1.0 PP0078 hypothetical protein 0.8 0.5 PP0087 carbonic anhydrase 0.5 0.9 PP0100 sulfate transporter, putative 0.6 0.9 PP0101 hypothetical protein 0.5 0.5 PP0102 hypothetical protein 1.0 1.4 PP0122 cytochrome c-type protein 1.1 1.5 PP0125 cytochrome c4 1.1 1.6 PP0126 proton sodium-glutamate aspartate symporter 0.6 1.2 PP0137 hypothetical protein 1.7 4.3 PP0153 acetyl-CoA hydrolase transferase family protein 0.6 0.5 PP0154 pyridine nucleotide transhydrogenase, 0.3 0.2 PP0155 beta subunit surface adhesion protein, putative 1.4 1.2 PP0168 transcriptional factor-related protein 0.5 0.8 PP0173 cytochrome c family protein 1.8 1.7 PP0180 alginate biosynthesis protein PprA 0.7 0.7 PP0185 nitrate ABC transporter, periplasmic nitrate- binding protein, putative 1.7 1.6 PP0207 succinate-semialdehyde dehydrogenase 2.8 3.7 PP0213 4-aminobutyrate aminotransferase 2.6 4.0 PP0214 ABC transporter, permease protein 0.9 1.5 PP0219 porin E 1.4 1.1 PP0234 NADH-dependent FMN reductase 1.0 1.5 PP0236 hypothetical protein 1.1 1.6 PP0244

133 Supplemental data

Fold Primary annotation (from KT2440 Fold change S12 KT2440 change database) (t=15 min) locus tag (t=30 min) hypothetical protein 0.6 0.8 PP0278 amino acid ABC transporter, permease protein 0.6 0.9 PP0280 gamma-aminobutyrate transporter, putative 1.4 1.1 PP0284 oxalate formate antiporter, putative 0.9 0.6 PP0288 imidazoleglycerol phosphate synthase, cyclase subunit 1.0 1.4 PP0293 low-specificity L-threonine aldolase 0.6 0.5 PP0321 pyruvate dehydrogenase, dihydrolipoamide acetyltransferase component 1.3 1.5 PP0338 pyruvate dehydrogenase, E1 component 1.4 1.6 PP0339 RNA polymerase sigma-70 factor, ECF subfamily 1.2 1.4 PP0352 CBS domain protein 0.5 0.7 PP0354 response regulator 0.6 0.7 PP0355 biotin biosynthesis protein BioC 0.9 0.6 PP0365 acyl-CoA dehydrogenase, putative 0.5 0.8 PP0368 acetylornithine aminotransferase, putative 0.6 0.5 PP0372 Pmp3 family protein 0.3 0.1 PP0373 prolyl oligopeptidase family protein 0.4 0.1 PP0375 coenzyme PQQ synthesis protein E 0.4 0.1 PP0376 coenzyme PQQ synthesis protein D 0.4 0.1 PP0377 coenzyme PQQ synthesis protein C 0.3 0.1 PP0378 coenzyme PQQ synthesis protein B 0.2 0.1 PP0379 coenzyme PQQ synthesis protein A 0.3 0.1 PP0380 O-sialoglycoprotein endopeptidase 1.0 1.5 PP0390 conserved hypothetical protein TIGR00023 1.0 1.5 PP0391 hypothetical protein 0.7 0.5 PP0397 rhodanese domain protein 0.6 0.8 PP0398 polyamine ABC transporter, ATP-binding protein 22.4 12.8 PP0411 polyamine ABC transporter, periplasmic polyamine-binding protein 18.4 15.3 PP0412 polyamine ABC transporter, permease protein 31.0 27.7 PP0413 polyamine ABC transporter, permease protein 10.7 15.5 PP0414 ribulose-phosphate 3-epimerase 1.2 1.5 PP0415 lipase, GDSL family 1.1 0.5 PP0418 hypothetical protein 0.6 0.6 PP0419 hypothetical protein 1.1 1.4 PP0427 ribosomal protein S14 1.1 1.4 PP0467 bacterioferritin 0.5 0.7 PP0482 formate dehydrogenase, iron-sulfur subunit 0.7 0.7 PP0490 formate dehydrogenase, cytochrome b556 subunit 0.5 0.4 PP0491 formate dehydrogenase accessory protein FdhE, putative 0.4 0.5 PP0492 L-seryl-tRNA selenium transferase 0.6 0.4 PP0493 selenocysteine-specific translation elongation factor 0.5 0.5 PP0494 outer membrane protein OprG 1.3 2.2 PP0504 ethanolamine ammonia-lyase, heavy subunit 0.7 0.5 PP0543 aldehyde dehydrogenase family protein 1.2 0.6 PP0545 sigma-54 dependent transcriptional regulator 1.0 0.7 PP0546 translation initiation factor SUI1 0.3 0.8 PP0566 biosynthetic arginine decarboxylase 0.2 0.9 PP0567 DNA-binding response regulator, LuxR family 0.7 0.4 PP0574

134 Supplemental data

Fold Primary annotation (from KT2440 Fold change S12 KT2440 change database) (t=15 min) locus tag (t=30 min) methyl-accepting chemotaxis transducer 1.0 0.5 PP0584 beta-alanine--pyruvate transaminase 8.0 1.2 PP0596 methylmalonate-semialdehyde dehydrogenase 6.8 0.9 PP0597 branched-chain amino acid ABC transporter, ATP- binding protein 0.7 0.7 PP0615 NADH dehydrogenase 1.4 1.7 PP0626 hypothetical protein 0.6 0.6 PP0647 amino acid ABC transporter, permease protein, putative 0.5 1.2 PP0656 hypothetical protein 1.0 1.5 PP0673 ABC transporter, ATP-binding protein 1.1 1.4 PP0674 glutamate dehydrogenase 2.8 0.6 PP0675 peptidyl-prolyl cis-trans isomerase, FKBP-type 1.2 1.8 PP0684 hypothetical protein 1.2 1.4 PP0685 phnA protein 1.0 1.4 PP0686 major facilitator family transporter 3.5 2.4 PP0702 transmembrane sensor, putative 1.8 1.5 PP0703 RNA polymerase sigma-70 factor, ECF subfamily 1.7 1.5 PP0704 peptidyl-tRNA hydrolase 1.0 1.4 PP0720 hypothetical protein 0.5 0.6 PP0728 phosphoenolpyruvate synthase-related protein 0.6 0.7 PP0729 hypothetical protein 0.4 0.4 PP0738 deoxyribodipyrimidine photolyase 0.4 0.2 PP0739 transcriptional regulator, MerR family 0.3 0.3 PP0740 hypothetical protein 0.5 0.4 PP0741 hypothetical protein 0.4 0.4 PP0742 hypothetical protein 0.5 0.4 PP0743 ferrochelatase 0.6 0.6 PP0744 hypothetical protein 0.6 0.7 PP0757 hypothetical protein 0.7 0.5 PP0764 hypothetical protein 0.9 0.6 PP0765 hypothetical protein 1.0 0.6 PP0766 hypothetical protein 0.5 0.6 PP0784 cyoups1 protein 0.6 0.8 PP0810 cyoups2 protein 0.5 0.6 PP0811 cytochrome o ubiquinol oxidase, subunit II 0.5 0.7 PP0812 cytochrome o ubiquinol oxidase, subunit I 0.5 0.8 PP0813 cytochrome o ubiquinol oxidase, subunit III 0.6 0.8 PP0814 cytochrome o ubiquinol oxidase, protein CyoD 0.6 0.8 PP0815 protoheme IX farnesyltransferase 0.6 0.8 PP0816 aminotransferase, class I 0.5 0.5 PP0817 phosphate ABC transporter, permease protein, putative 0.6 1.3 PP0826 hypothetical protein 0.5 0.7 PP0829 ornithine decarboxylase, putative 0.1 0.1 PP0864 glycine betaine carnitine choline ABC transporter, periplasmic binding protein, putative 1.4 1.1 PP0870 peptide chain release factor 3 1.2 1.5 PP0872 hypothetical protein 1.1 1.4 PP0874 sensor histidine kinase 0.5 0.5 PP0887 response regulator 0.5 0.4 PP0888 hypothetical protein 0.6 0.5 PP0889 hypothetical protein 0.6 0.6 PP0890 hypothetical protein 1.5 1.5 PP0908

