UNIVERSIDADE DE LISBOA FACULDADE DE CIÊNCIAS DEPARTAMENTO DE BIOLOGIA VEGETAL
Dibenzothiophene desulfurization by Gordonia alkanivorans strain 1B
Luís Manuel Gonçalves Alves
Ph.D. in Biology (Microbiology) 2007
On the Cover - Photo of colonies of Gordonia allkanivorans.
UNIVERSIDADE DE LISBOA FACULDADE DE CIÊNCIAS DEPARTAMENTO DE BIOLOGIA VEGETAL
Dessulfurização de dibenzotiofeno por Gordonia alkanivorans estirpe 1B
Luís Manuel Gonçalves Alves
ORIENTADORES:
Professor Doutor Rogério Paulo de Andrade Tenreiro
Professor Auxiliar da Faculdade de Ciências da Universidade de Lisboa
Doutor José António dos Santos Pereira de Matos Investigador Auxiliar do Instituto Nacional de Engenharia, Tecnologia e Inovação
DOUTORAMENTO EM BIOLOGIA (MICROBIOLOGIA) 2007
Na presente dissertação incluem-se resultados que foram alvo de publicação com outros autores. Para efeitos do disposto no nº 2 do Art. 8º do Decreto-Lei 388/70, O autor da dissertação declara que interveio na concepção e execução do trabalho experimental, na interpretação dos resultados e na redacção dos manuscritos publicados ou enviados para publicação.
Lisboa, 03 de Agosto, 2007
------(Luís Manuel Gonçalves Alves)
Para os meus pais, a minha esposa e o meu filho com carinho.
"A ciência nunca está concluída, está cada vez mais próxima da compreensão total e rigorosa da natureza, mas nunca chega a alcançá-la.”
Carl Sagan
Abbreviations
2-HBP – 2-Hydroxybiphenyl HPBS – 2-(2′-hydroxyphenyl) (2-hidroxibifenilo) benzene sulfinate 4,6-dmDBT – 4,6-dimethylDBT HPLC – High performance liquid (4,6-dimetilDBT) chromatography 4-mDBT – 4-methylDBT (4-metilDBT) Ile – Isoleucine Ala – Alanine IPTG – Isopropyl β-D-1- thiogalactopyranoside aa – Amino acid Lys – Lysine BDS – Biodesulfurization LRP – Lamas da reciclagem de papel (biodessulfurização) LB – Luria-Bertani broth bp – Base pair BT – Benzothiophene µ – Specific growth rate rDNA – Ribosomal DNA MS – Mass spectroscopy DNS – Dinitrosalysilic acid Asn – Asparagine DBT – Dibenzothiophene (dibenzotiofeno) NADH – Nicotinamide adenine dinucleotide DBTS – Dibenzothiophene sulfone OD– Optical density DCW – Dry cell weight ppm – Parts per million (partes por milhão) DO – Densidade óptica PCR – Polymerase chain reaction DSMZ – Deutsche Sammlung von rbs – Ribossome biding site Mikroorganismen und Zellkulturen GmbH rpm – Revolutions per minute dsz – Desulfurization RPS – Recycled paper sludge DszA – DBT-monooxygenase Ser – Serine DszB – DBTS-monooxygenase SD – Standard deviation DszC – 2'-Hydroxybiphenyl- SDS-PAGE – sodium dodecyl sulfate 2-sulfinate desulfinase polyacrylamide gel electrophoresis DszD – Flavin-reductase SFM – sulfur-free mineral DMF – Dimethylformamide Thr – Threonine DNA – Deoxyribonucleic acid tds – Thermal desulfurization Fig – Figure U – International units Val – Valine FMN – Flavin mononucleotide Trp – Tryptophan FP – Filter paper v/v – Volume/volume Gly – Glycine w/v – Weight/volume GC – Gas chromatography w/w – Weight/ weight HDS – Hydrodesulfurization (hidrodessulfurização)
Resumo
A crescente utilização de combustíveis fósseis levou a um aumento da emissão de óxidos de enxofre para a atmosfera, os quais são um dos principais causadores das chuvas ácidas. A legislação já aprovada prevê que em 2009 o nível máximo de enxofre nos combustíveis seja apenas 10 ppm, enquanto actualmente esse valor se situa nos 150 ppm. O processo de hidrodessulfurização (HDS) utilizado nas refinarias é baseado em técnicas físico-químicas muito dispendiosas, além de apresentar limitações na remoção do enxofre orgânico. Quanto mais estrita for a legislação sobre os níveis máximos de enxofre nos combustíveis fósseis, mais compostos recalcitrantes à HDS necessitam de ser removidos. Isto implica um aumento da intensidade do tratamento físico-químico e inerentemente dos seus custos. Como resultado, os compostos recalcitrantes à HDS representam uma barreira significativa para a obtenção de níveis de enxofre muito baixos nalgumas fracções petrolíferas.
Diversas entidades governamentais e companhias petrolíferas já reconheceram a dificuldade de cumprir as regulamentações ambientais de uma maneira eficiente e económica usando a tecnologia convencional de HDS uma vez que as unidades de HDS para a dessulfurização profunda são extremamente dispendiosas de construir e operar. Por isso, é muito importante o estudo de novos processos de dessulfurização que possam de alguma maneira substituir ou complementar a HDS. A dessulfurização biológica poderá ser uma dessas tecnologias a implementar nos próximos anos pela indústria petrolífera.
A alternativa ao tratamento físico-químico passa pelo recurso a processos biológicos (biodessulfurização) mais eficazes para a dessulfurização dos combustíveis fósseis, nomeadamente ao nível da remoção do enxofre ligado covalentemente a matrizes orgânicas. A biodessulfurização (BDS) ocorre em condições de funcionamento mais amenas sob condições de pressão atmosférica e temperatura ambiente, apresentando maior especificidade de reacção devido à natureza dos biocatalisadores, não requerendo
7
hidrogénio molecular e permitindo a manutenção no processo da emissão de CO2 a um nível baixo. Por estes motivos, a remoção de enxofre por processos biocatalíticos é hoje em dia considerada como alternativa ou como complemento do processo de HDS convencional usado na indústria petrolífera. Deste modo, na última década e meia tem-se verificado um aumento dos estudos envolvendo a utilização de microrganismos com a capacidade de remover especificamente o enxofre deste tipo de compostos.
A grande maioria dos trabalhos de BDS foi efectuada com estirpes de Rhodococcus erythropolis, especialmente a estirpe IGTS8 a qual se tornou a estirpe referência neste tipo de estudos. Assim, foi obtido um excelente conhecimento da BDS de dibenzotiofeno (DBT) e compostos análogos em termos de fisiologia bacteriana, enzimologia e biologia molecular. No entanto este conhecimento resumiu-se inicialmente ao género Rhodococcus e só mais recentemente se começou a estudar bactérias pertencentes a outros géneros.
Neste trabalho foi isolada e seleccionada uma bactéria a partir de solos contaminados por hidrocarbonetos, que demonstrou uma boa capacidade de dessulfurização de DBT. Este composto é utilizado como modelo na maioria dos estudos de BDS. A bactéria seleccionada foi identificada como Gordonia alkanivorans estirpe 1B após estudos de caracterização de microbiologia clássica, bioquímica e biologia molecular. Embora esta espécie tenha sido descrita pela primeira vez em 1998, nenhum trabalho de BDS associado a esta tinha sido publicado.
O crescimento desta bactéria, num meio de cultura com glucose como fonte de carbono e DBT como única fonte de enxofre, permite a dessulfurização deste composto formando-se 2-hidroxibifenilo (2-HBP) e libertando-se o enxofre na forma de sulfito. A taxa de dessulfurização específica obtida foi de 1,03 µmol g-1(biomassa) h-1 para uma taxa específica de crescimento de 0,019 h-1. Esta bactéria tem ainda a capacidade de utilizar outros compostos tiofénicos, tais como o benzotiofeno, tiofeno, 4-metil e 4,6-dimetil- dibenzotiofeno.
8
Foi identificado e sequenciado o operão de G. alkanivorans estirpe 1B, responsável pela capacidade de dessulfurização de dibenzotiofeno, sendo constituído por três genes. As sequências nucleotídicas obtidas apresentam uma semelhança na ordem dos 85 a 90% em relação às sequências dos genes homólogos de Rhodococcus erythropolis IGTS8, permitindo concluir que G. alkanivorans estirpe 1B utiliza a via metabólica 4S. A principal vantagem desta via metabólica é não ocorrer diminuição do potencial energético do composto dessulfurizado, facto importante para a dessulfurização de combustíveis fósseis.
Tendo em vista a utilização de fontes de carbono alternativas obtidas a partir de resíduos agro-industriais na formulação de meios de cultivo, foi estudado o efeito da presença ou ausência de alguns iões metálicos que compõem o meio de cultivo utilizado no crescimento laboratorial desta bactéria. Para os iões metálicos estudados, apenas a ausência de cobre e de zinco no meio de cultura diminuiu a quantidade de 2-HBP produzida. No entanto, a ausência de zinco reduziu a biomassa produzida, indicando que este ião pode ter uma importância relevante para o metabolismo de G. alkanivorans estirpe 1B.
Crescimentos da estirpe 1B, em meios de cultura contendo sulfato ou DBT como fonte de enxofre, permitiram verificar que o ião zinco apenas estimula o metabolismo bacteriano na presença de DBT. Isto sugere que o ião zinco tem um papel importante no sistema enzimático envolvido na dessulfurização. Estes resultados foram confirmados após estudos envolvendo células de G. alkanivorans estirpe 1B pré-crescidas na presença ou ausência de zinco. De facto, no ensaio com células pré-crescidas com zinco, foi possível obter uma produtividade específica de 2-HBP de 2,29 µmol g-1(biomassa) h-1, valor este 7,6 vezes superior ao obtido no ensaio com células pré-crescidas na ausência de zinco. A produtividade obtida na ausência de zinco sofreu um incremento de 70% quando as células pré-crescidas foram incubadas com 1 mg l-1 de zinco. Estes resultados permitiram verificar a necessidade de aumentar a quantidade de zinco inicialmente utilizado no meio de cultura (0,5 mg l-1) para 10 mg l-1 de modo a maximizar a taxa de dessulfurização de DBT.
9
A aplicação industrial da BDS depende do custo inerente à produção de biocatalisadores, o qual ainda é muito elevado para permitir que este processo seja viável em termos económicos. Assim, o estudo de fontes de carbono alternativas resultantes da hidrólise de materiais agro-industriais pode ser uma das estratégias para reduzir esses custos. Neste trabalho foram utilizadas lamas da reciclagem de papel (LRP) uma vez que é um material rico em celulose (35%). Os polissacáridos das LRP foram hidrolisados com uma mistura enzimática, a qual foi previamente dialisada para remoção de compostos contendo enxofre. O hidrolisado obtido continha, respectivamente, 36,3 e 53,9 g l-1 de glucose e açúcares totais, com um rendimento de hidrólise de 72%.
O hidrolisado de LRP foi inicialmente utilizado como fonte de carbono (10 g l-1 glucose), resultando um elevado crescimento bacteriano (DO600 nm = 10 após 4 dias de cultura) e no caso do hidrolisado obtido com a mistura enzimática dialisada obteve-se uma produtividade específica de 2-HBP de 1,1 µmol g-1(biomassa) h-1. Nestas condições G. alkanivorans estirpe 1B apresentou uma taxa de dessulfurização um pouco superior à obtida após crescimento em glucose comercial.
Este hidrolisado foi ainda testado como fonte de outros nutrientes para além de fonte de carbono. Obteve-se um bom crescimento bacteriano (DO600 nm = 9 após 5 dias de cultura) e dessulfurização de DBT quando o hidrolisado de LRP foi suplementado apenas com fosfatos e amónia. No entanto, a melhor suplementação foi a que incluiu adicionalmente magnésio e 10 mg l-1 de zinco, onde a taxa específica de crescimento foi de 0,03 h-1.
Apesar de neste trabalho não ter sido estudada a dessulfurização de nenhum combustível fóssil, procedeu-se no entanto a um estudo de dessulfurização de um combustível modelo. Este modelo era constituído por DBT, 4-metilDBT e 4,6-dimetilDBT dissolvidos em n-heptano. G. alkanivorans estirpe 1B, utilizando o hidrolisado de LRP como fonte de nutrientes, diminuiu o enxofre total presente neste combustível de 6 mM para 2,23 mM após 7 dias de cultivo, para uma taxa específica de crescimento de 0,062 h-1. Esta estirpe dessulfuriza preferencialmente DBT, seguido de 4-metilDBT e por fim 4,6-dimetilDBT,
10
embora a dessulfurização dos 3 compostos ocorra simultaneamente. A taxa específica máxima de dessulfurização obtida foi de 22,2 µmol g-1(biomassa) h-1, valor semelhante aos descritos para algumas estirpes selvagens de R. erythropolis e Gordonia sp.
Estes resultados permitiram mostrar que o hidrolisado de LRP pode ser utilizado como fonte de nutrientes para a dessulfurização de DBT e seus derivados por G. alkanivorans estirpe 1B, apresentando uma eficiência semelhante à obtida no meio de cultura convencional com glucose comercial. A utilização de LPR apresenta um duplo benefício: as LRP podem ser usadas como fonte de nutrientes num processo biotecnológico contribuíndo para a resolução do problema ambiental deste resíduo.
Em conclusão, os resultados obtidos na melhoria da actividade de dessulfurização por G. alkanivorans estirpe 1B e a utilização de um meio de cultura menos dispendioso revelaram-se promisores para uma futura aplicação biotecnológica desta bactéria.
Palavras-Chave: Biodesulfurização, Combustíveis fósseis, Dibenzotiofeno, Gordonia alkanivorans, Fontes de carbono alternativas.
11
Abstract
The decreased availability of low sulfur crude oils has resulted in a need to refine heavier, higher sulfur crude. The combustion of this crude produces sulfur oxides that contribute to air pollution and, consequently, many countries have changed their legislation to obtain a progressive reduction in sulfur content of fossil fuels. The microbiological process, biodesulfurization, offers the potential for an alternative/complementary method for lowering the sulfur content of petroleum products.
A mesophilic bacterium identified as Gordonia alkanivorans strain 1B was isolated from soil samples contaminated with hydrocarbons. This strain has the ability to desulfurize dibenzothiophene (DBT) and also other sulfur compounds, namely, benzothiophene, DBT sulfone, 4-methylDBT and 4,6-dimethylDBT. The identification, sequencing and characterization of desulfurization genes of strain 1B allowed to confirm that the desulfurization of DBT occurs through the specific sulfur pathway 4S.
In order to use cheaper alternative nutrient sources to decrease the costs associated to the production of biocatalysts, the culture medium requirements were studied, mainly the metal ion composition. It was found that zinc ion may have an important role for one or more desulfurization enzymes. The increase of zinc concentration from 0.5 to 10 mg l-1 allowed an improvement of DBT desulfurization of about 26%.
Polysaccharides of recycled paper sludge were hydrolyzed with an enzyme formulation and the hydrolyzate obtained was used for nutrient source supplementation of culture medium to grow Gordonia alkanivorans strain 1B. Under these conditions the bacterium was able to desulfurize a model oil containing DBT, 4-methylDBT and 4,6-dimethylDBT, with a reduction on total sulfur from 6 to 2.23 mM. The maximum specific desulfurization rate of strain 1B in model oil was 22.2 µmol g-1(DCW) h-1 very similar to some reported wild type Rhodococcus erythropolis and Gordonia sp. strains.
13
The improvement of desulfurization activity in addition to the use of a less expensive culture medium is an important achievement for a future application of this bacterium.
Keywords: Biodesulfurization, Fossil fuels, Dibenzothiophene, Gordonia alkanivorans, Alternative carbon sources.
14
Contents
CONTENTS
Chapter 1
General introduction …………………………………………………………. 21 [Partially in press – Enzimas em Biotecnologia – Produção, Aplicações e Indústria (Chapter 18)]
Fossil fuels desulfurization…………………………………………………………..21 Sulfur in fossil fuels……………………………………..……………………..21 The sulfur problematics……….…..……………………………………………22 Regulations………………….………………..………………………………...24 Hydrodesulfurization…….……...………………………………………………24 Biodesulfurization………………………………………………………………27 Desulfurizing microorganisms……..………..………………………………….28 Model compounds biodesulfurization……………………………………………….29 Benzothiophene biodesulfurization…………………………………………….30 Dibenzothiophene biodesulfurization…………………………………………..31 Dibenzothiophene metabolic pathways……....……………………………32 4S pathway enzymatics…………………………………………………….38 4S pathway genetics………………………………………………………..41 Biodesulfurization application……………………………………………………….45 Oil biodesulfurization…………………………………………………………...47 Coal biodesulfurization………………………………………………………….49 Bottlenecks for biodesulfurization application…………………………………...50 Scope of the thesis…………………………………………………………………...51
Chapter 2
Desulfurization of dibenzothiophene, benzothiophene and ……..…..……………71 other thiophene analogues by a newly isolated bacterium Gordonia alkanivorans strain 1B
[Paper published – Appl. Biochem. Biotechnol. 2005, 120: 199-208]
17 CONTENTS
Chapter 3
Sequencing, cloning and expression of the dsz genes required……………….…… 89 for dibenzothiophene sulfone desulfurization from Gordonia alkanivorans strain 1B [Paper published – Enzyme Microb. Technol. 2007, 40: 1598-1603]
Chapter 4
Effect of zinc and other metal ions on the performance ……….…………………110 of dibenzothiophene desulfurization by Gordonia alkanivorans strain 1B [Paper submitted –J. Ind. Microbiol. Biotechnol.]
Chapter 5
Dibenzothiophene desulfurization by Gordonia alkanivorans ………...…...……126 strain 1B using recycled paper sludge hydrolyzate
[Paper submitted – Chemosphere]
Chapter 6
Global analysis and conclusions ………...….…………………….………...…149
18
Chapter 1
GENERAL INTRODUCTION
FOSSIL FUELS DESULFURIZATION
SULFUR IN FOSSIL FUELS
After food, fossil fuel is humanity’s most important source of energy and many of the benefits we enjoy from our way of life are due to fossil fuel use. Indeed, about 85% of our energy comes from fossil fuels, another 8% comes from nuclear power and 7% from all other sources, mostly hydroelectric power and wood. There are three major fuels - coal, oil, and natural gas. Oil leads with a share of about 40% of the total world consumption, followed by coal (24%) and natural gas (22%). Almost all fossil fuels are used by burning, which causes pollution producing waste products due to impurities in the fuel, especially particulates and various gases such as sulfur dioxide, nitrogen oxides and volatile organic compounds (Gupta et al., 2005).
Organic compounds containing sulfur (S) constitute an important fraction of fossil fuels and due to their low biodegradability are considered recalcitrant compounds. In crude oil the contaminant S is present in inorganic form, e.g. hydrogen sulfide and elemental sulfur (Kropp & Fedorak, 1998), although the organic forms are predominant with over 200 sulfur containing organic compounds (Lu et al., 1999). These compounds can be divided in three groups: (I) aliphatic and aromatic thiols, and its oxidation products (bisulfides); (II) aliphatic and aromatic thioethers; (III) heterocycles based in thiophenic ring: thiophene, benzothiophene (BT), dibenzothiophene (DBT) and its alkyl derivatives (Kropp & Fedorak, 1998).
Petroleum recovered from different reservoirs varies widely in compositional and physical properties (Van Hamme et al., 2003) containing between 0.04 and 5% (w/w) sulfur, and in general, crude oils of higher density contain a higher percent of sulfur (Kropp et al., 1997). The organosulfur compounds in petroleum include thiols, thioethers, and thiophenes, but the
21 GENERAL INTRODUCTION
sulfur compounds that predominate in the so-called heavy fractions, where sulfur content is the highest, are primarily the condensed thiophenes (Lu et al., 1999).
In some cases the sulfur content is very high, e.g., the crude oil in California and Utah at the USA and in Germany contain 5.5, 13.9 and 9.6% of sulfur, respectively (Kropp & Fedorak, 1998). Sulfur in gasoline is mainly found in thiophenic and non-thiophenic compounds, and in diesel oil is found in benzo and dibenzothiophenes (Oldfield et al., 1998).
Total sulfur content varies among ranks of coal from 0.3% to 6%; on average, organic and inorganic sulfur comprise equal amounts (Wang & Krawiec, 1994), although there are some exceptions (Fairbairn & Bushell, 1992). The inorganic sulfur in coal is, predominantly, in the form of pyrite while organic sulfur is present in several forms; the principal moieties are thiols, sulfides, disulfides, and thiophenes (Wang & Krawiek, 1994).
THE SULFUR PROBLEMATICS
The growth of industrial civilization, in particular the use of fossil fuels for energy, has led to the environmental pollution with a range of compounds of non-biological origin. This concern will become crucial at least partially owing to the decreasing availability of low- sulfur fuels (Konishi et al., 1997).
In 1990 about 65 million barrels of oil were produced each day, with an average sulfur content of about 1.1% (Monticello, 1998). Since then the oil production has risen (Fig. 1) and in 2005 about 85 million barrels were produced each day (Energy Information Administration’s primary report of recent international petroleum statistics, USA). There is no indication that this trend will slow down any time in the near future. Recent estimates of the worldwide reserves of fossil fuels indicate that the proven reserves of natural gas, crude oil and coal are sufficient to continue at this rate for at least the next 70 years (Kerr, 2000).
22 GENERAL INTRODUCTION
90
85
80 75 70 65 60
55 50 45 Oil productionOil (million barrels/day) 40
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Year
Fig. 1. World oil production during the last 25 years. Data were obtained from the Energy Information Administration’s primary report of recent international petroleum statistics (USA).
Most of the hydrocarbons mined from the Earth are burned for energy and since most liquid and solid (i.e., oil and coal) reserves are contaminated with sulfur, direct combustion of this fuel will lead to the release of vast amounts of sulfur oxides into the atmosphere (Monticello, 2000). These oxides (together with acidic nitrogen oxides) are responsible for poor air-quality, acid rain (Oldfield et al., 1998) and for ozone layer depletion (Malik et al., 2001b).
In addition, SO2 is responsible for various health hazards, such as respiratory tract cancer and cardio-respiratory diseases (Malik et al., 2001b). Concentrations of SO2 higher than 100 ppm in the atmosphere are harmful to the respiratory system of humans and a short-period exposure to 400-500 ppm is lethal (Schmidt et al., 1973). The combination of SO2 with dust in the atmosphere or with fog increases the noxious effect. The plant kingdom is also very sensitive to the SO2 concentration; exposure to 1-2 ppm SO2 provokes damages in few hours (Schmidt et al., 1973).
23 GENERAL INTRODUCTION
In oil spill accidents, some sulfur heterocyclic compounds are introduced in the environment. Some of these compounds (e.g. benzothiophene and its derivatives) present mutagenic and carcinogenic activities and acute toxicity to the organisms living in that ecosystem. Condensed thiophenes are bioacumulated in organism tissues, which associated with their mutagenic, carcinogenic and toxic potential, significantly contribute to the negative impact in the oil spills (Kropp & Fedorak, 1998).
REGULATIONS
During the last decade a drastic reduction in the allowed sulfur concentration in transportation fuels was observed. In Portugal this sulfur reduction has also been observed. Considering the values for diesel, in 1995 the maximum concentration of sulfur allowed was 2000 ppm, decreasing to 500 ppm in 1996 (Portaria nº 949/94 25th October). In 2001 this value was set at 350 ppm (Decreto-Lei nº 104/2000 3rd June) and nowadays is 50 ppm (Decreto-Lei nº 235/2004 16th December). Considering the values for gasoline, in 1995 the maximum concentration of sulfur allowed was 1000 ppm for leaded gasoline and 500 ppm for unleaded gasoline (Portaria nº 1489/95 29th December). In 2002 this value was set at 150 ppm (Decreto-Lei nº 104/2000 3rd June) and nowadays is 50 ppm (Decreto-Lei nº 235/2004 16th December).
In the near future, several countries along the world only allow the use of transportation fuels with the maximum concentration of sulfur of 15 ppm and with a target value of 10 ppm in 2010 (Swaty, 2005; Rashtchi et al., 2006). In Portugal this value was regulated to be 10 ppm in January 2009 (Decreto-Lei nº 235/2004 16th December).
HYDRODESULFURIZATION
The environmental restrictions in the industrialized countries require the use of fossil fuels with low sulfur content. However, the supply of low sulfur crude oils is being exhausted and
24 GENERAL INTRODUCTION
consequently they rule a higher price on the market than higher sulfur crudes (Gray et al., 1996). Thus, one of the strategies to reduce the emission levels to the atmosphere is to remove the sulfur from fossil fuels before their combustion.