135 Supplemental data

Fold Primary annotation (from KT2440 Fold change S12 KT2440 change database) (t=15 min) locus tag (t=30 min) hypothetical protein 1.5 1.4 PP0909 xenobiotic reductase B 1.2 1.5 PP0920 Transcriptional regulator, ArsR family 1.3 1.4 PP0921 glutamyl-tRNA(Gln) amidotransferase, A subunit 1.3 1.6 PP0931 glutamyl-tRNA(Gln) amidotransferase, C subunit 1.2 1.4 PP0932 sigma54 modulation protein 0.6 0.7 PP0951 hypothetical protein 0.6 0.7 PP0971 valyl-tRNA synthetase 1.1 1.5 PP0977 glycine cleavage system T protein 1.3 2.5 PP0986 L-serine dehydratase, iron-sulfur-dependent, single chain form 1.4 2.8 PP0987 glycine cleavage system P protein 1.1 2.5 PP0988 glycine cleavage system H protein 1.0 2.3 PP0989 hypothetical protein 0.3 0.4 PP0998 carbamate kinase 0.5 0.5 PP0999 ornithine carbamoyltransferase, catabolic 0.5 0.5 PP1000 arginine deiminase 0.6 0.6 PP1001 glyceraldehyde 3-phosphate dehydrogenase 1.2 0.4 PP1009 DNA-binding response regulator GltR 1.4 0.7 PP1012 sensor histidine kinase 1.5 0.5 PP1013 sugar ABC transporter, permease protein 2.2 0.3 PP1016 sugar ABC transporter, permease protein 2.0 0.3 PP1017 sugar ABC transporter, ATP-binding subunit 1.5 0.4 PP1018 porin B 1.5 0.5 PP1019 hypothetical protein 1.9 0.5 PP1020 transcriptional regulator HexR 2.4 0.5 PP1021 GMP synthase 0.9 1.4 PP1032 multicopper oxidase 1.5 1.2 PP1034 periplasmic binding domain transglycosylase SLT domain fusion protein 1.4 1.1 PP1036 phosphoribosylformylglycinamidine synthase 0.9 1.4 PP1037 hypothetical protein 1.0 1.8 PP1038 transcriptional regulator, PadR family 1.4 1.2 PP1057 amino acid ABC transporter, ATP-binding protein 1.3 0.4 PP1068 amino acid ABC transporter, permease protein 1.5 0.3 PP1069 amino acid ABC transporter, permease protein 1.4 0.3 PP1070 amino acid ABC transporter, periplasmic amino acid-binding protein 1.2 0.3 PP1071 conserved hypothetical protein TIGR00011 0.7 0.9 PP1077 ornithine carbamoyltransferase 1.9 1.3 PP1079 bacterioferritin 0.4 0.7 PP1082 antioxidant, AhpC Tsa family 1.1 1.5 PP1084 membrane protein, putative 0.6 0.8 PP1093 hypothetical protein 0.9 1.4 PP1106 OmpA family protein 0.7 0.5 PP1121 helicase, putative 1.3 1.2 PP1125 branched-chain amino acid ABC transporter, ATP- binding protein 1.2 0.6 PP1137 branched-chain amino acid ABC transporter, ATP- binding protein 1.3 0.5 PP1138 high-affinity branched-chain amino acid transport protein 1.3 0.5 PP1139 branched-chain amino acid ABC transporter, permease protein 1.4 0.6 PP1140

136 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) branched-chain amino acid ABC transporter, periplasmic amino acid-binding protein 1.1 0.5 PP1141 hypothetical protein 0.9 1.7 PP1149 lipoprotein, putative 0.6 1.1 PP1153 acetolactate synthase, catabolic, putative 0.5 0.9 PP1157 oxidoreductase, FAD-binding 1.3 1.4 PP1163 TRAP dicarboxylate transporter, DctP subunit 1.6 1.3 PP1169 ribonucleoside reductase, beta subunit 1.3 1.4 PP1177 hypothetical protein 1.1 1.4 PP1178 ribonucleoside reductase, alpha subunit 1.1 1.3 PP1179 outer membrane protein H1 0.8 1.8 PP1185 transcriptional regulatory protein PhoP 0.8 1.5 PP1186 sensor protein PhoQ 0.8 1.4 PP1187 C4-dicarboxylate transport protein 4.3 0.2 PP1188 exsB protein 0.9 1.4 PP1226 membrane protein, putative 9.9 3.4 PP1227 methyl-accepting chemotaxis transducer 27.8 10.3 PP1228 amino acid transporter, putative 56.3 14.5 PP1229 hypothetical protein 1.7 1.4 PP1230 hypothetical protein 1.1 1.3 PP1247 malate:quinone oxidoreductase 0.7 4.4 PP1251 PhoH family protein 0.6 0.8 PP1291 general amino acid ABC transporter, periplasmic binding protein 1.8 0.9 PP1297 general amino acid ABC transporter, permease protein 2.3 1.4 PP1298 general amino acid ABC transporter, permease protein 1.9 1.1 PP1299 general amino acid ABC transporter, ATP-binding protein 1.9 0.9 PP1300 transcriptional regulator, AsnC family 1.4 1.0 PP1307 hypothetical protein 0.6 0.7 PP1310 transcriptional regulator, AraC family 1.5 1.1 PP1313 oxidoreductase, aldo keto reductase family 0.9 0.6 PP1314 hypothetical protein 1.2 1.6 PP1353 multidrug efflux MFS transporter, putative 1.1 1.5 PP1354 muropeptide permease AmpG 0.8 0.6 PP1355 hypothetical protein 0.7 0.6 PP1372 3-carboxy-cis,cis-muconate cycloisomerase 0.7 0.7 PP1379 BenF-like porin 1.4 1.3 PP1383 multidrug solvent RND outer membrane protein TtgC 1.1 1.4 PP1384 multidrug solvent RND transporter TtgB 1.1 1.3 PP1385 multidrug solvent RND membrane fusion protein 1.0 1.3 PP1386 hypothetical protein 0.6 1.2 PP1390 dicarboxylate MFS transporter 1.0 1.5 PP1400 acyl-transferase 0.4 0.8 PP1408 hypothetical protein 3.3 0.5 PP1415 tricarboxylate transport protein TctA, putative 4.0 0.4 PP1416 tricarboxylate transport protein TctB, putative 4.1 0.2 PP1417 tricarboxylate transport protein TctC, putative 3.3 0.3 PP1418 porin, putative 2.6 0.2 PP1419 L-aspartate oxidase 0.9 0.6 PP1426

137 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) glucose dehydrogenase (pyrroloquinoline- quinone) 1.4 1.6 PP1444 metabolite-proton symporter 1.5 1.2 PP1458 hypothetical protein 1.0 1.3 PP1503 phosphoenolpyruvate carboxylase 0.7 0.6 PP1505 adenylate kinase 0.5 0.3 PP1506 cold shock protein CspA 1.1 0.4 PP1522 hypothetical protein 1.4 1.3 PP1548 hypothetical protein 1.5 1.2 PP1568 hypothetical protein 1.4 1.2 PP1574 phage tail tape meausure protein, lambda family 1.4 1.2 PP1577 translation elongation factor Ts 1.2 1.4 PP1592 2-dehydro-3-deoxyphosphooctonate aldolase 1.0 1.4 PP1611 enolase 1.0 1.5 PP1612 D-isomer specific 2-hydroxyacid dehydrogenase family protein 1.4 1.9 PP1616 esterase, putative 1.3 2.0 PP1617 RNA polymerase sigma factor RpoS 0.7 0.5 PP1623 hypothetical protein 1.0 2.0 PP1627 CinA domain protein 1.0 1.4 PP1628 recA protein 1.4 1.4 PP1629 recX protein 1.3 1.4 PP1630 ISPpu10, transposase 1.5 1.1 PP1653 nicotinate-nucleotide-- dimethylbenzimidazolephosphoribosyltransferase 1.1 1.4 PP1679 cobalamin (5 -phosphate) synthase 0.9 1.4 PP1681 hypothetical protein 1.0 1.4 PP1693 major facilitator family transporter 1.4 1.1 PP1698 acyltransferase 0.6 0.6 PP1700 sodium:solute symporter family protein 1.2 0.5 PP1743 hypothetical protein 0.8 0.7 PP1762 prephenate dehydrogenase, putative 3- phosphoshikimate 1-carboxyvinyltransferase 1.0 1.4 PP1770 integration host factor, beta subunit 0.5 0.7 PP1773 hypothetical protein 0.7 0.6 PP1774 mannosyltransferase, putative 1.2 1.6 PP1780 hypothetical protein 1.4 1.3 PP1833 TonB-dependent receptor 1.2 1.7 PP1847 membrane protein, putative 1.0 1.5 PP1850 organic hydroperoxide resistance protein 0.7 0.7 PP1859 heat shock protein HtpX 0.6 0.8 PP1871 ribonuclease E 1.5 1.1 PP1905 molybdenum cofactor biosynthesis protein A, putative 1.4 1.3 PP1969 lipoprotein, putative 0.8 1.5 PP1970 hypothetical protein 0.8 0.7 PP1973 glutamyl-tRNA synthetase 1.1 1.5 PP1977 3-isopropylmalate dehydratase, large subunit 0.5 0.5 PP1985 3-isopropylmalate dehydratase, small subunit 0.5 0.5 PP1986 methlytransferase, UbiE COQ5 family 0.5 0.4 PP1987 3-isopropylmalate dehydrogenase 0.6 0.5 PP1988 aspartate-semialdehyde dehydrogenase 0.7 0.6 PP1989 semialdehyde dehydrogenase family protein 0.9 1.4 PP1992 amidophosphoribosyltransferase 1.1 1.4 PP2000

138 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) O-succinylhomoserine sulfhydrylase 1.2 1.4 PP2001 oxidoreductase, short chain dehydrogenase reductase family 1.1 1.4 PP2002 membrane protein, putative 2.1 0.8 PP2019 transcriptional regulator, AraC family 1.1 1.4 PP2072 hypothetical protein 0.5 0.8 PP2080 cmpX protein 0.6 0.6 PP2087 hypothetical protein 1.1 1.7 PP2096 hypothetical protein 1.2 1.4 PP2099 hypothetical protein 1.9 2.6 PP2115 hypothetical protein 1.2 1.4 PP2116 ABC transporter, ATP-binding protein 1.4 1.6 PP2131 hypothetical protein 1.4 1.6 PP2133 glyceraldehyde 3-phosphate dehydrogenase 1.0 1.4 PP2149 copper-binding protein, plastocyanin azurin family 0.5 0.7 PP2159 hypothetical protein 1.0 1.3 PP2200 copper resistance protein B, putative 0.5 0.7 PP2204 peptidase, U32 family 1.2 2.2 PP2206 acyl-CoA dehydrogenase family protein 1.1 0.5 PP2216 hypothetical protein 0.6 1.4 PP2221 hypothetical protein 0.6 0.7 PP2235 peptidase, M24 family protein 0.6 0.7 PP2238 aerotaxis receptor Aer-1 1.5 1.4 PP2257 sensory box protein 1.4 1.3 PP2258 sigma-54 dependent transcriptional regulator 1.5 1.3 PP2259 head-to-tail joining protein 1.5 1.1 PP2279 hypothetical protein 1.7 1.0 PP2296 hydrolase, TatD family 0.9 1.4 PP2311 hypothetical protein 1.0 1.5 PP2321 carboxyvinyl-carboxyphosphonate phosphorylmutase, putative 1.0 0.6 PP2334 methylcitrate synthase, putative 1.3 0.5 PP2335 type 1 pili usher protein CsuD 1.4 1.4 PP2362 hypothetical protein 0.9 0.5 PP2364 hypothetical protein 1.2 1.5 PP2374 5-methyltetrahydrofolate--homocysteine methyltransferase 1.2 1.5 PP2375 esterified fatty acid cis trans isomerase 0.6 0.6 PP2376 Sco1 SenC family protein 0.4 0.4 PP2379 hypothetical protein 0.3 0.4 PP2380 major facilitator family transporter 1.0 1.7 PP2392 cobalt-zinc-cadmium resistance protein CzcB, putative 1.3 1.4 PP2409 acetyltransferase, GNAT family 0.3 0.2 PP2421 carboxymuconolactone decarboxylase family protein 0.3 0.1 PP2422 alkyl hydroperoxide reductase, C subunit 0.6 0.5 PP2439 alkyl hydroperoxide reductase, F subunit 0.4 0.4 PP2440 hypothetical protein 0.6 0.7 PP2446 hypothetical protein 0.6 0.8 PP2447 hypothetical protein 21.6 19.5 PP2448 L-asparaginase II 1.7 0.7 PP2453

139 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) ribose ABC transporter, periplasmic ribose- binding protein 2.0 0.3 PP2454 ribose ABC transporter, ATP-binding protein 2.3 0.4 PP2455 ribose ABC transporter, permease protein 2.9 0.4 PP2456 ribose operon repressor 2.0 0.4 PP2457 ribokinase 1.5 0.3 PP2458 ribose ABC transporter, periplasmic ribose- binding protein 1.5 0.3 PP2459 inosine-uridine preferring nucleoside hydrolase 1.1 0.5 PP2460 phenylalanyl-tRNA synthetase, alpha subunit 1.2 1.4 PP2469 phenylalanyl-tRNA synthetase, beta subunit 1.3 1.5 PP2470 hypothetical protein 0.6 0.5 PP2511 GTP cyclohydrolase I 0.7 0.7 PP2512 protocatechuate 4,5-dioxygenase, putative 1.4 1.1 PP2518 gamma-aminobutyrate transporter, putative 48.7 89.1 PP2543 hypothetical protein 1.0 1.4 PP2562 hypothetical protein 1.1 1.4 PP2622 hypothetical protein 2.0 2.2 PP2629 hypothetical protein 1.5 1.6 PP2630 hypothetical protein 1.6 1.6 PP2631 hypothetical protein 1.2 1.5 PP2632 hypothetical protein 1.8 2.0 PP2633 cellulose synthase, putative 1.3 1.6 PP2634 cellulose synthase, putative 1.1 1.7 PP2635 cellulose synthase, putative 1.3 2.0 PP2636 endo-1,4-beta-D-glucanase 2.0 2.8 PP2637 cellulose synthase operon C protein, putative 1.6 2.8 PP2638 universal stress protein family 0.6 1.0 PP2648 hypothetical protein 0.1 0.1 PP2662 hypothetical protein 0.1 0.0 PP2663 sensory box histidine kinase response regulator 0.1 0.0 PP2664 DNA-binding response regulator AgmR 0.3 0.0 PP2665 ABC efflux transporter, permease protein 0.3 0.0 PP2667 ABC efflux transporter, ATP-binding protein 0.3 0.1 PP2668 outer membrane protein, putative 0.2 0.0 PP2669 hypothetical protein 0.2 0.0 PP2670 sensor histidine kinase 0.3 0.1 PP2671 DNA-binding response regulator, LuxR family 0.4 0.0 PP2672 pentapeptide repeat family protein 0.7 0.0 PP2673 quinoprotein ethanol dehydrogenase 0.8 0.0 PP2674 cytochrome c-type protein 0.3 0.0 PP2675 periplasmic binding protein, putative 0.3 0.0 PP2676 hypothetical protein 0.2 0.0 PP2677 hydrolase, putative 0.3 0.0 PP2678 quinoprotein ethanol dehydrogenase, putative 0.4 0.0 PP2679 aldehyde dehydrogenase family protein 0.4 0.0 PP2680 coenzyme PQQ synthesis protein D, putative 0.3 0.0 PP2681 alcohol dehydrogenase, iron-containing 0.4 0.0 PP2682 sensory box histidine kinase response regulator 0.8 0.2 PP2683 transcriptional regulator, LysR family 1.4 1.1 PP2695 flavin reductase domain protein 0.8 1.4 PP2697 hypothetical protein 0.5 1.3 PP2699 hypothetical protein 0.8 0.6 PP2705 hypothetical protein 1.2 1.4 PP2710

140 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) hypothetical protein 1.5 1.1 PP2722 cyclopropane-fatty-acyl-phospholipid synthase 0.5 0.5 PP2734 hypothetical protein 0.7 0.7 PP2735 hypothetical protein 0.5 0.4 PP2736 acyl carrier protein, putative 1.7 2.7 PP2777 3-oxoacyl-(acyl-carrier-protein) synthase II 1.7 2.4 PP2778 beta-ketoacyl synthase, putative 2.0 2.9 PP2779 3-oxoacyl-(acyl-carrier-protein) synthase II 1.8 2.8 PP2780 beta-ketoacyl synthase 2.0 3.1 PP2781 pyridoxalphosphate dependent aminotransferase, class III 2.4 3.8 PP2782 3-oxoacyl-(acyl-carrier-protein) reductase 2.3 3.8 PP2783 oxidoreductase, short-chain dehydrogenase reductase family 2.6 4.1 PP2784 hypothetical protein 2.3 3.5 PP2785 hypothetical protein 2.6 2.9 PP2786 transporter, putative 2.7 3.0 PP2787 transcriptional regulator, MerR family 1.3 1.5 PP2788 oxidoreductase, putative 1.1 1.7 PP2789 sigma-54 dependent sensory box protein 1.1 1.4 PP2790 oxidoreductase, putative 1.3 1.3 PP2798 hypothetical protein 0.9 1.3 PP2828 transcriptional regulator, LysR family 1.0 1.4 PP2833 urease accessory protein UreJ 1.1 1.3 PP2847 hypothetical protein 1.1 1.5 PP2873 transporter, bile acid Na+ symporter family 1.1 1.4 PP2877 hypothetical protein 1.0 1.4 PP2883 hypothetical protein 0.8 1.7 PP2928 carboxynorspermidine decarboxylase 0.9 1.6 PP2929 L-serine dehydratase, putative 2.6 4.1 PP2930 conserved hypothetical protein TIGR00148 2.6 3.4 PP2931 amidase family protein 12.5 18.5 PP2932 hypothetical protein 2.7 2.6 PP2938 membrane protein, putative 1.5 1.3 PP2968 transposase family protein 1.3 1.2 PP2975 oxidoreductase, FMN-binding 1.3 1.3 PP2994 DNA topology modulation kinase FlaR, putative 1.0 1.8 PP2995 hypothetical protein 0.6 0.6 PP3004 DNA-binding protein Roi-related protein 1.4 1.3 PP3032 ClpP protease, putative 1.4 1.3 PP3045 membrane protein, putative 0.6 0.6 PP3082 hypothetical protein 1.3 1.1 PP3089 hypothetical protein 1.3 1.2 PP3100 oxidoreductase, aldo keto reductase family 1.1 0.6 PP3120 CoA-transferase, subunit A, putative 1.3 1.4 PP3122 hypothetical protein 0.6 0.6 PP3134 serine O-acetyltransferase, putative 0.6 0.9 PP3136 transcriptional regulator, LysR family 1.2 1.5 PP3143 hypothetical protein 2.6 1.2 PP3145 oxidoreductase, putative 4.0 1.3 PP3146 periplasmic polyamine-binding protein, putative 8.4 0.8 PP3147 glutamine synthetase, putative 11.6 1.1 PP3148 outer membrane ferric siderophore receptor, putative 2.2 1.7 PP3155