Refineries remove organic sulfur from crude oil-derived fuels by hydrodesulfurization (HDS). HDS is a catalytic process that converts organic sulfur to hydrogen sulfide gas by reacting crude oil fractions with hydrogen at pressures between 150 and 3,000 psi and temperatures between 290 and 455 °C in the presence of metal catalysts (Grossman et al.,
1999). The metal catalysts mainly used are CoMo/Al2O3 and sometimes NiMo/Al2O3. At these conditions, sulfur concentration can be reduced from 1-5% to 0.1% (Izumi et al., 1994).
Although HDS can easily remove the inorganic sulfur or simple organic sulfur compounds, it is not effective for removing complex polycyclic sulfur compounds (Ohshiro & Izumi, 1999; del Olmo et al., 2005). Molecules such as 4,6-dimethyldibenzothiophene (4,6- dmDBT), with alkyl groups adjacent to the sulfur atom, are often used as model molecules in deep HDS studies because they are very difficult to desulfurize and create problems in deep HDS (Niquille-Röthlisberger & Prins, 2006). Indeed, it was shown that DBT and its alkylated derivatives still remain in HDS treated gas oil (Onaka et al., 2001a).
The imposition of increasingly stringent restrictions on the levels of sulfur allowed in transportation fuels implies that more recalcitrant compounds to the HDS process must be removed. To achieve this goal it is necessary to increase the intensity of physical-chemical treatment and inherently its costs (Grossman et al., 2001). Deep hydrodesulfurization technology must be implemented to attain this low level of sulfur (Prins et al., 2006). This technology involves much higher pressures, temperatures and residence times (Reichmuth et al., 2000), requiring the use of sophisticated and expensive catalysts (Klein et al., 1999). In these conditions, the process to remove the organic sulfur is too expensive (McFarland, 1999). In addition, increasing the severity of HDS also elicits undesirable effects on fuel quality, as other chemical components are reduced at the higher temperatures and pressures
25 GENERAL INTRODUCTION
needed to achieve low sulfur levels (Folson et al., 1999), thereby reducing the octane number of the fuel (Reichmuth et al., 2000).
In general, the distribution of sulfur in crude oil is such that the proportion of sulfur increases along with the boiling point of the distillate fraction. As a result, the higher the boiling range of the fuel the higher the sulfur content will tend to be. For example, a middle- distillate-range fraction, e.g. diesel fuel, will typically have higher sulfur content than the lower-boiling-range gasoline fraction (Grossman et al., 1999).
Organic sulfur compounds in the lower-boiling fractions of petroleum, e.g. the gasoline range, are mainly thiols, sulfides, and thiophenes, which are readily removed by HDS. However, middle-distillate fractions, e.g. the diesel and fuel oil range, contain significant amounts of benzothiophenes and dibenzothiophenes, which are considerably more difficult to remove by HDS (Grossman et al., 2001). Moreover, 4- and 6-substituents in DBT extremely retard the desulfurization reaction because of its steric hindrance. Probably, these substituents located near sulfur in DBT will inhibit the adsorption of this atom to the catalyst surface (Kabe et al., 1992). Due to their resistance to HDS, these compounds represent a significant barrier to reaching very low sulfur levels in middle- and heavy- distillate-range fuels (Grossman et al., 2001).
The governments and the petroleum refining companies have recognized that it is difficult to meet the environmental regulations in a cost-effective way by the conventional hydrodesulfurization technology because hydrodesulfurization units for the high extent of desulfurization are extremely expensive to build and operate (Ohshiro & Izumi, 1999). In fact, attempts are underway worldover to bring down the severity of refining operations through the development of milder physical and chemical processes (Bhatia & Sharma, 2006). For these reasons, it is of great importance to study new desulfurization processes that can substitute or complement HDS. Biological desulfurization might be one of those technologies to be implemented by oil industry.
26 GENERAL INTRODUCTION
BIODESULFURIZATION
Biodesulfurization (BDS) constitutes one possibility to be implemented instead of HDS process, due to the specificity of microorganisms to eliminate sulfur from HDS recalcitrant compounds (del Olmo et al., 2005). Biocatalytic processes are noted for their mild operating conditions, greater reaction specificity afforded by the nature of biocatalysis (Kaufman et al., 1998) and for not requiring molecular hydrogen (Reichmuth et al., 2000), allowing the maintenance of CO2 emissions at a low level (Ishii et al., 2000b). For these reasons, the sulfur removal by biocatalytic processes is considered an alternative or a complementary step for the conventional HDS process used in the refining industry of fossil fuels.
The first biodesulfurization studies were reported in the 50s and 60s, but without significant results. Only in the last 15 years BDS studies presented a development that allows considering a future application of microorganisms to desulfurize fossil fuels. Biological desulfurization of petroleum may occur either oxidatively or reductively. In the oxidative approach, organic sulfur is converted to sulfate, and this may be removed in process water. This route is attractive because it does not require further processing of the sulfur. In the reductive desulfurization scheme, organic sulfur is converted into hydrogen sulfide, which may then be catalytically converted into elemental sulfur; this is also an approach of utility at the refinery (Kaufman et al., 1998).
Oil and coal are complex substrates and the fact that a microorganism can metabolize DBT in laboratorial conditions does not necessarily imply that it can remove the organic sulfur from the fossil fuels. It is necessary to take into account the problem of microbial sulfur compounds accessibility in addition to the existence of steric hindrance associated to the structure of these compounds that difficult the activity of microbial enzymatic systems. A fundamental aspect to the BDS application is the possibility of keeping intact the carbon structure of the fossil fuel. Thus, the selected microorganisms must have the capability to use DBT as sulfur source but not as carbon source.
27 GENERAL INTRODUCTION
DESULFURIZING MICROORGANISMS
Several procaryotes presenting the ability to utilize sulfur from poliaromatic hydrocarbon compounds have been described: Acinetobacter sp. (Malik, 1978), Agrobacterium sp. (Constanti et al., 1996), Arthrobacter sp. (Dahlberg et al., 1993; Lee et al., 1995), Bacillus sp. (Kirimura et al., 2001), Beijerinckia sp. (Laborde & Gibson, 1977), Brevibacterium sp. (Van Afferden et al., 1990), Corynebacterium sp. (Omori et al., 1992), Desulfomicrobium sp. (Onadera-Yamada et al., 2001), Desulfovibrio sp. (Kim et al., 1990; Yeong et al., 1990; Sohrabi et al., 2006), Gordonia sp. (Rhee et al., 1998; Santos et al., 2006; Jia et al. 2006), Klebsiella sp. (Dudley & Frost, 1994), Mycobacterium sp. (Furuya et al., 2001; Takada et al., 2005; Li et al., 2007), Nocardia sp. (Wang & Krawiec, 1994), Paenibacillus sp. (Konishi et al., 1997), Pseudomonas sp. (Setti et al., 1992; De Fatima et al., 1996; Luo et al., 2003), Rhodococcus sp. (Denome et al., 1993a; Izumi et al., 1994; Ohshiro et al., 1994: Lee et al., 1995; Matsui et al., 2002), Rhizobium sp. (Malik, 1978; Frassinetti et al., 1998), Sinorhizobium sp. (Tanaka et al., 2001), Sphingomonas sp. (Darzins & Mrachko, 2000; Gai et al., 2007), Sulfolobus sp. (Kargi, 1987; Ju & Kankipati, 1998), Xanthomonas sp. (Constanti et al., 1996). The majority of the BDS studies involve aerobic microorganisms despite some anaerobic bacteria have also been described. However, the low rate and extent of petroleum desulfurization by currently available anaerobic cultures and the lack of knowledge on the biochemistry and genetics of such microorganisms makes the development of a commercial anaerobic process unlikely (Kilbane, 2006).
The genus Gordonia
The first report describing this genus in biodesulfurization studies was the work involving Gordona sp. strain 213E (Gilbert et al., 1998). The genus Gordona was proposed by Tsukamura (1971), but the three original species of the genus were subsequently reclassified in the genus Rhodococcus (Goodfellow & Alderson, 1977). However, Rhodococcus species could be divided into these two aggregate groups by serological and chemotaxonomic
28 GENERAL INTRODUCTION
properties, such as mycolic acid composition and menaquinone profiles. Stackebrandt et al. (1988) found that the two aggregate groups are phylogenetically distinct on the basis of 16S rRNA sequences and revived the genus Gordona.
Nowadays, the name Gordonia, instead of Gordona, is being recognized because it is etymologically correct (Stackebrandt et al., 1997). The members of the genus Gordonia are widely distributed in nature (Takeuchi & Hatano, 1988). Other Gordonia species have also been isolated from activated sludge in aeration tanks of biological sewage-treatment plants (Lemmer & Kroppenstedt, 1984) and from the packing material of a biofilter used for biological odor abatement (Bendinger et al., 1995; Klatte et al., 1996). Nevertheless, little is known about Gordonia species or strains that degrade toxic aromatic compounds, unlike the genus Rhodococcus, a phylogenetic neighbor which is a very important taxon from the point of view of bioremediation (Finnerty, 1992).
MODEL COMPOUNDS BIODESULFURIZATION
Crude oil can contain 50% of organic sulfur in the form of condensed thiophenes (Frassinetti et al., 1998) that after refining at high temperature lead to DBT concentrations higher than 70% of total sulfur (Kropp & fedorak, 1998). The straight-run middle-distillate feed stock contains 95% of the organosulfur compounds as thiophenic compounds, that include thiophene, BT and DBT. Each of these basic aromatic structures can have a variety of alkyl substituents, thereby increasing the number of unique compounds (Folsom et al., 1999). Furthermore, these compounds are recalcitrant and persisting in biosphere, which are released in the environment through industrial processes, including gasification and liquefaction of coal and refining of crude oil, and through oil spilling accidents (Berthou & Vignier, 1986).
Alkyl-substituted DBTs seem to be the most difficult to remove among all organosulfur compounds, with certain isomers surviving even to deep HDS treatment (Amorelli et al.,
29 GENERAL INTRODUCTION
1992). DBT can be regarded as a model for the main aromatic organosulfur nucleus of the coal matrix (Oldfield et al., 1998).
In this context, for more than 4 decades, dibenzothiophene has been the model compound for the biodegradation or biodesulfurization studies of sulfur heterocycles (Bressler & Fedorak, 2001). In the majority of the reported works on biodesulfurization, DBT and other analogous compounds with alkyl groups are utilized as carbon and energy source or as sulfur source (Monticello et al., 1985). Another advantage to use this compound as model is the fact that it does not present any mutagenic potential (Kropp & Fedorak, 1998).
Although DBT is the preferential model compound, some reports refer benzothiophene in BDS studies (Matsui et al., 2001). The importance of BT as model compound is relevant due to the fact that it presents a different metabolic pathway.
BENZOTHIOPHENE BIODESULFURIZATION
Studies on the carbon-sulfur bond-targeted type of BT desulfurization have only been performed in some bacteria: Desulfovibrio desulfuricans (Setti et al., 1993), Gordonia sp. 213E (Gilbert et al. 1998), Paenibacillus sp. A11-2 (Konishi et al. 2000a), Rhodococcus sp. T09 (Matsui et al. 2000) and Sinorhizobium sp. KT55 (Tanaka et al. 2001).
The pathway for BT desulfurization by Paenibacillus sp. A11-2 (Konishi et al., 2000a) and Sinorhizobium sp. KT55 (Tanaka et al., 2001) is BT → BT sulfoxide → BT sulfone → benzo[e][1,2]oxathiin S-oxide → o-hydroxystyrene (Fig. 2). BT suffers a double oxidation of the sulfur atom forming the corresponding sulfoxide and sulfone. Conversion of BT S,S- dioxide into benzo[e][1,2]oxathiin S-oxide can be explained by oxidative cleavage of one of the two C-S bonds in BT S,S-dioxide and circularization of the cleavage product under acidic conditions. In the final step, benzo[e][1,2]oxathiin S-oxide seems to lose its sulfur by the desulfination reaction (Konishi et al., 2000a).
30 GENERAL INTRODUCTION
Benzothiophene S
S
O
Benzothiophene sulfone S
O O
Benzo[e][1,2]oxathiin S- (F S oxide O O
o-hydroxystyrene
OH
Fig. 2. The postulated pathway of benzothiophene desulfurization by Paenibacillus sp. strain A11-2 (adapted from Konishi et al., 2000a).