141 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) benABC operon transcriptional activator BenR 0.6 0.3 PP3159 major facilitator family transporter 1.6 1.2 PP3176 Smp-30 Cgr1 family protein 1.2 1.6 PP3180 ring-oxidation complex protein 1 1.4 1.2 PP3278 glutathione-regulated potassium-proton antiporter 1.7 1.4 PP3311 outer membrane ferric siderophore receptor, putative 1.3 1.1 PP3325 TonB-dependent receptor, putative 1.6 1.6 PP3340 nickel ABC transporter, ATP-binding protein 1.1 1.5 PP3345 feruloyl-CoA-synthetase 2.3 1.0 PP3356 vanillin dehydrogenase 1.9 1.0 PP3357 enoyl-CoA hydratase aldolase 1.5 1.2 PP3358 bacterial surface antigen family protein 3.1 0.3 PP3373 hypothetical protein 2.5 0.4 PP3374 endonuclease I 2.6 0.4 PP3375 2-ketogluconate 6-phosphate reductase 1.4 0.4 PP3376 2-ketogluconate transporter, putative 3.1 0.3 PP3377 2-ketogluconate kinase 2.4 0.5 PP3378 epimerase KguE, putative 2.1 0.4 PP3379 transcriptional regulator PtxS 1.6 0.5 PP3380 cobalamin biosynthesis protein cobE, putative 0.7 0.6 PP3409 methyl-accepting chemotaxis transducer sensory box protein 0.8 0.6 PP3414 transcriptional regulator, LacI family 0.7 0.6 PP3415 gluconokinase 1.4 0.9 PP3416 gluconate transporter 1.4 0.7 PP3417 hypothetical protein 1.8 0.8 PP3418 ThiJ PfpI family protein 0.8 0.6 PP3431 glyceraldehyde-3-phosphate dehydrogenase, putative 0.9 0.5 PP3443 hypothetical protein 0.5 0.5 PP3445 threonine dehydratase, biosynthetic 0.2 0.2 PP3446 transcriptional regulatory protein RstA, putative 1.5 1.4 PP3454 multidrug efflux RND membrane fusion protein 1.7 2.6 PP3455 multidrug efflux RND transporter 1.6 2.4 PP3456 hypothetical protein 1.4 1.2 PP3461 hypothetical protein 1.3 1.2 PP3469 hypothetical protein 0.3 0.4 PP3504 hypothetical protein 0.5 0.8 PP3505 AMP-binding domain protein 2.0 1.0 PP3553 Na+ H+ antiporter family protein 1.0 1.7 PP3556 serine transporter SdaC 0.9 2.0 PP3589 aromatic-amino-acid aminotransferase 1.1 1.6 PP3590 malate dehydrogenase, putative 1.3 0.5 PP3591 amino acid ABC transporter, periplasmic amino acid-binding protein 1.4 0.9 PP3593 D-amino acid dehydrogenase, small subunit family protein 1.1 0.6 PP3596 hypothetical protein 71.4 51.0 PP3598 D-glucarate permease 1.6 1.1 PP3600 hypothetical protein 0.7 0.4 PP3611 hypothetical protein 1.1 1.4 PP3620 N-acetyl-gamma-glutamyl-phosphate reductase 0.6 0.7 PP3633

142 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) transcriptional regulator, AraC family 0.5 0.7 PP3659 decarboxylase family protein 0.8 0.6 PP3662 hypothetical protein 1.4 1.0 PP3697 glycolate oxidase, subunit GlcD 1.7 1.1 PP3745 glycolate oxidase, subunit GlcE 1.4 1.0 PP3746 response regulator 0.7 0.7 PP3757 hydrolase, isochorismatase family 0.7 0.6 PP3806 polyamine ABC transporter, periplasmic polyamine-binding protein 1.4 1.1 PP3814 cytochrome c family protein 0.8 0.3 PP3822 molybdate ABC transporter, periplasmic molybdate-binding protein 1.5 1.1 PP3828 alcohol dehydrogenase, zinc-containing 0.6 1.0 PP3839 D-aminopeptidase putative 1.4 1.2 PP3844 hypothetical protein 1.5 1.2 PP3903 hypothetical protein 1.1 1.3 PP3913 hypothetical protein 1.5 1.3 PP3916 phage integrase, putative 1.4 1.1 PP3920 hypothetical protein 1.4 1.1 PP3988 hypothetical protein 1.1 1.6 PP3991 xanthine uracil permease family protein 0.9 1.4 PP3992 isocitrate dehydrogenase, NADP-dependent, prokaryotic-type 0.8 0.7 PP4011 isocitrate dehydrogenase, NADP-dependent, monomeric-type 1.2 2.2 PP4012 hypothetical protein 1.1 1.4 PP4015 adenylosuccinate lyase 1.0 1.5 PP4016 hypothetical protein 1.0 1.4 PP4017 acetyltransferase, GNAT family 1.1 1.5 PP4018 DNA topoisomerase III 1.3 1.1 PP4019 hypothetical protein 1.5 1.1 PP4028 hypothetical protein 1.4 1.1 PP4028 hypothetical protein 0.5 0.5 PP4054 1,4-alpha-glucan branching enzyme 0.8 0.6 PP4058 long-chain-fatty-acid-CoA ligase, putative 1.0 0.6 PP4063 hypothetical protein 1.4 1.1 PP4087 isocitrate lyase 0.3 0.0 PP4116 NADH dehydrogenase I, A subunit 1.2 1.5 PP4119 NADH dehydrogenase I, B subunit 1.6 1.9 PP4120 NADH dehydrogenase I, C,D subunit 1.7 2.1 PP4121 NADH dehydrogenase I, E subunit 1.8 2.1 PP4122 NADH dehydrogenase I, F subunit 1.8 2.4 PP4123 NADH dehydrogenase I, G subunit 2.1 2.9 PP4124 NADH dehydrogenase I, H subunit 2.0 2.9 PP4125 NADH dehydrogenase I, I subunit 2.1 3.1 PP4126 NADH dehydrogenase I, J subunit 1.9 2.8 PP4127 NADH dehydrogenase I, K subunit 1.8 2.8 PP4128 NADH dehydrogenase I, L subunit 2.1 3.0 PP4129 NADH dehydrogenase I, M subunit 1.8 2.7 PP4130 NADH dehydrogenase I, N subunit 1.6 2.7 PP4131 peptide ABC transporter, ATP-binding protein 0.7 0.7 PP4150 ThrB family protein aminotransferase, class III 2.0 1.2 PP4154 succinyl-CoA synthetase, alpha subunit 1.4 1.6 PP4185 succinyl-CoA synthetase, beta subunit 1.2 1.4 PP4186

143 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) succinate dehydrogenase, flavoprotein subunit 1.3 1.3 PP4191 succinate dehydrogenase, hydrophobic membrane anchor protein 1.2 1.4 PP4192 transcriptional regulator, GntR family 0.9 0.6 PP4197 siderophore biosynthesis protein, putative 0.6 0.7 PP4245 cytochrome c oxidase, cbb3-type, subunit I 0.2 0.4 PP4250 cytochrome c oxidase, cbb3-type, subunit II 0.3 0.4 PP4251 cytochrome c oxidase, cbb3-type, CcoQ subunit 0.2 0.4 PP4252 cytochrome c oxidase, cbb3-type, subunit III 0.2 0.3 PP4253 hypothetical protein 0.9 0.5 PP4254 cytochrome c oxidase, cbb3-type, subunit I 0.6 0.1 PP4255 cytochrome c oxidase, cbb3-type, subunit II 0.6 0.1 PP4256 cytochrome c oxidase, cbb3-type, CcoQ subunit 0.8 0.2 PP4257 cytochrome c oxidase, cbb3-type, subunit III 0.7 0.2 PP4258 iron-sulfur cluster-binding protein 0.6 0.4 PP4259 hypothetical protein 0.7 0.6 PP4260 cation-transporting P-type ATPase 0.6 0.5 PP4261 oxygen-independent coproporphyrinogen III oxidase 1.1 1.8 PP4264 hypothetical protein 1.2 1.5 PP4292 hypothetical protein 1.2 1.7 PP4293 phenazine biosynthesis protein, PhzF family 1.3 1.6 PP4315 hypothetical protein 0.7 0.7 PP4317 hypothetical protein 0.5 0.4 PP4319 thiol:disulfide interchange protein DsbE 0.7 0.5 PP4321 cytochrome c-type biogenesis protein CcmF 0.8 0.5 PP4322 cytochrome c-type biogenesis protein CcmE 0.8 0.6 PP4323 cytochrome c-type biogenesis protein CcmD 0.5 0.5 PP4324 flagellar motor protein MotB 0.7 0.9 PP4335 protein-glutamate methylesterase CheB 0.7 0.8 PP4337 flagellar biosynthetic protein FliR 0.9 1.4 PP4353 flagellar assembly protein FliO 0.9 1.4 PP4356 flagellar motor switch protein FliN 1.0 1.4 PP4357 flagellar protein FliL 0.9 1.3 PP4359 hypothetical protein 0.5 0.8 PP4360 anti-sigma F factor antagonist, putative 0.5 0.5 PP4364 flagellar motor switch protein FliG 0.9 1.5 PP4368 response regulator FleR 0.8 1.5 PP4371 flagellar P-ring protein precursor FlgI 1.0 1.4 PP4383 flagellar L-ring protein precursor FlgH 0.8 1.4 PP4384 flagellar basal-body rod protein FlgG 0.8 1.5 PP4385 flagellar basal-body rod protein FlgF 0.7 1.4 PP4386 flagella basal body P-ring formation protein FlgA 0.9 1.4 PP4394 glutamine synthetase, putative 2.9 1.3 PP4399 succinylarginine dihydrolase 0.5 0.9 PP4477 succinylglutamic semialdehyde dehydrogenase 0.5 0.9 PP4478 arginine N-succinyltransferase, beta subunit 0.6 0.9 PP4479 arginine N-succinyltransferase, alpha subunit 0.7 1.0 PP4480 transcriptional regulator, AraC family 0.6 1.2 PP4482 basic amino acid ABC transporter, permease protein 0.6 1.1 PP4484 basic amino acid ABC transporter, permease protein 0.6 1.1 PP4485 acetyl-coA synthetase 1.5 0.1 PP4487