DIBENZOTHIOPHENE BIODESULFURIZATION
Despite the obvious chemical similarity between DBT and BT, the two desulfurization pathways are mutually exclusive. Thus BT cannot be desulfurized via the DBT-specific pathway and DBT cannot be desulfurized through the BT-specific pathway (Gilbert et al. 1998). Indeed, cells of R. erythropolis IGTS8 pre-grown in BT cannot desulfurize DBT. This suggests that the enzymatic system of BT desulfurization could be different from DBT desulfurization system. However, it was reported that the first step of BT desulfurization (sulfur oxidation) could be catalyzed by the same monooxygenase involved in DBT desulfurization (Kobayashi et al., 2000). Therefore, the organisms having these metabolic pathways are complementary in terms of their potential roles in development of a microbial fuel desulfurization technology (Oldfield et al., 1998).
31 GENERAL INTRODUCTION
It has been reported that DBT is aerobically or anaerobically metabolized by microorganisms. Desulfovibrio desulfuricans M6, a sulfate-reducing bacterium, degraded DBT anaerobically, and biphenyl was isolated as the major degradation product (Kim et al., 1990). An anaerobic process for sulfur removal will be attractive because it does not liberate sulfate as a by-product that must be disposed by some appropriate treatment. However, anaerobic microorganisms effective enough for the practical petroleum desulfurization have not been found yet (Ohshiro & Izumi, 1999).
There are many reports concerning aerobic DBT metabolism. DBT desulfurization occurs inside the cell with its entrance in the cytoplasm possibly from the organic phase, after transient adsorption (Monticello, 1998). However, Gallardo et al. (1997) reported that DBT is present in the aqueous phase before its entrance into the cell. The solubility of DBT in water is very low, about 0.005 mM, although it can be increased with the surfactants produced by the cells.
Increased rates of DBT desulfurization in higher hydrocarbon fractions were reported, and this might suggest transfer of DBT through the interface between the aqueous and hydrocarbon phase or adsorption of cells at the interface (Maghsoudi et al., 2001). On the other hand, the presence of a significant amount of bacterial cells in the organic phase and in the water/organic interphase in 1-10 µl droplets during desulfurization of DBT in high hexadecane concentrations was reported (Kaufman et al., 1998), presumably due to biosurfactant production, which will have impact on separations in a commercial process (McFarland, 1999).
Dibenzothiophene metabolic pathways
To current knowledge, DBT metabolism by aerobic microorganisms can be divided in 3 different pathways: Kodama pathway (Kodama et al., 1973), Van Afferden pathway (Van Afferden et al., 1990) and 4S pathway (Denome et al., 1993a).
32 GENERAL INTRODUCTION
Kodama pathway
In 1973, Kodama and co-workers reported that DBT was partially degraded by strains of Pseudomonas sp. through some oxidations by a mechanism similar to the naphtalene degradation (Fig. 3). The dihydroxylation of a DBT aromatic ring causes the destruction of that ring, obtaining 3-hydroxy-2-formyl-benzothiophene as final product, in which the thiophenic nucleus containing the sulfur atom persists (Mormile & Atlas, 1988). This final product presents biological toxicity levels similar to the initial substrate (Gallagher et al., 1993).
Dibenzothiophene S
OH
OH
S
OH O COOH
S
OH 3-hydroxy-2-formyl- benzothiophene S CHO
Fig. 3. Kodama pathway – Dibenzothiophene is partially degraded by attack of a specific aromatic dioxygenase. The product, 3-hydroxy 2-formyl benzothiophene, which retains the sulfur moiety, is not further degraded (adapted from Kodama et al., 1973).
33 GENERAL INTRODUCTION
More recently, 3-hydroxy-2-formyl-benzothiophene was totally biodegraded and mineralized by a mixed bacterial culture (Bressler & Fedorak, 2001). This is the most widely used pathway by bacteria that can metabolize DBT (Kilbane & Jackowsky, 1992).
Since the DBT analogous compounds present in fossil fuels contain alkyl and aryl groups in those positions, this metabolic pathway cannot degrade them. Thus, this is a metabolic pathway without practical interest to biodesulfurization.
Van Afferden pathway
In 1990, Van Afferden et al. described a different metabolic pathway by Brevibacterium sp., in which DBT is converted, in stoichiometric quantities, to benzoate and sulfite (Fig. 4). DBT is initially oxidized to DBT sulfone (DBTS), followed by the opening of the thiophenic ring initiated by aromatic dioxygenase action, to yield 2,3-dihydroxybiphenyl 2’- sulfinate. A second round of dioxygenase attack opens the 2,3-dihydroxybenzene nucleus.
Benzoate is mineralized to CO2 and H2O, remaining in the final only 9% of the carbon from DBT as dissolved organic carbon in the medium (Van Afferden et al., 1990). DBT is used as nutrient either as carbon and sulfur sources. This DBT degradation pathway has only a partial interest in terms of BDS of fossil fuels, due to the complete mineralization of the carbon structure of the hydrocarbon. This necessarily provokes a decrease of the potential chemical energy of the fossil fuels.
Unfortunately, this pathway has not been studied in more detail and there is no further information concerning its enzymology or genetics. Although its less importance in terms of BDS, the bacteria using this metabolic pathway are potentially useful in the formulation of mixed microbial inocula for poliaromatic hydrocarbon bioremediation processes.
34 GENERAL INTRODUCTION
Dibenzothiofene S
S
O
Dibenzothiofene sulfone S
O O
S OH H OH O O
2,3-dihydroxybiphenyl 2’-sulfinate
SO2H OH OH
O
HOOC
SO2H OH
- - SO3 SO 4 COOH Benzoate
H2O + CO2
Fig. 4. Van Afferden Pathway – Following oxidation of dibenzothiophene to dibenzothiophene sulfone, thiophene ring-opening is triggered by the action of an angular dioxygenase. The product of this reaction is 2,3-dihydroxybiphenyl 2’- sulfinate. In a second round of dioxygenase attack, the catechol ring is opened and the product degraded to benzoate. Ultimately, benzoate is mineralized to CO2 and water (adapted from Van Afferden et al., 1993).
35 GENERAL INTRODUCTION
4S pathway
The third reported DBT desulfurization pathway is the usually denominated 4S pathway, due to the formation of 4 sulfur compounds during the metabolic pathway (sulfoxide- sulfone-sulfonate-sulfate) (Kilbane, 1989; Kilbane & Jackowski, 1992; Omori et al., 1992).
This pathway is specific for the removal of the sulfur atom present in DBT, in which the thiophenic group is progressively oxidized without degradation of the carbon structure. The enzymatic system involved in the desulfurization of DBT was reported for the first time by Ohshiro et al. (1994). 4S pathway is a multienzymatic system with four different activities (Gray et al., 1996) as depicted in Fig. 5. DBT desulfurization is an energetically expensive process, since it has been estimated to require 4 mol NADH per mol of DBT desulfurized (Oldfield et al., 1997).
The molecular oxygen is also an important factor to the activity of the first two enzymes of this pathway. Ohshiro et al. (1995) reported that DBT degradation was repressed in a reaction mixture with high concentration of cells, in which there is oxygen limitations. The final product (2-hydroxybiphenyl; 2-HBP) concentration is the main limiting factor to the DBT biodesulfurization (Omori et al., 1992). 2-HBP is the active ingredient in the disinfectant Lysol and, perhaps not surprisingly, was found to be toxic to Arthrobacter sp. strain ECRD-1 at a concentration of 50 mg l-1 when applied to freshly inoculated cultures (Lee et al., 1995). 2-HBP presents a strong inhibitory effect on the cells in the process of DBT desulfurization, due to the retroactive inhibition of the desulfurization enzymes (Nekodzuka et al., 1997). This inhibition was reported in a study using a Gordonia sp. strain CYKS1 in which the presence of 0.2 mM 2-HBP in the culture medium increased by 50% the cell duplication time (Rhee et al., 1998). Even a small concentration of 2-HBP decreased about 20% the desulfurization and cell growth rates (Kim et al., 2004).
36 GENERAL INTRODUCTION
Cx-DBT
NADH-FMN- reductase
Desulfinase
Fig. 5. A conceptual diagram of some of the steps in the desulfurization of oil by bacterial cells. DBT is converted to 2-HBP due to the activity of four enzymes. DBTO – dibenzothiophene sulfoxide; DBTO2 – dibenzothiophene sulfone; HPBS – 2-(2′- hydroxyphenyl) benzene sulfinate; DBT-MO – dibenzothiophene monooxygenase; DBTO2- MO – DBT sulfone monooxygenase (adapted from Monticello, 2000).
However, in bibliography is extensively reported that these effects are reduced in biphasic media because 2-HBP shows a really high affinity for organic solvents (Caro et al., 2007; Kobayashi et al., 2001; Yang et al., 2005).
Without reduction of DBT carbon content, the microorganisms that utilize 4S pathway are very promising for application in BDS processes (Rhee et al., 1998), namely bacteria belonging to the genera Arthrobacter, Brevibacterium, Gordonia, Paenibacillus and
37 GENERAL INTRODUCTION
Rhodococcus (Duarte et al., 2001; Mohebali et al., 2007), as well as Agrobacterium (Oldfield et al., 1998).
The enzymatic activities involved in DBT desulfurization are associated to the soluble fraction and none of them are associated to the membrane fraction (Gray et al., 1996), as opposed to the enzymes involved in the metabolism of other very hydrophobic molecules, such as alkanes (Monticello, 2000).
The microorganisms that use this pathway to metabolize DBT can release a potentially noxious atom from the thiophenic compound producing a biodegradable compound (sulfate), only with a slight reduction of the hydrocarbon calorific value (Wang & Krawiec, 1996).
In an in-vitro desulfurization assay, it was shown that the only intermediate accumulated to any extent was 2-(2-hydroxyphenyl) benzenesulfinate, being produced at a rate about five times faster than its consumption. Therefore, the last step in the pathway (catalyzed by the desulfinase) was rate limiting (Gray et al., 1996).
The original Rhodococcus IGTS8 desulfurizing enzymes have little activity towards thiophenes or benzothiophenes, and thus gasoline desulfurization will require new biocatalysts with improved efficiency toward thiophenic sulfur (McFarland, 1999). This enzyme system will act not only on DBT but also on thioxanthen-9-one, as well as on DBT derivatives, 4-methyl DBT, 2-ethyl DBT, 3-ethyl DBT, 3,4,6-trimethyl DBT, 3,4,6,7- tetramethyl DBT (Kobayashi et al., 2000), 4,6-dimethyl DBT, 2,8-dimethyl DBT and 3,4- benzo DBT (Ohshiro et al., 1997).
4S pathway enzymatics
The first enzyme of 4S pathway is a DBT monooxygenase (DszC) which catalyses the DBT oxidation to DBTS in two steps. The second is also a monooxygenase (DszA) which converts DBTS to 2-(2’-hydroxyphenyl) benzenesulfinate, and in the final step a liase
38 GENERAL INTRODUCTION
(DszB) converts it to 2-hydroxybiphenyl and sulfite. In this metabolic pathway, a FMN- reductase (DszD) has an important role in the activity of the monooxygenases, since it is responsible for the maintenance of reduced flavin levels (Fig. 5).
DBT monooxygenase
This enzyme is classified as unspecific monooxygenase (EC 1.14.14.1) by the Nomenclature Committee of the International Union of Biochemistry and Molecular Biology (NC- IUBMB). There is not yet a specific classification for this particular enzyme. This enzyme catalyzes the sequential conversion of the DBT to a DBT sulfone using FMNH2, as a co- substrate (Xi et al., 1997):
DBT → DBT-sulfoxide → DBT-sulfone
The first oxidation step (constant rate 0.06 min−1) occurs with one-tenth of the rate of the second step (constant rate 0.5 min−1). DBT monooxygenase from R. erythropolis D-1 is a homohexamer with a subunit molecular weight of 45 kDa. Its activity is maximal at a temperature of 40 °C and a pH of 8.0.
DBT monooxygenase was also purified from the thermophilic bacterium Paenibacillus sp. strain A11-2. The molecular mass of the purified enzyme and its subunit were determined to be 200 kDa and 43 kDa by gel filtration and sodium dodecyl sulfate polyacrylamide gel electrophoresis, respectively, indicating a tetrameric structure. The optimal temperature and pH for this enzyme are 65 ºC and pH 9 (Konishi et al., 2002).
DBT sulfone monooxygenase
The enzyme classification of DBT sulfone monooxygenase is the same than the DBT monooxygenase (EC 1.14.14.1). DBT sulfone monooxygenase is widely studied and has been purified from some bacterial strains (R. erythropolis IGTS8, R. erythropolis D-1, Paenibacillus sp. strain A11-2 and Bacillus subtilis WU-S2B). This monooxygenase,
39 GENERAL INTRODUCTION
present in strain D-1, has a molecular weight of 97 kDa consisting of two subunits with identical masses of 50 kDa (Ohshiro et al., 1999) and oxidizes DBT sulfone to 2-(2′- hydroxyphenyl) benzene sulfinate (HPBS) using FMNH2 as a co-substrate (Xi et al., 1997) with a reaction rate 5-10-fold higher than DszC (Gray et al., 1996).
Being thermophilic, the enzymes isolated from Paenibacillus sp. strain A11-2 (Konishi et al. 2000b) and Bacillus subtilis WU-S2B (Ohshiro et al., 2005) exhibit different characteristics from the enzyme isolated from the mesophiles. This thermophilic enzyme has an optimum temperature of 45 °C and is stable till 60 °C. It works at an optimum pH of 5.5.
2'-hydroxybiphenyl-2-sulfinate desulfinase
2'-hydroxybiphenyl-2-sulfinate desulfinase (EC 3.13.1.3) is the rate-limiting enzyme of 4S pathway and catalyzes the conversion of HPBS to 2-HBP. After cleavage, the sulfur product is incorporated into the biomass of the microorganism (Watkins et al., 2003). It is the least studied enzyme as it is produced in very small amounts (Gupta et al., 2005). This enzyme was purified from R. erythropolis IGTS8 and it consists of a monomer with a subunit molecular weight of 40 kDa, showing enzyme activity over a wide temperature range (25– 50 °C), with the optimum at 35 °C (Watkins et al., 2003). The working pH range for this enzyme is 6.0–7.5, with the optimum at pH 7.0. A cysteine residue is shown to be present in the sequence of this enzyme and it is found to be critical for enzyme activity (Nakayama et al., 2002).
This enzyme was also purified and characterized from thermophilic bacteria, such as Paenibacillus sp. strain A11-2 (Konishi & Maruhashi, 2003). The molecular mass of the purified enzyme is 39 kDa, with a monomeric structure. The optimal temperature and pH for the reaction involving this enzyme are 55 ºC and 8, respectively.
40 GENERAL INTRODUCTION
Flavin reductase
Flavin reductase (EC 1.5.1.30), also named NADH-FMN oxidoreductase, is associated to the supply of free FMNH2 needed to 4S monooxygenases activity. All the monooxygenases of 4S pathway showed a requirement for equimolar quantities of flavin reductase for their respective oxygenation reactions. Xi et al. (1997) reported an enhanced desulfurization activity of DBT monooxygenase and DBT sulfone monooxygenase, under in vitro conditions, with increased concentrations of flavin reductase.
The purified flavin reductase of R. erythropolis D1 contains no chromogenic cofactors and it was found to have a molecular mass of 86 kDa and four identical 22-kDa subunits. The optimal temperature and pH for enzyme activity were 35 °C and 6.0, respectively, and the enzyme retained 30% of its activity after heat treatment at 80 °C for 30 min (Matsubara et al., 2001).
The flavin reductase purified from the thermophilic strain Paenibacillus sp. strain A11-2 (Konishi et al. 2000b) was also characterized. It is a homodimer with a subunit molecular weight of 25 kDa. This enzyme was completely FMN-dependent, and FAD could not act as a cofactor. This enzyme has the optimal temperature and pH at 55 °C and 5.5, respectively.
4S pathway genetics
Although the sox (sulfur oxidation) designation was first used, the dsz (desulfurization) designation for the 4S pathway genes has generally been adopted because several other unrelated genes were already labeled sox. To avoid misunderstanding with those other genes, the dsz designation (Dsz for gene product) has generally been accepted (Gupta et al., 2005).
Rhodococcus erythopolis IGTS8 and Rhodococcus sp. strain X309 were among the first strains to be characterized at the molecular level (McFarland, 1999). There are indications
41 GENERAL INTRODUCTION
of the conservative nature of the dsz genotype and of the establishment of differences and similarities among desulfurization strains isolated from different geographic locations (Denis-Larose et al., 1997). Complete sequence identity of the dsz operon was observed between Arthrobacter sp. DS7 and Rhodococcus sp. IGTS8 (Serbolisca et al., 1999). However, evidence for the presence of bacteria with divergent dszA gene sequences in an oil-polluted soil was reported (Duarte et al., 2001).
The work performed by Denome et al. (1993b) indicates that a single genetic pathway controls the metabolism of dibenzothiophene, naphthalene and phenanthrene in Pseudomonas sp. strain C18. The majority of genetic studies involving desulfurization genes were performed with Rhodococcus strains and, therefore, it is important to increase the knowledge of regulatory mechanisms for other bacterial genus.
In this context, some works were more recently reported using the bacteria Bacillus subtilis (Kirimura et al., 2004), Burkholderia sp. (Di Gregorio et al., 2004), Gordonia nitida (Park et al., 2003), Mycobacterium sp. G3 (Nomura et al., 2005; Takada et al., 2005), Mycobacterium phlei (Kirimura et al., 2004) and Paenibacillus sp. A11-2 (Ishii et al., 2000a). dsz operon
The dsz genes of Rhodococcus sp. are arranged in an operon-regulated system located on a circular plasmid of approximately 150 kb (Oldfield et al., 1998), 120 kb (Santos et al., 2007) or on a 100-kb plasmid in other strains (Gupta et al., 2005). In this plasmid-encoded pathway, three genes (dszABC) arranged in an operonic manner and spanning a 4-kb region are responsible for the metabolism of DBT to 2-HBP and sulfate (Denis-Larose et al., 1997), though some differences occur at gene expression level (Li et al., 1996). It is transcribed in the same direction, coding for three proteins DszA, DszB, DszC, respectively (Piddington et al., 1995). The termination codon for dszA and the initiation codon for dszB overlap 2-bp and there is no such overlap between dszB and dszC (Oldfield et al., 1998),
42 GENERAL INTRODUCTION
existing a 13-bp gap (Piddington et al., 1995). Although expressed as an operon, DszB is present at concentrations several-fold less in the cytoplasm, as compared with DszA and DszC (Li et al. 1996). These genes, when cloned on a Dsz negative phenotype, confer the ability to desulfurize DBT to 2-HBP.
The evidences that support the hypothesis that the cluster is expressed as an operon are: first, disabling or removing the promoter region prevented expression of all measurable enzymatic activities; second, replacing the promoter region with alternative promoters relieved the sulfur repression normally observed at each step of desulfurization; and third, replacing the native promoter region with the E. coli tac promoter allowed expression of DBT desulfurization in E. coli (Piddington et al., 1995).
Analogous to the dsz operon in mesophiles, the tds (thermal desulfurization) operon is located on a 8.7-kb DNA fragment in the thermophile Paenibacillus sp. A11-2 (Ishii et al. 2000a; Konishi et al. 2000a). The tdsA, tdsB, and tdsC nucleotide sequences and the deduced amino acid sequences showed significant homology to the dszA, dszB and dszC genes of R. erythropolis IGTS8. Several gram-positive and gram-negative organisms are known to have desulfurization genes; and they show 70% homology (McFarland 1999). dsz operon regulation
Promoter and regulatory regions of the dsz operon were also studied and it was found that synthesis of enzymes is strongly repressed in the presence of readily bioavailable sulfur (Li et al. 1996), i.e., sulfate, sulfide, methionine and cysteine. The promoter for the dsz gene cluster in strain IGTS8 is located in a 385-bp region immediately upstream of dszA. This 385 bp of 5’ untranslated dsz DNA contains a number of interesting elements; it appears to contain a promoter and at least three regions that affect Dsz activity. The first, at -263 to - 244, reduced repression, but deletions did not affect repression or gene expression. The second region, between -144 and -121, was able to bind a protein that could be an activator; deletion of this region reduced gene expression but not repression. The third region,
43 GENERAL INTRODUCTION
between -57 and -98, may be a repressor binding site that overlaps the promoter -10 and -35 regions (Li et al., 1996). To date, no repressor or activator proteins that act upon the regulatory sequence of dsz have been isolated (Reichmuth et al., 2000). The three genes are transcribed in the same direction beginning in the position -46 bp.
The enzyme production is induced by DBT and its analogs, DBT sulfone or thioxanthen 9- one, and it is strongly repressed by sulfate or the compounds with sulfur, such as methionine and cysteine, even in the presence of DBT (Kayser et al., 1993; Ohshiro et al., 1996). Site- directed mutagenesis will be a better way to construct mutants with full promoter activity and no repression (Ohshiro & Izumi, 1999).
Gene expression
Many environmental pollutants are readily degraded by naturally occurring microbes. Very often, however, the rate of degradation may not be optimal for practical large-scale bioremediation. Genetic engineering of biodegradative pathways offers the potential of expanding the existing capabilities found in nature. Expression of biocatalytic pathways in foreign microorganisms can significantly enhance the efficiency of the biodegradation process (Chen et al., 1999).
This is the case of the dsz operon expression in Pseudomonas aeruginosa EGSOX that, in addition to the ability to produce a biosurfactant that increase the aqueous concentrations of hydrophobic compounds, allows DBT desulfurization 4-fold faster than the wild strain of R. erythropolis (Gallardo et al., 1997). It has been shown (Rambosek et al. 1999) that the reconstruction of several promoters containing the R. erythropolis IGTS8 dsz genes, with the dszD gene, helps to increase the rate of DBT desulfurization. The same conclusion was reported in the work involving R. erythropolis KA2-5-1 (Konishi et al., 2005a). In addition, when flavin reductase, flavin mononucleotide reductase or various oxidoreductases were over-expressed in recombinant constructs, the desulfurization rate increased up to 100-fold (Gray et al., 1996).
44 GENERAL INTRODUCTION
BIODESULFURIZATION APPLICATION
An industrial-scale process for petroleum biodesulfurization using aerobic microorganisms has not yet been demonstrated. However, through an improved understanding of the biochemistry and genetics of the desulfurization pathway, it is anticipated that improved biocatalysts with activities suitable for an industrial process will be developed (Kilbane, 2006).
Until the present date, studies on sulfur oxidative pathways have mainly been focused in model compounds, which limit the ability to demonstrate the commercial potential of BDS (Grossman et al., 2001). However, some reported works involved several fractions of crude oil refining, including gasoline and diesel (Rhee et al., 1998; Folsom et al., 1999; Grossman et al., 1999; Grossman et al., 2001; Furuya et al., 2003; Guobin et al., 2005; Li et al., 2005). Efforts to increase the rate of sulfur removal from aromatic sulfur heterocycles have been possible due to the use of genetic engineering techniques or the use of immobilization matrices (Gupta et al., 2005).
The selection of the petroleum feedstock in biodesulfurization will play a large role in the overall economic viability of the process. BDS may be utilized as a pretreatment to crude oil before entering pipelines. It may also be applied as an alternative to hydrotreating the crude at the refinery or it may be applied in the polishing of refinery products such as diesel or gasoline. As pretreatment, the BDS unit may be used to treat marginally sour crudes (0.6- 0.7% S), converting them to sweet crudes (<0.5% S). For this application, the extent of desired desulfurization is quite low, and this may serve as an attractive initial niche for BDS (Kaufman et al., 1998).
Inherent to all of the current bioprocessing of fossil feedstocks schemes is the need to contact a biocatalyst-containing aqueous phase with an immiscible or partially miscible organic substrate (Van Hamme et al., 2003). Factors such as liquid-liquid and gas-liquid mass transport, amenability for continuous operation and high throughput, capital and
45 GENERAL INTRODUCTION
operating costs, as well as ability for biocatalyst recovery and emulsion breaking, are significant issues in the selection of a reactor for aqueous-organic contacting (Kaufman et al., 1998). Biodesulfurization studies of fossil fuels usually involved intact cells as biocatalyst, which avoids the Dsz enzymes purification and facilitate the BDS industrialization. The immobilization of cells can be used to desulfurize DBT efficiently (Shan et al., 2005) being the life-time of immobilized cell biocatalysts more than 600 h (Hou et al., 2005).