144 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) glutamine synthetase, putative 3.3 1.3 PP4547 oxidoreductase, putative 3.7 1.2 PP4548 long-chain-fatty-acid-CoA ligase 1.0 0.7 PP4549 hydrolase, alpha beta fold family 0.6 0.8 PP4551 hypothetical protein 0.6 0.7 PP4557 transcriptional regulator, LysR family 0.8 0.7 PP4601 FecR protein, putative 1.1 1.7 PP4612 outer membrane iron(III) dicitrate receptor 1.5 1.1 PP4613 hypothetical protein 0.4 0.6 PP4615 maleylacetoacetate isomerase, putative 0.9 0.5 PP4619 hypothetical protein 0.6 0.9 PP4635 5-methyltetrahydropteroyltriglutamate- homocysteine S-methyltransferase family protein 0.6 1.3 PP4637 hypothetical protein 1.5 0.6 PP4640 carbon starvation protein CstA 1.4 0.5 PP4641 large conductance mechanosensitive channel protein 0.6 0.7 PP4645 hypothetical protein 0.6 0.5 PP4649 ubiquinol oxidase subunit II, cyanide insensitive 0.2 0.2 PP4650 ubiquinol oxidase subunit I, cyanide insensitive 0.6 0.7 PP4651 gamma-glutamyltransferase 2.8 0.7 PP4659 osmY-related protein 1.7 2.4 PP4707 ribosome-binding factor A 1.0 1.4 PP4711 conserved hypothetical protein TIGR00253 1.3 1.5 PP4720 hypothetical protein 1.2 1.5 PP4721 transcription elongation factor GreA 1.3 1.4 PP4722 carbamoyl-phosphate synthase, large subunit 1.2 1.5 PP4723 carbamoyl-phosphate synthase, small subunit 1.3 1.8 PP4724 L-lactate transporter 4.4 0.2 PP4735 L-lactate dehydrogenase 0.7 0.2 PP4736 D-lactate dehydrogenase, putative 0.5 0.3 PP4737 amino acid ABC transporter, periplasmic amino acid-binding protein 1.4 1.1 PP4748 N-methylproline demethylase, putative 1.3 1.1 PP4753 acyl-CoA dehydrogenase, putative 0.6 0.5 PP4780 PhoH family protein 0.6 0.7 PP4787 conserved hypothetical protein TIGR00043 0.5 0.7 PP4788 metal ion transporter, putative 0.4 0.7 PP4789 apolipoprotein N-acyltransferase 0.7 0.8 PP4790 membrane-bound lytic murein transglycosylase, putative 0.7 0.7 PP4798 membrane protein, putative 0.8 0.7 PP4815 phosphoribosylaminoimidazolecarboxamide formyltransferase IMP cyclohydrolase 0.8 1.4 PP4822 phosphoribosylamine--glycine ligase 0.8 1.4 PP4823 dnaK protein, putative 0.8 1.6 PP4849 branched-chain amino acid ABC transporter, periplasmic amino acid-binding protein 1.5 0.7 PP4867 azurin 0.5 0.2 PP4870 iron ABC transporter, periplasmic iron-binding protein, putative 1.5 1.4 PP4881 iron ABC transporter, permease protein 1.5 1.6 PP4882 transporter, NCS1 nucleoside transporter family 1.6 1.3 PP4921 thiamin biosynthesis protein ThiC 1.6 1.2 PP4922

145 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) multidrug efflux SMR transporter 1.0 1.4 PP4930 hypothetical protein 1.0 1.3 PP4945 sodium proline symporter 1.4 1.1 PP4946 sensory box protein response regulator 0.8 0.5 PP4959 phosphoglycerate kinase 1.2 1.5 PP4963 D-erythrose-4-phosphate dehydrogenase 1.0 1.4 PP4964 transketolase 1.0 1.4 PP4965 transcriptional regulator, ArsR family 1.2 1.5 PP4966 S-adenosylmethionine synthetase 1.2 1.7 PP4967 Na+ H+ antiporter, putative 0.2 0.5 PP4974 long-chain acyl-CoA thioester hydrolase family protein 0.5 0.7 PP4975 5,10-methylenetetrahydrofolate reductase 1.0 1.8 PP4977 channel protein, hemolysin III family 0.5 0.7 PP4986 type IV pili methyl-accepting chemotaxis transducer PilJ 0.6 0.9 PP4989 GTP-binding protein TypA BipA 1.2 1.4 PP5044 RNA methyltransferase, TrmH family, group 2 1.5 1.7 PP5052 2,3-biphosphoglycerate-independent phosphoglycerate mutase 1.0 1.5 PP5056 glutamate synthase, small subunit 1.9 0.5 PP5075 glutamate synthase, large subunit 1.6 0.5 PP5076 hypothetical protein 1.0 1.4 PP5127 lipoprotein, putative 0.8 1.4 PP5165 sigma-54 dependent transcriptional regulator 1.1 1.4 PP5166 hypothetical protein 1.0 2.0 PP5167 sulfate ABC transporter, ATP-binding protein 0.6 0.9 PP5168 sulfate ABC transporter, permease protein 0.5 1.1 PP5169 sulfate ABC transporter, permease protein 0.5 1.4 PP5170 sulfate ABC transporter, periplasmic sulfate- binding protein 0.8 1.6 PP5171 hypothetical protein 0.9 2.1 PP5172 hypothetical protein 10.3 1.6 PP5176 putrescine ABC transporter, permease protein 8.2 2.1 PP5177 putrescine ABC transporter, permease protein 6.7 2.1 PP5178 putrescine ABC transporter, ATP-binding protein 7.0 2.0 PP5179 putrescine ABC transporter, periplasmic putrescine-binding protein 8.3 2.1 PP5180 putrescine ABC transporter, periplasmic putrescine-binding protein 5.6 2.1 PP5181 aminotransferase, class III 19.5 5.7 PP5182 glutamine synthetase, putative 7.2 2.2 PP5183 glutamine synthetase, putative 10.1 5.1 PP5184 N-acetylglutamate synthase 0.5 0.8 PP5185 glycine cleavage system T protein 1.4 1.2 PP5194 ABC transporter, ATP-binding protein permease protein, putative 0.9 1.4 PP5207 ChaC-related protein 0.7 0.8 PP5211 oxidoreductase, iron-sulfur-binding 0.6 0.9 PP5212 hypothetical protein 0.6 0.8 PP5219 DNA-binding response regulator, LuxR family 0.4 0.3 PP5241 xanthine phosphoribosyltransferase 0.9 1.3 PP5265 acetyl-CoA hydrolase transferase family protein 1.0 1.5 PP5266 transcriptional regulator, Cro CI family 1.1 1.6 PP5268

146 Supplemental data

Primary annotation (from KT2440 Fold change Fold S12 KT2440 database) (t=15 min) change locus tag (t=30 min) alanine racemase, catabolic 0.9 1.4 PP5269 D-amino acid dehydrogenase, small subunit 1.4 1.4 PP5270 leucine-responsive regulatory protein 1.6 1.4 PP5271 hypothetical protein 1.3 1.4 PP5272 phospholipase D family protein 1.3 1.5 PP5276 major facilitator family transporter 39.0 26.1 PP5277 aldehyde dehydrogenase family protein 18.9 14.7 PP5278 hypothetical protein 12.3 14.9 PP5279 phosphopantothenoylcysteine decarboxylase phosphopantothenate--cysteine ligase 1.3 1.4 PP5285 deoxyuridine 5-triphosphate nucleotidohydrolase 1.3 1.7 PP5286 acetylglutamate kinase 0.6 0.9 PP5289 catabolite repression control protein 1.3 1.6 PP5292 hypothetical protein 1.1 1.5 PP5293 amino acid transporter, putative 21.8 12.8 PP5297 hypothetical protein 24.0 15.3 PP5298 glutamine synthetase, putative 50.3 25.4 PP5299 transcriptional regulator, LysR family 1.4 1.5 PP5300 ferric siderophore transport system protein ExbB 1.5 1.5 PP5306 ferric siderophore transport system, inner membrane protein ExbD 1.8 1.8 PP5307 ferric siderophore transport system, periplasmic binding protein TonB 1.7 1.7 PP5308 DNA-binding protein HU-alpha 2.1 0.5 PP5313 long-chain acyl-CoA thioester hydrolase family protein 0.6 0.6 PP5331 hypothetical protein 0.7 0.6 PP5332 hypothetical protein 0.9 1.6 PP5334 phosphoribosylaminoimidazole carboxylase, ATPase subunit 0.9 1.4 PP5335 phosphoribosylaminoimidazole carboxylase, catalytic subunit 0.9 1.5 PP5336 aspartate ammonia-lyase 0.8 1.6 PP5338 acetyltransferase, GNAT family 0.8 1.7 PP5344 oxaloacetate decarboxylase, alpha subunit 1.0 0.7 PP5346 acetyl-CoA carboxylase, biotin carboxylase 0.9 0.6 PP5347 transcriptional regulator, LysR family 1.2 1.8 PP5348 Rieske 2Fe-2S family protein 1.4 1.1 PP5373 hypothetical protein 1.1 1.4 PP5381 hypothetical protein 0.6 0.9 PP5389 glucosamine--fructose-6-phosphate aminotransferase, isomerizing 1.2 1.6 PP5409 UDP-N-acetylglucosamine pyrophosphorylase 1.0 1.6 PP5411

147 Supplemental data

Supplemental data for Chapter 4

Table S1: Proteins identified via mass spectrometry from the isolated protein band of the zymogram. Protein Protein match (No. Protein accession Protein description mass of peptides (Da) matched) 1 gi|206974153 peroxiredoxin [Bacillus cereus H3081.97] 20864 571