Traditionally, impeller-based stirred reactors are used for such mixing, because of their ease of operation and wide acceptance in the chemical and biological processing industries (Fig. 6).
Fig. 6. A conceptual process flow diagram for the biodesulfurization process (adapted from Monticello, 1998).
46 GENERAL INTRODUCTION
This kind of reactors promotes the contact between the aqueous and organic phases by imparting energy to the entire bulk solution, achieving water or oil droplet sizes of 100-300 µm in diameter when surfactants are not present. To obtain droplets of about 5 µm, the energy consumed by the reactor will be 5-fold higher (Kaufman et al., 1997). In BDS processes, oil is mixed with an aqueous medium that contains biocatalytically active bacterial cells. Recovery of oil from the oil–water–bacteria mixture follows the biodesulfurization step as a separated batch process (Konishi et al., 2005b).
To date, there are some microbial desulfurization studies at laboratory scale involving petroleum fractions and coal. Energy BioSystems Corporation (ECB) was the only commercial venture dedicated to the development of biodesulfurization technology (Kilbane, 2006). EBC’s concept for a biodesulfurization process was not only to treat diesel, but also to produce a value-added surfactant byproduct to achieve a more economical process (Monticello, 2000).
There was a plan to construct a demonstration-scale biodesulfurization process at the Petro Star refinery in Valdez, Alaska. The date for the construction of a demonstration plant was progressively postponed (Kilbane, 2006).
OIL BIODESULFURIZATION
Early biodesulfurization research used model compounds like DBT, sometimes in aqueous systems bearing little resemblance to the conditions the biocatalyst would encounter in commercial applications (Ohshiro & Izumi, 1999). In fact, the desulfurization rates of diesel oil were much smaller than those obtained for pure DBT (Rhee et al., 1998). Biodesulfurization of petroleum results in total sulfur removals between 30 and 70% for mid-distillates (Grossman et al., 1999), 24 to 78% for hydrotreated diesel (Rhee et al., 1998; Maghsoudi et al., 2001; Ma et al., 2006), 20-60% for light gas oils (Chang et al., 1998; Ishii et al., 2005) and 25-60% for crude oils (McFarland, 1999).
47 GENERAL INTRODUCTION
Taking into account that BDS can be a complementary method to HDS, the study of fractions of pre-desulfurized oil is important. Grossman et al. (2001) reported a treatment by Rhodococcus sp. strain ECRD-1 of middle distillate oil whose sulfur content was virtually all substituted DBTs containing 669 ppm of total sulfur. Analysis of the sulfur content of the treated oil revealed that 92% of the sulfur had been removed, reducing the sulfur content from 669 ppm to 56 ppm.
In addition, studies of desulfurization with Rhodococcus erythropolis I-19, involving hydrodesulfurized middle distillate oil, showed that after 0, 1, 3 and 6 h, the sulfur concentrations were 1850, 1620, 1314 and 949 ppm, respectively. The first 230-ppm drop in total sulfur, observed after 1 h, corresponded primarily to a biotransformation of DBT and midboiling- range sulfur compounds. Between 1 and 3 h, another 300-ppm sulfur reduction occurred, with some evidence for more highly alkylated DBTs being affected. At 3 h, most of the DBT and much of the C1-DBTs were consumed. Between 3 and 6 h, desulfurization shifted to the higher-boiling-range sulfur compounds, resulting in an additional 365-ppm drop in total sulfur. Analysis of this middle distillate oil biodesulfurized from 1850 to 615 ppm sulfur showed the majority of the remaining sulfur to be thiophenes (75%), with 11% sulfides, 2% sulfoxides and 12% sulfones (Folsom et al., 1999). More recently, Zhang et al. (2007) reported a total reduction of 97% (to 6 µg ml–1) of the sulfur content of previous hidrodesulfurized diesel oil.
There are also some studies on desulfurization of oil fractions involving thermophilic bacteria such as Paenibacillus sp. (Konishi et al., 1997). When Paenibacillus sp. strains A11-1 and A11-2 were cultured in the presence of light gas oil containing 800 ppm of total sulfur at a high temperature, the bacteria grew well. Light gas oil is known to contain small amounts of sulfur and limited species of heterocyclic organosulfur compounds composed mainly of alkylated DBT derivatives. In conformity with the stimulated bacterial growth, the content of sulfur in the oil phase was significantly decreased, indicating that both Paenibacillus strains can desulfurize at high temperatures from the processed light gas oil. However, these strains presented a very low desulfurization rate (Onaka et al., 2001b),
48 GENERAL INTRODUCTION
lower than the desulfurization rate obtained with R. rhodochrous (Konishi et al., 1997). The use of thermophilic bacteria has some advantages since it is not necessary to cool-down the oil fractions after the HDS, which makes this process less expensive (Konishi et al., 2000b). Another advantage is the fact of reducing the possibility of contamination by undesirable bacteria that can negatively affect the BDS process (Kirimura et al., 2001).
Although the obtained removals are significant, this level of desulfurization is insufficient to meet the required sulfur levels for all oil derivatives (Grossman et al., 1999).
COAL BIODESULFURIZATION
The isolation of mesophilic bacterial strains belonging to the genera Chryseomonas, Moraxella, Pseudomonas, Xanthomonas (Gómez et al., 1999), Leptospirillum and Thiobacillus, and thermophilic prokaryotic strains, as Acidianus brierleyi, Metallosphaera sedula, Sulfolobus acidocaldarius and Thiobacillus caldus (Kargi & Weissman, 1987; Schippers et al., 1999), allowed, in general, a high efficiency of metabolization of sulfur present in coal, mainly pyritic sulfur.
Thiobacillus ferrooxidans is the most widely used organism for biodesulfurization of coal (Olson & Kelly, 1986). It is an aerobic chemoautotrophic bacterium which derives its metabolic energy from oxidation of reduced iron and sulfur components of the pyrite (Lees et al., 1996).
The effect of particle size, pulp density, cell density and various other process parameters on biodesulfurization process of coal has been extensively studied (Malik et al., 2001b). Bioleaching is a complex phenomenon governed by a chain of reactions representing direct (bacterial) and indirect attack on sulfide (Schippers et al., 1999). During this process, the insoluble pyrite is solubilized to aqueous ferrous sulfate and other intermediate products that are subsequently oxidized to ferric iron and soluble sulfate (Malik et al., 2001a). Therefore, the iron solubilization and sulfur removal are linked to each other. Usualy, the optimal coal
49 GENERAL INTRODUCTION
density in the reaction mixture to remove pyritic sulfur is 10% (Malik et al., 2001a). However, even an almost 40% (w/v) pulp density does not reduce the specific removal rate (Klein et al., 1999).
BOTTLENECKS FOR BIODESULFURIZATION APPLICATION
A significant stumbling block to the commercialization of BDS is the rate at which whole bacterial cells can remove sulfur (Gray et al., 2003). Biocatalyst activity, the oil/water volume ratio and biocatalyst stability constitute the most important technical bottlenecks in the development of biodesulfurization processes. The highest bioconversion values were obtained by unspecific aerobic microorganisms such as Rhizobium meliloti [1200 mg DBT removed (g-1 (DCW) h)-1]. However, these destroy the hydrocarbon structure of the sulfur compound (Klein et al., 1999).
The involvement of three enzymes and two coenzymes in biocatalytical desulfurization makes the use of isolated free or immobilized enzymes difficult. Consequently, schemes using whole cells appear more feasible because they will allow cofactor regeneration in situ.
Another problem to the implementation of a BDS process is the fact that the sulfur requirement of bacteria is low when compared to the level of sulfur found in fuels. In Rhodococcus sp., the cells were found to require 0.1 mM of sulfur for normal growth (Reichmuth et al., 2000). Only 1% of bacterial dry weight is sulfur (Stoner et al., 1990), which implies a very low need in relation to this element. The sulfur content in fossil fuels is about 100 mM and thus, the bacteria cease desulfurization before its total removal.
The utilization of organic solvents and emulsifiers supports protein solubilization and enzymatic reactions in hydrophobic environments (Lee & Yen, 1990). These compounds allow the organosulfur compounds, which have very low water solubility, to be more available to enzymes and microbial cells (Luisi & Laane, 1986).
50 GENERAL INTRODUCTION
The majority of the BDS processes previously reported consists of triphasic systems composed of cells, water and oil (Kaufman et al., 1997; Setti et al., 1997; McFarland et al., 1998). In these cases, oil is mixed together with the cells as a suspension, which produces a sort of surfactant, emulsifying the oil. The separation of the organic and aqueous fractions is very difficult due to those emulsifiers (Naito et al., 2001). Furthermore, the recovery of the cells is also difficult.
Another barrier to commercial acceptance of BDS involves the logistics of sanitary handling, shipment, storage and use of living bacterial cells within the production field or refinery environment. The original BDS process had unacceptable catalyst longevity of about l-2 days. One current design includes the production and regeneration of the biocatalyst within the BDS process, with biocatalyst longevity in the range of 200-400 hours (McFarland, 1999).
In conclusion, commercial application of BDS is dependent upon a better understanding of the underlying BDS mechanisms and the cooperation of scientists and engineers. Enhanced biocatalyst performance may result from improvements in specific activity, stability and longevity, specificity, and new approaches to manufacturing/regenerating the biocatalyst. The key engineering issues include reactor design, separations (oil, water and biocatalyst), byproduct disposition and product quality (McFarland, 1999). The critical success factors for commercial implementation will be cost-effectiveness for the biocatalyst production and the ability of the bioprocess to integrate as seamlessly as possible into existing petrochemical operations (Santos et al., 2007).
SCOPE OF THE THESIS
The aim of this work was to study the dibenzothiophene desulfurization by a novel bacterium belonging to a different genus than the well known Rhodococcus sp. Little was known for other bacterial genus, in terms of physiological, genetic and enzymatic studies.
51 GENERAL INTRODUCTION
From a practical point of view, it is important that desulfurizing strains use a non- destructive metabolic pathway. Some assays involving alternative carbon sources, aiming to decrease the cost of biocatalysts production, were also carried out.
In this context, several stages were performed aiming to attain different objectives:
● Stage 1: Screening and isolation of novel bacteria able to desulfurize DBT as the only sulfur source. Selection, characterization and identification of a bacterial strain which shows a high capability to desulfurize DBT. The goal of this stage of the work was to obtain a new bacterium which degrades DBT and other sulfur compounds present in fossil fuels, preferentially using the 4S metabolic pathway.
● Stage 2: Confirmation of DBT desulfurization pathway used by the selected bacterium, through identification and sequencing of the bacterial genes involved in this metabolism. Comparison of these nucleotide sequences with the homologous sequences described in literature, especially with those from R. erythropolis strain IGTS8.
● Stage 3: Optimization of culture conditions for the selected strain, mainly the metal ion composition of culture medium, to increase its ability to desulfurize DBT. Study of the effect of the presence or absence of some metal ions in culture medium, on bacterial growth and DBT desulfurization.
● Stage 4: Screening and selection of alternative carbon sources derived from agro- industrial residues which can efficiently be used by the selected bacterium to desulfurize DBT. Assessment of the possibility of using the selected syrups, also as source of other nutrients (other than C-source), to grow the selected bacterium. The aim of this stage was to minimize the supplementation of the culture medium using the alternative carbon source, without decreasing the bacterial ability to desulfurize DBT.
52 GENERAL INTRODUCTION
REFERENCES
Amorelli, A., Amos, Y.D., Halsig, C.P., Kosman, J.J., Jonker, R.J., de Wind, M., Vrieling, J., 1992. Estimate feedstock processability. Hydrocarb. Process. 71, 92-101.
Bhatia, S., Sharma, D.K., 2006. Emerging role of biorefining of heavier crude oils and integration of biorefining with petroleum refineries in the future. Pet. Sci. Technol. 24, 1125-1159
Bendinger, B., Rainey, F. A., Moormann, M., Kroppenstedt, R. M., Klatte, S., 1995. Gordona hydrophobica sp. nov., isolated from biofilters for waste gas treatment. Int. J. Syst. Bacteriol. 45, 544-548.
Berthou, F., Vignier, V., 1986. Analyses and fate of dibenzothiophene derivatives in the marine environment. Int. J. Environ. Anal. Chem. 27, 81-96.