2 gi|229015725 Alkyl hydroperoxide reductase subunit C 20878 220 [Bacillus cereus AH1273] 3 gi|30264444 5~-methylthioadenosine/S- 25410 80 adenosylhomocysteine nucleosidase [Bacillus anthracis str. Ames] 4 gi|30021641 cold shock protein [Bacillus cereus ATCC 7362 59 14579] 5 gi|49480324 phosphopentomutase [Bacillus 44142 56 thuringiensis serovar konkukian str. 97- 27] 6 gi|42779297 aldo/keto reductase family 31852 48 oxidoreductase [Bacillus cereus ATCC 10987] 7 gi|206976195 NADH-dependent flavin oxidoreductase, 41563 44 Oye family [Bacillus cereus H3081.97] 8 gi|217961965 trigger factor [Bacillus cereus AH187] 47185 38 9 gi|30018378 elongation factor Tu [Bacillus cereus ATCC 43026 31 14579] 10 gi|52142140 aldo/keto reductase family 31859 27 oxidoreductase [Bacillus cereus E33L] 11 gi|206975479 oxidoreductase, aldo/keto reductase 31836 26 family [Bacillus cereus H3081.97] 12 gi|52145111 elongation factor G [Bacillus cereus E33L] 76533 25 13 gi|229032524 Cold shock protein cspB [Bacillus cereus 7233 25 AH1271] 14 gi|47564809 putative transcriptional regulator [Bacillus 13556 24 cereus G9241] 15 gi|30265108 aldo/keto reductase family 31860 24 oxidoreductase [Bacillus anthracis str. Ames] 16 gi|30262795 putative manganese-dependent inorganic 33955 21 pyrophosphatase [Bacillus anthracis str. Ames] 17 gi|30260978 quinone family NAD(P)H dehydrogenase 21225 18 [Bacillus anthracis str. Ames] 18 gi|52140575 NAD(P)H dehydrogenase (quinone) 21696 17 [Bacillus cereus E33L] 19 gi|42783441 molecular chaperone DnaK [Bacillus 65892 16 cereus ATCC 10987] 20 gi|42784284 phosphopyruvate hydratase [Bacillus 46428 15 cereus ATCC 10987] 21 gi|42779212 adenylate kinase [Bacillus cereus ATCC 24039 15 10987] 22 gi|206974887 ribosomal protein S1 [Bacillus cereus 42100 14 H3081.97]

148 Supplemental data

Protein Protein match (No. Protein accession Protein description mass of peptides (Da) matched) 23 gi|30018319 regulatory protein SpoVG [Bacillus cereus 10928 14 ATCC 14579] 24 gi|206976545 ferrochelatase [Bacillus cereus H3081.97] 35353 14 25 gi|206975063 metallo-beta-lactamase family protein 36562 13 [Bacillus cereus H3081.97] 26 gi|42784343 cold shock protein CspC [Bacillus cereus 7560 13 ATCC 10987] 27 gi|206974481 conserved hypothetical protein [Bacillus 47941 12 cereus H3081.97] 28 gi|30023337 F0F1 ATP synthase subunit beta [Bacillus 50990 12 cereus ATCC 14579] 29 gi|30265417 HAD superfamily hydrolase [Bacillus 30522 12 anthracis str. Ames] 30 gi|30260567 tellurium resistance protein [Bacillus 21202 11 anthracis str. Ames] 31 gi|206974918 histidinol-phosphate aminotransferase 41828 11 [Bacillus cereus H3081.97] 32 gi|30023042 Acyl-CoA dehydrogenase [Bacillus cereus 66507 11 ATCC 14579] 33 gi|206976024 porphobilinogen deaminase [Bacillus 33970 11 cereus H3081.97] 34 gi|42779235 arginase [Bacillus cereus ATCC 10987] 32299 11 35 gi|30023167 carboxylesterase [Bacillus cereus ATCC 28323 10 14579] 36 gi|30022242 2-oxoisovalerate dehydrogenase beta 35909 10 subunit [Bacillus cereus ATCC 14579] 37 gi|42779595 tellurium resistance protein [Bacillus 20991 9 cereus ATCC 10987] 38 gi|30260417 4-hydroxyphenylpyruvate dioxygenase 42052 9 [Bacillus anthracis str. Ames] 39 gi|42782685 aconitate hydratase [Bacillus cereus ATCC 99236 9 10987] 40 gi|217958855 camelysin [Bacillus cereus AH187] 21674 9 41 gi|217960824 metallo-beta-lactamase family protein 31689 8 [Bacillus cereus AH187] 42 gi|30018736 Serine protein kinase [Bacillus cereus 73282 8 ATCC 14579] 43 gi|206977523 transcriptional regulator, GntR family 24405 8 [Bacillus cereus H3081.97] 44 gi|30019505 ribonucleotide-diphosphate reductase 36994 7 subunit beta [Bacillus cereus ATCC 14579] 45 gi|206974851 conserved hypothetical protein [Bacillus 16768 7 cereus H3081.97] 46 gi|42782928 succinyl-CoA synthetase subunit beta 41894 7 [Bacillus cereus ATCC 10987] 47 gi|208702147 hypothetical protein BCH308197_A0066 46061 7 [Bacillus cereus H3081.97] 48 gi|30018338 Hsp33-like chaperonin [Bacillus cereus 32252 7 ATCC 14579] 49 gi|30263040 pyrrolidone-carboxylate peptidase 23612 7 [Bacillus anthracis str. Ames] 50 gi|30023418 putative heme peroxidase [Bacillus cereus 28791 7 ATCC 14579]

149 Supplemental data

Protein Protein match (No. Protein accession Protein description mass of peptides (Da) matched) 51 gi|30260278 glutamyl-tRNA synthetase [Bacillus 55638 7 anthracis str. Ames] 52 gi|42779183 DNA-directed RNA polymerase subunit 132096 7 beta [Bacillus cereus ATCC 10987] 53 gi|42784513 CTP synthetase [Bacillus cereus ATCC 59999 6 10987] 54 gi|206975459 conserved domain protein [Bacillus cereus 17878 6 H3081.97] 55 gi|30261823 hypothetical protein BA1774 [Bacillus 26855 6 anthracis str. Ames] 56 gi|42783124 mutT/nudix family protein [Bacillus cereus 21603 6 ATCC 10987] 57 gi|30264900 DNA-binding response regulator [Bacillus 26737 6 anthracis str. Ames] 58 gi|30022714 Thiol peroxidase [Bacillus cereus ATCC 18055 6 14579] 59 gi|206977441 conserved hypothetical protein [Bacillus 25407 6 cereus H3081.97] 60 gi|42783507 transcription elongation factor GreA 17428 6 [Bacillus cereus ATCC 10987] 61 gi|217958223 hypothetical protein BCAH187_A0766 37706 6 [Bacillus cereus AH187] 62 gi|206977333 Xaa-His dipeptidase [Bacillus cereus 51359 6 H3081.97] 63 gi|30265330 F0F1 ATP synthase subunit alpha [Bacillus 54667 6 anthracis str. Ames] 64 gi|30020419 methylisocitrate lyase [Bacillus cereus 32982 6 ATCC 14579] 65 gi|42780794 cold shock protein CspB [Bacillus cereus 7174 6 ATCC 10987] 66 gi|42783163 phosphocarrier protein HPr [Bacillus 9176 5 cereus ATCC 10987] 67 gi|42784288 glyceraldehyde-3-phosphate 35928 5 dehydrogenase [Bacillus cereus ATCC 10987] 68 gi|30019202 transcriptional regulator Hpr [Bacillus 21773 5 cereus ATCC 14579] 69 gi|30021934 hypothetical protein BC3844 [Bacillus 13302 5 cereus ATCC 14579] 70 gi|30264918 naphthoate synthase [Bacillus anthracis 29844 5 str. Ames] 71 gi|30022999 4-nitrophenylphosphatase [Bacillus cereus 28036 5 ATCC 14579] 72 gi|30022599 thioredoxin [Bacillus cereus ATCC 14579] 11585 5 73 gi|30022558 ATP-dependent protease La [Bacillus 86591 5 cereus ATCC 14579] 74 gi|121357 RecName: Full=Glutamine synthetase; 50448 5 AltName: Full=Glutamate--ammonia ligase 75 gi|30264861 S-ribosylhomocysteinase [Bacillus 18072 5 anthracis str. Ames] 76 gi|206974521 putative copper homeostasis protein CutC 24811 5 [Bacillus cereus H3081.97] 77 gi|229021520 SMI1 / KNR4 [Bacillus cereus AH1273] 17923 5 150 Supplemental data

Protein Protein match (No. Protein accession Protein description mass of peptides (Da) matched) 78 gi|30264237 3-methyl-2-oxobutanoate dehydrogenase, 36901 5 alpha subunit [Bacillus anthracis str. Ames] 79 gi|30023338 F0F1 ATP synthase subunit gamma 31586 5 [Bacillus cereus ATCC 14579] 80 gi|30264231 hypothetical protein BA4378 [Bacillus 16232 5 anthracis str. Ames] 81 gi|206975417 glyoxalase family protein [Bacillus cereus 33863 5 H3081.97] 82 gi|47566274 hypothetical protein BCE_G9241_1400 21506 5 [Bacillus cereus G9241] 83 gi|30021938 acyl carrier protein [Bacillus cereus ATCC 8809 4 14579] 84 gi|30260268 lysyl-tRNA synthetase [Bacillus anthracis 57644 4 str. Ames] 85 gi|206975360 tetratricopeptide domain protein [Bacillus 42897 4 cereus H3081.97] 86 gi|30022341 hydroxyacylglutathione hydrolase 23192 4 [Bacillus cereus ATCC 14579] 87 gi|42783200 pyrimidine-nucleoside phosphorylase 46240 4 [Bacillus cereus ATCC 10987] 88 gi|118478354 NAD(P)H nitroreductase [Bacillus 23598 4 thuringiensis str. Al Hakam] 89 gi|42781158 general stress protein [Bacillus cereus 16696 4 ATCC 10987] 90 gi|30022355 superoxide dismutase [Mn] [Bacillus 24738 4 cereus ATCC 14579] 91 gi|217960661 hypothetical protein BCAH187_A3277 18349 4 [Bacillus cereus AH187] 92 gi|30263796 GntR family transcriptional regulator 27564 4 [Bacillus anthracis str. Ames] 93 gi|30022960 phosphoglucomutase [Bacillus cereus 64302 4 ATCC 14579] 94 gi|30023440 azoreductase [Bacillus cereus ATCC 22744 4 14579] 95 gi|206976478 putative comA operon protein [Bacillus 14220 4 cereus H3081.97] 96 gi|30023498 30S ribosomal protein S6 [Bacillus cereus 11292 4 ATCC 14579] 97 gi|30018290 seryl-tRNA synthetase [Bacillus cereus 48931 4 ATCC 14579] 98 gi|42780330 ornithine--oxo-acid transaminase [Bacillus 43600 3 cereus ATCC 10987] 99 gi|30021868 stage V sporulation protein S [Bacillus 8921 3 cereus ATCC 14579] 100 gi|30018632 hypothetical protein BC0424 [Bacillus 11845 3 cereus ATCC 14579] 101 gi|4584148 pyruvate carboxylase [Bacillus cereus 110294 3 ATCC 10987] 102 gi|30262422 cold shock protein CspA [Bacillus anthracis 7321 3 str. Ames] 103 gi|206975465 conserved hypothetical protein [Bacillus 19272 3 cereus H3081.97]

151 Supplemental data

Protein Protein match (No. Protein accession Protein description mass of peptides (Da) matched) 104 gi|30022683 phosphoesterase, DHH family protein 34914 3 [Bacillus cereus ATCC 14579] 105 gi|196038957 putative lacX protein [Bacillus cereus 33300 3 NVH0597-99] 106 gi|42782768 transketolase [Bacillus cereus ATCC 72484 3 10987] 107 gi|42783066 branched-chain alpha-keto acid 46013 3 dehydrogenase subunit E2 [Bacillus cereus ATCC 10987] 108 gi|30022178 ferric uptake regulation protein [Bacillus 17878 3 cereus ATCC 14579] 109 gi|30260304 50S ribosomal protein L2 [Bacillus 30291 3 anthracis str. Ames] 110 gi|42781687 D-alanine--D-alanine ligase [Bacillus 33974 3 cereus ATCC 10987] 111 gi|47569179 conserved hypothetical protein protein 26018 3 [Bacillus cereus G9241] 112 gi|206977867 hypothetical protein BCH308197_1301 72788 3 [Bacillus cereus H3081.97] 113 gi|42784488 ribose-5-phosphate isomerase B [Bacillus 16146 3 cereus ATCC 10987] 114 gi|42783696 electron transfer flavoprotein, beta 28213 3 subunit [Bacillus cereus ATCC 10987] 115 gi|42783531 aspartyl-tRNA synthetase [Bacillus cereus 66525 3 ATCC 10987] 116 gi|30022198 bifunctional 3,4-dihydroxy-2-butanone 4- 44315 3 phosphate synthase/GTP cyclohydrolase II protein [Bacillus cereus ATCC 14579] 117 gi|30021888 dihydrodipicolinate synthase [Bacillus 31243 3 cereus ATCC 14579] 118 gi|42783592 Maf-like protein [Bacillus cereus ATCC 22400 3 10987] 119 gi|42780240 hypothetical protein BCE_1165 [Bacillus 18952 3 cereus ATCC 10987] 120 gi|30021887 Zn-dependent hydrolase [Bacillus cereus 61122 3 ATCC 14579] 121 gi|30022292 lipoate-protein ligase A [Bacillus cereus 32312 3 ATCC 14579] 122 gi|206973900 metallo-beta-lactamase family protein 26501 3 [Bacillus cereus H3081.97]

152 Samenvatting

Summary Samenvatting

153 Summary

Summary

Due to increasing demands for, and decreasing supply of, fossil reserves, sustainable solutions are needed for cheaper and cleaner alternative feedstock for the production of chemicals and fuels. Renewable resources such as lignocellulosic biomass are promising feedstocks for the production of bio-fuels and value-added products. Biocatalysts are considered important tools in such processes.

Pseudomonas putida S12 has a broad metabolic potential and is exceptionally tolerant towards a range of toxic organic solvents and aromatic compounds. It has various mechanisms to cope with hydrophobic toxic compounds; it can modify its inner and outer membranes to counteract the effect of solvents, or actively extrude solvents through membrane-associated efflux pumps. These mechanisms make this bacterium a very suitable host for the production of aromatic compounds. Another factor that possibly contributes to the tolerance of P. putida S12 to stress and solvents is the polyamine putrescine, which exerts a protective effect in stressed microbial cells. The metabolic pathway of putrescine was investigated in P. putida S12 (Chapter 2) . A transcriptomics approach was employed to identify the genes that were potentially involved in putrescine metabolism. A number of the identified genes, particularly those involved in a transamination/dehydrogenation pathway, were inactivated to assess their role in putrescine metabolism. Several of the resulting (multiple) knockout strains were affected in their ability to degrade putrescine. However, none were obtained that completely lost this capacity. Evidence was found for an alternative degradation pathway in P. putida S12 via a gamma-glutamylation route. The high level of redundancy in putrescine degrading enzymes and pathways illustrates the multifaceted nature of putrescine metabolism in P. putida S12, as well as the importance of polyamine homeostasis for normal cell functioning.

The utilisation of lignocellulosic biomass as a renewable feedstock for sustainable energy and chemicals production, is gaining increased attention. Still, intensive research is

154 Samenvatting required to enable efficient utilisation of all components contained within lignocellulose, and to expand the range of value-added products to be obtained from this resource. At present, Kraft lignin is the predominant lignin type available. This byproduct from the paper and pulping industry is commonly burnt to generate power and heat. Microorganisms capable of growing on this complex substrate may be a source of novel enzymes which can be of use for the valorization of lignin. Chapter 3 describes the isolation of three soil isolates that are able to grow on Kraft lignin and evaluates their ligninolytic potential. Of the three isolates, Pandoraea norimbergensis LD001 and Pseudomonas sp. LD002 showed relatively good growth on lignin, but a limited ability to decolourize various lignin-model dyes. Bacillus sp. LD003 showed least efficient growth on lignin, but exhibited the most extensive dye-decolourizing capacity. This was best illustrated by its ability to readily decolourize the recalcitrant thiazine dye Azure B (AB), which is frequently used to assess ligninolytic activity.

Based on its dye-decolourizing capacity, Bacillus sp. LD003 was selected as a promising source of novel types of ligninolytic enzymes. In Chapter 4 , it was attempted to isolate and characterize the enzyme(s) involved in the decolourization of AB, which may have novel and valuable properties. It was observed that the dye-decolourizing activity was intracellular rather than extracellular, as no decolourization was observed for the culture supernatant. Several candidate enzymes were identified via mass spectrometry and the corresponding genes were synthesized and cloned into Escherichia coli . A quinone NAD(P)H dehydrogenase (QD) was successfully expressed and shown to decolourize AB in solid as well as liquid phase assays. QDs play a role in the reduction of quinones, presumably to protect the cell from oxidative damage. Various quinones have been identified as degradation intermediates of lignin, lignin model compounds and aromatic pollutants. Thus, QD possibly plays a role in the metabolism of lignin-derived compounds.

The work described in this thesis provides insights into the metabolism of the polyamine putrescine in P. putida S12, which presumably plays a role in the response to solvent stress. Furthermore, novel bacterial isolates have been investigated that could potentially serve as a source of lignin degrading enzymes. Application of such enzymes combined with

155 Summary the solvent tolerance of P. putida S12 could open up a new avenue for the valorization of the ubiquitously available renewable resource lignin. It has also been made clear, however, that the recalcitrant nature and complex structure of lignin greatly complicates the research into biological lignin depolymerization. Much is to be learned and possibly gained from bacterial lignin degradation, which justifies more and intensified research.

Samenvatting

Door de stijgende vraag naar, fossiele grondstoffen tegen een afnemende voorraad zijn duurzame, alternatieve grondstoffen nodig voor goedkopere en schonere productie van chemicaliën en brandstoffen. Hernieuwbare grondstoffen zoals lignocellulose zijn veelbelovend voor de productie van biobrandstoffen en meer hoogwaardige producten. Biokatalysatoren kunnen in dergelijke processen een belangrijke rol spelen.

Pseudomonas putida S12 heeft een flexibel metabolisme en is zeer tolerant voor een scala aan toxische organische oplosmiddelen en aromatische verbindingen. Deze bacterie beschikt over verschillende mechanismen om weerstand te bieden aan hydrofobe toxische stoffen: de samenstelling van de binnen- en buitenmembraan kan aangepast worden om schade door organische oplosmiddelen te voorkomen. Deze oplosmiddelen kunnen bovendien actief verwijderd worden door middel van membraan-gebonden efflux pompen. P. putida S12 is daarom bijzonder geschikt voor de productie van aromatische verbindingen. Een andere factor die mogelijk bijdraagt aan de stress- en oplosmiddeltolerantie van P. putida S12 is het polyamine putrescine waarvan bekend is dat het aan stress blootgestelde microbiële cellen kan beschermen. De metabole route van putrescine in P. putida S12 is onderzocht (Hoofdstuk 2) waarbij de betrokken genen werden geïdentificeerd via een transcriptomics aanpak. Een aantal genen werd geïnactiveerd om de rol in het putrescine-metabolisme vast te kunnen stellen, met nadruk op genen die mogelijk betrokken zijn in transaminering/dehydrogenerings routes. Verschillende van de (meervoudige) knock-out mutanten vertoonden een verminderde capaciteit om putrescine af te breken. Er werd echter geen mutant verkregen die het vermogen putrescine af te breken volledig had verloren. Wel werden aanwijzingen

156 Samenvatting gevonden die duidden op het bestaan van een alternatieve afbraakroute in P. putida S12, via gamma-glutamylering. De overvloed aan genen en verschillende routes die blijkbaar betrokken zijn bij putrescine catabolisme duidt op een meervoudige rol van putrescine in P. putida S12 en onderstreept het belang van polyamine homeostase voor het juist functioneren van de cel.