Bressler, D.C., Fedorak, P.M., 2001. Purification, stability, and mineralization of 3-hydroxy-2- formylbenzothiophene, a metabolite of dibenzothiophene. Appl. Environ. Microbiol. 67, 821- 826.
Caro, A., Boltes, K., Letón, P., García-Calvo, E., 2007. Dibenzothiophene biodesulfurization in resting cell conditions by aerobic bacteria. Biochem. Eng. J. 35, 191–197.
Chang, J.H., Rhee, S.-K., Chang, Y.K., Chang, H.N., 1998. Desulfurization of diesel oils by a newly isolated dibenzothiophene-degrading Nocardia sp. Strain CYKS2. Biotechnol. Prog. 14, 851- 855.
Chen, W., Brühlmann, F., Richins, R.D., Mulchandani, A., 1999. Engineering of improved microbes and enzymes for bioremediation. Curr. Opin. Biotechnol. 10, 137-141.
Constanti, M., Giralt, J., Bordons, A., 1994. Desulfurization of dibenzothiophene by bacteria. World J. Microbiol. Biotechnol. 10, 510-516.
Constanti, M., Giralt, J., Bordons, A., 1996. Degradation and desulfurization of dibenzothiophene sulfone and other sufur compounds by Agrobacterium MC501 and a mixed culture. Enzyme Microb. Technol. 19, 214-219.
53 GENERAL INTRODUCTION
Dahlberg, M., Rohrer, R.L., Fauth, D. J., Sprecher, R., Olson, G.J., 1993. Biodesulfurization of Dibenzothiophene sulfone by Arthrobacter sp. and studies with oxidized Illinois No.6 coal. Fuel. 72, 1645-1649.
Darzins, A., Mrachko, G.T., 2000. Sphingomonas biodesulfurization catalyst. U.S. Patent nº 6133016.
De Fatima, W.M.J., Setti, L., Lanzarini, G., Pifferi, G., 1996. Dibenzothiophene biodegradation by a Pseudomonas sp. in poorly degradable organic solvents. Process Biochem. 31, 711-717. del Olmo, C.H., Santos, V.E., Alcon, A., Garcia-Ochoa, F., 2005. Production of a Rhodococcus erythropolis IGTS8 biocatalyst for DBT biodesulfurization: influence of operational conditions. Biochem. Eng. J. 22, 229-237.
Denis-Larose, C., Labbé, D., Bergeron, H., Jones, A.M., Greer, C.W., Al-Hawari, J., Grossman, M.J., Sankey, B.M., Lau, P.C.K., 1997. Conservation of plasmid-encoded dibenzothiophene desulfurization genes in several Rhodococci. Appl. Environ. Microbiol. 63, 2915-2919.
Denome, S.A., Olson, E.S., Young, K.D., 1993a. Identification and cloning of genes involved in specific desulfurization of dibenzothiophene by Rhodococcus sp. strain IGTS8. Appl. Environ. Microbiol. 59, 2837-2843.
Denome, S.A., Stanley D.C., Olson, E.S., Young, K.D., 1993b. Metabolism of dibenzothiophene and naphthalene in Pseudomonas strains – complete DNA-sequence of an upper naphthalene catabolic pathway. J. Bacteriol. 175, 6890-6901.
Di Gregorio, S., Zocca, C., Sidler, S., Toffanin, A., Lizzari, D., Vallini, G., 2004. Identification of two new sets of genes for dibenzothiophene transformation in Burkholderia sp. DBT1. Biodegradation. 15, 111-123.
Duarte, G.F., Rosado, A.S., Seldin, L., Araujo, W., Van Elsas, J.D., 2001. Analysis of bacterial community structure in sulfurous-oil-containing soils and detection of species carrying dibenzothiophene desulfurization (dsz) genes. Appl. Environ. Microbiol. 67, 1052-1062.
Dudley, M.W., Frost, J.W., 1994. Biocatalytic desulfurization of arylsulfonates. Bioorg. Med. Chem. 2, 681–690.
54 GENERAL INTRODUCTION
Fairbairn, D.J., Bushell, M.E., 1992. Evaluation of continous enrichment as a method for isolating microbial cultures of coal desulfurization. Enzyme Microb. technol. 14, 48-52.
Finnerty, W. R., 1992. The biology and genetics of the genus Rhodococcus. Annu. Rev. Microbiol. 46, 193-218.
Folsom, B.R., Schieche, D.R., DiGrazia, P.M., Werner, J., Palmer, S., 1999. Microbial desulfurization of alkylated dibenzothiophenes from a hydrodesulfurized middle distillate by Rhodococcus erythropolis I-19. Appl. Environ. Microbiol. 65, 4967-4972.
Frassinetti, S., Setti, L., Corti, A., Farrinelli, P., Montevecchi, P., Vallini, G., 1998. Biodegradation of dibenzothiophene by a nodulating isolate of Rhisobium meliloti. Can. J. Microbiol. 44, 289-297.
Furuya, T., Ishii, Y., Noda, K., Kino, K., Kirimura, K., 2003. Thermophilic biodesulfurization of hydrodesulfurized light gas oils by Mycobacterium phlei WU-F1. FEMS Microbiol. Lett. 221, 137-142.
Furuya, T., Kirimura, K., Kino, K., Usami, S., 2001. Thermophilic biodesulfurization of dibenzothiophene and its derivatives by Mycobacterium phlei WU-F1. FEMS Microbiol. Lett. 204, 129-133.
Gai, Z.H., Yu, B., Li, L., Wang, Y., Ma, C.Q., Feng, J.H., Deng, Z.X., Xu, P., 2007. Cometabolic degradation of dibenzofuran and dibenzothiophene by a newly isolated carbazole-degrading Sphingomonas sp. strain. Appl. Environ. Microbiol. 73, 2832-2838.
Gallangher, J.R., Olson, E.S., Stanley, D.C., 1993. Microbial desulfurization of dibenzothiophene: a sulfur-specific pathway. FEMS Microbiol. Lett. 107, 31-36.
Gallardo, M.E., Ferrandez, A., De Lorenzo, V., Garcia, J.L., Diaz, E., 1997. Designing recombinant Pseudomonas strains to enhance biodesulfurization. J. Bacteriol. 179, 7156-7160.
Gilbert, S.C., Morton, J., Buchanan, S., Oldfield, C., McRoberts, A., 1998. Isolation of a unique benzothiophene-desulfurizing bacterium, Gordona sp. strain 213E (NCIMB 40816), and characterization of the desulphurization pathway. Microbiology. 144, 2545-2553.
Gómez, F., Amils, R., Marín, I., 1999. Bioremoval of organic and inorganic sulphur from coal samples. Appl. Microbiol. Biotechnol. 52, 118-121.
55 GENERAL INTRODUCTION
Goodfellow, M., Alderson, G., 1977. The actinomycete genus Rhodococcus: a home for rhodochrous complex. J Gen Microbiol 100, 99-122.
Gray, K.A., Mrachko, G.T., Squires, C.H., 2003. Biodesulfurization of fossil fuels. Curr. Opin. Microbiol. 6, 229-235.
Gray, K.A., Pogrebinsky, O.S., Mrachko, G.T., Xi, L., Monticello, D.J., Squires, C.H., 1996. Molecular mechanisms of biocatalytic desulfurization of fossil fuels. Nature Biotechnol. 14, 1705-1709.
Grossman, M.J., Lee, M.K., Prince, R.C., Garrett, K.K., George, G.N., Pickering, I.J., 1999. Microbial desulfurization of a crude oil middle-distillate fraction: analysis of the extent of sulfur removal and the effect of removal on remaining sulfur. Appl. Environ. Microbiol. 65, 181-188.
Grossman, M.J., Lee, M.K., Prince, R.C., Minak-Bernero, V., George, G.N., Pickering, I.J., 2001. Deep desulfurization of extensively hydrodesulfurized middle distillate oil by Rhodococcus sp. strain ECRD-1. Appl. Environ. Microbiol. 67, 1949-1952.
Guobin, S., Jianmin, X., Huaiying, Z., Huizhou, L., 2005. Deep desulfurization of hydrodesulfurized diesel oil by Pseudomonas delafieldii R-8. J. Chem. Technol. Biotechnol. 80, 420-424.
Gupta, N., Roychoudhury, P.K., Deb, J.K., 2005. Biotechnology of desulfurization of diesel: prospects and challenges. Appl. Microbiol. Biotechnol. 66, 356-366.
Hou, Y., Kong, Y., Yang, J., Zhang, J., Shi, D., Xin, W., 2005. Biodesulfurization of dibenzothiophene by immobilized cells of Pseudomonas stutzeri UP-1. Fuel 84, 1975–1979.
Ishii, Y., Konishi, J., Okada, H., Hirasawa, K., Onaka, T., Suzuki, M., 2000a. Operon structure and functional analysis of the genes encoding thermophilic desulfurizing enzymes of Paenibacillus sp. A11-2. Biochem. Biophys. Res. Commun. 270, 81-88.
Ishii, Y., Konishi, J., Suzuki, M., Maruhashi, K., 2000b. Cloning and expression of the gene encoding the thermophilic NAD(P)H-FMN oxidoreductase coupling with the desulfurization enzymes from Paenibacillus sp. A11-2. J. Biosci. Bioeng. 90, 591-599.
Ishii, Y., Kozaki, S., Furuya, T., Kino, K., Kirimura, K., 2005. Thermophilic biodesulfurization of various heterocyclic sulfur compounds and crude straight-run light gas oil fraction by a newly isolated strain Mycobacterium phlei WU-0103. Curr. Microbiol. 50, 63-70.
56 GENERAL INTRODUCTION
Izumi, Y., Ohshiro, T., Ogino, H., Hine, Y., Shimao, M., 1994. Selective desulfurization of dibenzothiophene by Rhodococcus erythropolis D-1. Appl. Environ. Microbiol. 60, 223-226.
Jia X., Wen J.P., Sun Z.P., Caiyin Q.G., Xie S.P., 2006. Modeling of DBT biodegradation behaviors by resting cells of Gordonia sp. WQ-01 and its mutant in oil-water dispersions. Chem. Eng. Sci. 61, 1987-2000.
Ju, L.-K., Kankipati, P., 1998. Toxicity of dibenzothiophene to thermophile Sulfolobus acidocaldarius grown in sucrose medium. J. Biotechnol. 63, 219-227.
Kabe, T., Ishihara, A., Tajima, H., 1992. Hydrodesulfurization of sulfur-containing polyaromatic compounds in light oil. Ind. Eng. Chem. Res. 31, 1577-1580.
Kargi, F., 1987. Biological oxidation of thionthrene, thioxanthene and dibenzothiophene by the thermophilic organism Sulfolobus acidocaldarius. Biotechnol. Lett. 9, 478-482.
Kargi, F., Weissman, J.G., 1987. Kinetic parameter estimation in microbial desulfurization of coal. Biotechnol. Bioeng. 30, 1063-1066.
Kaufman, E.N., Harkins, J.B., Barole, A.P., 1998. Comparison of batch-stirred and electrospray reactors for biodesulfurization of dibenzothiophene in crude oil and hydrocarbon feedstocks. Appl. Biochem. Biotechnol. 73, 127-143.
Kaufman, E.N., Harkins, J.B., Rodriguez, M., Tsouris, C., Selvaraj, P.T., Murphy, S.E., 1997. Development of an electro-spray bioreactor for crude oil processing. Fuel Process Technol. 52, 127-144.
Kayser, K.J., Bielaga-Jones, B.A., Jackowsky, K., Odusan, O., Kilbane, J.J., 1993. Utilization of organosulphur compounds by axenic and mixed cultures of Rhodococcus rhodochrous IGTS8. J. Gen. Microbiol. 139, 3123-3129.
Kerr, R., 2000. USGS optimistic on world oil prospects. Science. 289, 237.
Kilbane II, J.J., 2006. Microbial biocatalysts development to upgrade fossil fuels. Curr. Opin. Biotechnol. 17, 305-314.
Kilbane, J.J., 1989. Desulfurization of coal: the microbial solution. Trends Biotechnol. 7, 97-100.
Kilbane, J.J., Jackowsky, K., 1992. Biodesulfurization of water-soluble coal-derived material by Rhodococcus rhodochrous IGTS8. Biotechnol. Bioeng. 40, 1107-1114.