Het gebruik van lignocellulose als hernieuwbare grondstof voor de duurzame productie van energie en chemicaliën ondervindt toenemende aandacht. Niettemin is intensief onderzoek nodig om het gebruik van alle componenten in lignocellulose mogelijk te maken en om het aantal hoogwaardige producten dat uit deze grondstof verkregen kan worden, uit te breiden. Het type lignine dat op dit moment het meest op grote schaal geproduceerd wordt is Kraft lignine. Dit bijproduct van de papier- en pulpindustrie wordt gewoonlijk verbrand om energie en hitte te produceren. Micro-organismen die dit complexe substraat kunnen benutten zijn mogelijk een bron voor nieuwe enzymen die gebruikt kunnen worden voor de valorisatie van lignine. Hoofdstuk 3 beschrijft de isolatie van drie bacteriestammen die op Kraft lignine kunnen groeien inclusief het lignolytisch potentieel van deze micro-organismen. Twee van de isolaten, Pandoraea norimbergensis LD001 en Pseudomonas sp. LD002, groeiden betrekkelijk goed op lignine maar ontkleurden diverse lignine-achtige kleurstoffen slechts matig. Bacillus sp. LD003 groeide weliswaar minder goed op lignine, maar vertoonde de beste ontkleurende activiteit op de recalcitrante thiazine-kleurstof Azure B (AB), die vaak wordt gebruikt om lignolytische activiteit aan te tonen.

Op basis van het AB-ontkleurend vermogen werd Bacillus sp. LD003 geselecteerd als een potentiële bron voor nieuwe lignolytische enzymen. In Hoofdstuk 4 werd gepoogd enzymen betrokken bij de ontkleuring van AB te isoleren en te karakteriseren. Deze enzymen hebben mogelijk nieuwe en waardevolle eigenschappen. De kleurstof bleek intracellulair te worden ontkleurd aangezien ontkleuring niet werd waargenomen in het supernatant van de bacterie-cultures. Mogelijke kandidaat-enzymen werden geïdentificeerd via massaspectrometrie. De corresponderende genen werden gesynthetiseeerd en gekloneerd in Escherichia coli . Een quinon-NAD(P)H-dehydrogenase

157 Summary

(QD) kon functioneel tot expressie worden gebracht en bleek bovendien in staat om AB te ontkleuren. QD’s spelen een rol in de reductie van quinonen, waarschijnlijk om microbiële cellen te beschermen tegen oxidatieve schade. Verschillende quinonen komen voor als afbraakproducten van lignine, lignine-achtige verbindingen en aromatische componenten. Een rol van QD in het metabolisme van lignine-afgeleide verbindingen lijkt daarom zeer waarschijnlijk.

Het werk in dit proefschrift beschrijft nieuwe inzichten in het metabolisme van het polyamine putrescine in P. putida S12, dat waarschijnlijk een rol speelt in de respons op oplosmiddel-gerelateerde stress. Daarnaast zijn er nieuwe bacteriën bestudeerd waaruit mogelijk lignine-afbrekende enzymen verkregen kunnen worden. Dergelijke enzymen, gecombineerd met de oplosmiddeltolerantie van P. putida S12, kunnen nieuwe mogelijkheden bieden voor het valoriseren van de alom beschikbare hernieuwbare grondstof lignine. Het is echter tevens duidelijk geworden dat het onderzoek naar de biologische depolymerisatie van lignine wordt bemoeilijkt door de recalcitrante eigenschappen en complexe structuur van lignine. Met meer, en met name meer intensief onderzoek valt er nog veel te ontginnen en te winnen met betrekking tot bacteriële lignine afbraak.

158

Curriculum Vitae Publications Acknowledgements

159 Curriculum Vitae

Curriculum Vitae

Luaine Jongman Bandounas was born on 16 August 1979 in Vanderbijlpark, South Africa. In December 1997, she graduated from Sasolburg High School in Sasolburg, South Africa. She went on to complete a BSc in Microbiology in 2000, followed by a BSc (Honours) in Medical Virology in 2001 at the University of Pretoria, South Africa. In 2005, she graduated cum laude with an MSc in Molecular Cell Biology and Bioinformatics from the University of Amsterdam. During this period she completed 2 internships: the first was performed at the Swammerdam Institute for Life Sciences (SILS), University of Amsterdam, which involved investigating the heat resistance of Bacillus subtilis spores. The second internship was completed at the Academic Centre for Dentistry Amsterdam (ACTA), where the disinfectant resistance and gene expression of Streptococcus mutans was investigated in dual species biofilms. She started her PhD at Delft University of Technology in 2005. This research was initially performed at TNO, Quality of Life, Bioconversion in Apeldoorn under the supervision of Dr. Jan de Bont (2005-2006) and Dr. Jan Wery (2006-2007). In January 2007, the research group relocated to the Kluyver laboratory in Delft, where the remainder of the work was performed at the Department of Biotechnology, supervised by Prof. Dr. Han de Winde and Dr. Harald Ruijssenaars. The results of her research are presented in this thesis. Since September 2010, she is working for The Federation of European Microbiological Societies (FEMS) at their Central Office in Delft, as Editorial Administrator.

160 Publications

Publications

Bandounas L , Ballerstedt H, de Winde JH, Ruijssenaars HJ, 2011. Redundancy in putrescine catabolism in solvent tolerant Pseudomonas putida S12. Journal of Biotechnology, 154 (1) : 1-10.

Bandounas L , Wierckx N, de Winde JH, Ruijssenaars HJ , 2011. Isolation and characterization of novel bacterial strains exhibiting ligninolytic potential. BMC Biotechnology , 11 : 94.

Bandounas L , Pinkse M, de Winde JH, Ruijssenaars HJ. Decolourization of the lignin-model dye, Azure B by a ligninolytic Bacillus sp. and initial identification of enzymes involved. Submitted .

Wierckx N, Koopman F, Bandounas L , de Winde JH, Ruijssenaars HJ, 2009. Isolation and characterization of Cupriavidus basilensis HMF14 for biological removal of inhibitors from lignocellulosic hydrolysate . Microbial Biotechnology, 3 (3) : 336-343.

Luppens SB, Kara D, Bandounas L , Jonker MJ, Wittink FR, Bruning O, Breit TM, Ten Cate JM, Crielaard W, 2008. Effect of Veillonella parvula on the antimicrobial resistance and gene expression of Streptococcus mutans grown in a dual-species biofilm. Oral Microbiology and Immunology, 23 (3) :183-9.

161 Acknowledgements

Acknowledgements

After an extremely enjoyable, yet sometimes tumultuous past few years of my PhD, I take great pleasure in writing the final part of my thesis. Some of my more fond memories include the many social activities and the relaxed working atmosphere; while the more stressful experiences involve the relocation of facilities from Apeldoorn to Delft, as well as changing research projects in the later stages of my research. Nevertheless, I survived and finally managed to finish.

I would like to thank my promoter Han de Winde, for the courage and enthusiasm he displayed after agreeing to supervise 7 PhD students once our research group had relocated to Delft. His reassuring words, optimistic nature and advice were always a welcome part of our discussions, which provided me with the necessary calm and peace I sometimes needed during the more turbulent times. I am very grateful to my supervisor, Harald Ruijssenaars. During, and after our relocation, he stepped up to the plate and took on the responsibility of guiding and supporting me through my research. His door was always open for help and advice. Furthermore, he also had the tiresome task of editing my manuscripts; his effort does not go unappreciated!

I would also like to thank Jan de Bont and Jan Wery for setting up the projects and for giving me the opportunity to do my PhD. Thank you both for your support and supervision during the initial stage of my research.

Next I have to mention the colleagues I spent most of my time with during this period. It was not always easy for me to express myself properly in Dutch and probably not that easy for you to understand me either, therefore I was sometimes rather quiet because by the time I had thought of how I should comment in Dutch, the conversation had already moved on! Nevertheless, I really had an enjoyable time with everyone in the lab and at conferences and social activities. First of all I would like to thank the “old PhD’s” Nick, Karin and Rita for their advice and assistance. It was a pleasure working with all of you.

162 Acknowledgements

Suzanne, Frank and Jean-Paul, thanks for your interest, your help and all the memorable and enjoyable experiences we had together. Sabrina and Mirjam, thanks for your advice and friendship, I really enjoyed your company. Hendrik, thank you for the microarray work and all your advice. Maaike and Lars, thanks for your cloning efforts! I would also like to thank Dorien, Hugo, Bas, Jasperien, Jan-Harm, Corjan, Yang and all the people at the Kluyver lab who played a part in creating a wonderful working environment. To my gym partner Louise, it was great to get rid of our irritation and frustration at the gym! Of course I also need to thank the two students I had the pleasure of supervising, first of all, Alexandra, I really enjoyed our time together, thank you for your enthusiasm and hard work! I would also like to thank Ismael, whom I supervised at one of the most stressful periods of my PhD, right up until the last day of my contract!

I would also like to acknowledge the interest and support of my colleagues at FEMS!

Now to thank the most important people in my life, namely my family. During my PhD there were some really stressful and emotional moments, which made it difficult to stay motivated and positive at times, but with the unconditional support and encouragement of my family, I managed to get through it. Mom, I really appreciate that you were always there to support me and lift my spirits when I was feeling down, even though you were miles away. To my brother Warren, thanks for your support and interest, even when you didn’t always understand what I was talking about. To Bert, my stepdad, thank you for everything you have done for me and my family in the last few years. A special acknowledgement goes to my late grandmother, Ouma Mini. Oupa Piet en Ouma Gretha, ik ben nu uiteindelijk klaar met mijn “studie” en ik weet dat jullie heel trots zijn, hartelijk bedankt voor jullie enthousiasme en steun. I would also like to acknowledge my parents- in-law, as well as Tessa and Iman for their help and support. To my husband, Costa, during this period you have experienced many of the highs and lows with me and I want to thank you for your constant encouragement and your unwavering confidence in me. I really appreciate your patience and support during this time.

Luaine

163