57 GENERAL INTRODUCTION
Kim, Y.J., Chang, J.H., Cho, K.-S., Ryu, H.W., Chang, Y.K., 2004. physiological study on growth and dibenzothiophene (DBT) dsulfurization characteristics of Gordonia sp. CYKS1. Korean J. Chem. Eng. 21, 436-441.
Kim, H. Y., Kim, T. S., Kim, B. H., 1990. Degradation of organic sulfur compounds and the reduction of dibenzothiophene to biphenyl and hydrogen sulfide by Desulfovibrio desulfuricans Mb. Biotechnol. Lett. 12, 757-760.
Kirimura, K., Furuya, T., Nishii, Y., Ishii, Y., Kino, K., Usami, S., 2001. Biodesulfurization of dibenzothiophene and its derivatives through the selective cleavage of carbon-sulfur bonds by a moderately thermophilic bacterium Bacillus subtilis WU-S2B. J. Biosci. Bioeng. 91, 262-266.
Kirimura, K., Harada, K., Iwasawa, H., Tanaka, T., Iwasaki, Y., Furuya, T., Ishii, Y., Kino, K., 2004. Identification and functional analysis of the genes encoding dibenzothiophene-desulfurizing enzymes from thermophilic bacteria. Appl. Environ. Microbiol. 65, 703-713.
Klatte, S., Kroppenstedt, R. M., Schumann, P., Altendorf, K., Rainey, F. A., 1996. Gordona hirsuta sp. nov. Int. J. Syst. Bacteriol. 46, 876-880.
Klein, J., Catcheside, D.E.A., Fakoussa, R., Gazso, L., Fritsche, W., Höfer, M., Laborda, F., Margarit, I., Rehm, H.-J., Reich-Walber, M., Sand, W., Schacht, S., Schmiers, H., Setti, L., Steinbüchel, A., 1999. Biological processing of fossil fuels. Appl. Microbiol. Biotechnol. 52, 2- 15.
Kobayashi, M., Horiuchi, K., Yoshikawa, O., Hirasawa, K., Ishii, Y., Fujino, K., 2001. Kinetic analysis of microbial desulfurization of model and light gas oils containing multiple alkyl dibenzothiophenes. Biosci. Biotechnol. Biochem. 65, 298–304.
Kobayashi, M., Toshimitsu, O., Onaka, T., Ishii, Y., Konishi, J., Takaki, M., Okada, H., Ohta, Y., Koizumi, K., Susuki, M., 2000. Desulfurization of alkylated forms of both dibenzothiophene and benzothiophene by a single bacterial strain. FEMS Microbiol. Lett. 187, 123-126.
Kodama, K., Umehara, K., Shimizu, K., Nakatani, S., Monioda, Y., Yamada, K., 1973. Identification of microbial products from dibenzothiophene and its proposed oxidation pathway. Agric. Biol. Chem. 37, 45-50.
58 GENERAL INTRODUCTION
Konishi, J., Ishii, Y., Onaka, T., Ohta, Y., Suzuki, M., Maruhashi, K., 2000b. Purification and characterization of dibenzothiophene sulfone monooxygenase from the thermophilic bacterium Paenibacillus sp. strain A11-2. J. Biosci. Bioeng. 90, 607-613.
Konishi, J., Ishii, Y., Onaka, T., Okumura, K., Susuki, M., 1997. Thermofilic carbon-sulfur bond targeted biodesulfurization. Appl. Environ. Microbiol. 63, 3164-3169.
Konishi, J., Maruhashi, K., 2003. 2-(2'-hydroxyphenyl)benzene sulfinate desulfinase from the thermophilic desulfurizing bacterium Paenibacillus sp. strain A11-2: purification and characterization. Appl. Microbiol. Biotechnol. 62, 356-361.
Konishi, J., Okada, H., Hirasawa, K., Ishii, Y., Maruhashi, K., 2002. Comparison of the substrate specificity of the two bacterial desulfurization systems. Biotechnol. Lett. 24, 1863-1867.
Konishi, J., Onaka, T., Ishii, Y., Susuki, M., 2000a. Demonstration of the carbon-sulfur bond targeted desulfurization of benzothiophene by thermophilic Paenibacillus sp. Strain A11-2 capable of desulfurizing dibenzothiophene. FEMS Microbiol. Lett. 187, 151-154.
Konishi, M., Kishimoto, M., Omasa, I., Katakura, Y., Shioya, S., Ohtake, H., 2005a. Effect of sulfur sources on specific desulfurization activity of Rhodococcus erythropolis KA2-5-1 in exponential fed-batch culture. J. Biosci. Bioeng. 24, 49-54.
Konishi, M., Kishimoto, M., Tamesui, N., Omasa, I., Shioya, S., Ohtake, H., 2005b. The separation of oil from an oil-water-bacteria mixture using a hydrophobic tubular membrane. Biochem. Eng. J. 24, 49-54.
Kropp, K. G., Andersson, J.T., Fedorak, P.M., 1997. Bacterial transformations of 1,2,3,4- tetrahidrodibenzothiophene and dibenzothiophene. Appl. Environ. Microbiol. 63, 3032-3042.
Kropp, K., Fedorak, P., 1998. A review of the occurrence, toxicity, and biodegradation of condensed thiophenes found in petroleum. Can. J. Microbiol. 44, 605-622.
Laborde, A.L., Gibson, D.T., 1977. Metabolism of dibenzotiophene by a Beijerinckia sp. Appl. Environ. Microbiol. 34, 783-790.
Lee, K.I., Yen, T.F., 1990. Sulfur removal from coal through multiphase media containing biocatalysts. J. Chem. Tech. Biotechnol. 3, 71-79.
59 GENERAL INTRODUCTION
Lee, M.K., Senius, J.V., Grossman, M.J., 1995. Sulfur-specific microbial desulfurization of sterically hindered analogs of dibenzothiophene. Appl. Environ. Microbiol. 61, 4362-4366.
Lees, H., Kwok, S.C., Suzuki, I., 1996. The thermodynamics of iron oxidation by the ferrobacilli. Can. J. Microbiol. 15, 43-46.
Lemmer, H., Kroppenstedt, R. M., 1984. Chemotaxonomy and physiology of some actinomycetes isolated from scumming activated sludge. Syst. Appl. Microbiol. 5, 124-135.
Li, F.L., Xu, P., Feng, J.H., Meng, L., Zheng, Y., Luo, L.L., Ma, C.Q., 2005. Microbial desulfurization of gasoline in a Mycobacterium goodii X7B immobilized-cell system. Appl. Environ. Microbiol. 71, 276-281.
Li, F.L., Zhang, Z.Z., Feng, J.H., Cai, X.F., Xu, P., 2007. Biodesulfurization of DBT in tetradecane and crude oil by a facultative thermophilic bacterium Mycobacterium goodii X7B. J. Biotechnol. 127, 222-228.
Li, M.Z., Squires, C.H., Monticello, D.J., Childs, J.D., 1996. Genetic analysis of the dsz promoter and associated regulatory regions of Rhodococcus erythropolis IGTS8. J. Bacteriol. 178, 6409- 6418.
Lu, J., Nakajima-Kambe, T., Shigeno, T., Ohbo, A., Nomura, N., Nakahara, T., 1999. Biodegradation of dibenzothiophene and 4,6-dimethyldibenzothiophene by Sphingomonas paucimobilis strain TZS-7. J. Biosci. Bioeng. 88, 293-299.
Luo, M.F., Xing, J.M., Gou, Z.X., Li, S., Liu, H.Z., Chen, J.Y., 2003. Desulfurization of dibenzothiophene by lyophilized cells of Pseudomonas delafieldii R-8 in the presence of dodecane. Biochem. Eng. J. 13, 1-6.
Luisi, P.L., Laane, C., 1986. Solubilization of enzymes in apolar solvents via reverse micelles. T ibtech. 4, 153-161.
Ma, C.-Q., Feng, J.-H., Zeng, Y.-Y., Cai, X.-F., Sun, B.-P., Zhang, Z.-B., Blankespoor, H.D., Xu, P., 2006. Methods for the preparation of a biodesulfurization biocatalyst using Rhodococcus sp. Chemosphere. 65, 165-169.
Maghsoudi, S., Vossoughi, M., Kheirolomoom, A., Tanaka, E., Katoh, S., 2001. Biodesulfurization of hydrocarbons and diesel fuels by Rhodococcus sp. strain P32C1. Biochem. Eng. J. 8, 151-156.
60 GENERAL INTRODUCTION
Malik, A., Dastidar, M.G., Roychoudhury, P.K., 2001a. Biodesulphurization of coal: effect of pulse feeding and leachate recycle. Enzyme Microb. Technol. 28, 49-56.
Malik, A., Dastidar, M.G., Roychoudhury, P.K., 2001b. Biodesulphurization of coal: Mechanism and rate limiting factors. J. Environ. Sci. Health., Part A. 36, 1113-1128.
Malik, K.A., 1978. Microbial removal of organic sufur from crude oil and the environment: some new perspectives. Process Biochem. 13, 10-12.
Matsubara, T., Ohshiro, T., Nishina, Y., Izumi, Y., 2001. Purification, characterization, and overexpression of flavin reductase involved in dibenzothiophene desulfurization by Rhodococcus erythropolis D-1. Appl. Environ. Microbiol. 67, 1179-1184.
Matsui, T., Noda, K., Tanaka, Y., Maruhashi, K., Kurane, R., 2002. Recombinant Rhodococcus sp. Strain T09 can desulfurize DBT in the presence of inorganic sulfate. Curr. Microbiol. 45, 240- 244.
Matsui, T., Onaka, T., Maruhashi, K., Kurane, R., 2001. Benzo[b]thiophene desulfurization by Gordonia rubropertinctus strain T08. Appl. Microbiol. Biotechnol. 57, 212-215.
Matsui, T., Onaka, T., Tanaka, Y., Tezuka, T., Suzuki, M., Kurane, R., 2000. Alkylated benzothiophene desulfurization by Rhodococcus sp. strain T09. Biosci. Biotechnol. Biochem. 64, 596-599.
McFarland, B.L., 1999. Biodesulfurization. Current Opinion Microbiol. 2, 257-264.
Mohebali, G., Ball, A.S., Rasekh, B., Kaytash, A., 2007. Biodesulfurization potential of a newly isolated bacterium, Gordonia alkanivorans RIPI90A. Enz. Microbial. Technol. 40, 578-584.
Monticello, D.J., 1998. Riding the fossil fuel biodesulfurization wave. Chemtech. 28, 38-45.
Monticello, D.J., 2000. Biodesulfurization and the upgrading of petroleum distillates. Curr. Opin. Microbiol. 11, 540-546.
Monticello, D.J., Bakker, D., Finnerty, W.R., 1985. Plasmid-mediated degradation of dibenzothiophene by Pseudomonas species. Appl. Environ. microbiol. 49, 756-760.
Mormile, M.R., Atlas, R.M., 1988. Mineralization of the dibenzothiophene: biodegradation products 3-hydroxy-2-formyl benzothiophene and dibenzothiophene sulfone. Appl. Environ. Microbiol. 54, 3183-3184.
61 GENERAL INTRODUCTION
Naito, M., Kawamoto, T., Fujino, K., Kobayashi, M., Maruhashi, K., Tanaka, A., 2001. Long-term repeated biodesulfurization by immobilized Rhodococcus erythropolis KA2-5-1. Appl. Microbiol. Biotechnol. 55, 374-378.
Nakayama, N., Matsubara, T., Ohshiro, T., Moroto, Y., Kawata, Y., Koizumi, K., Hirakawa, Y., Suzuki, M., Maruhashi, K., Izumi, Y., Kurane, R., 2002. A novel enzyme, 2'-hydroxybiphenyl-2- sulfinate desulfinase (DszB), from a dibenzothiophene-desulfurizing bacterium Rhodococcus erythropolis KA2-5-1: gene overexpression and enzyme characterization. Biochim. Biophys. Acta, Proteins Proteomics. 1598, 122-130.
Nekodzuka, S., Toshlaki, N., Nakallma-Kambe, T., Nobura, N., Lu, J., Nakahara, Y., 1997. Specific desulfurization of dibenzothiophene by Mycobacterium strain G3. Biocatalysis Biotransformation. 15, 21-27.
Niquille-Röthlisberger, A., Prins, R., 2006. Hydrodesulfurization of 4,6-dimethyldibenzothiophene and dibenzothiophene over alumina-supported Pt, Pd and Pt-Pd catalysts. J. Catal. 242, 207-216.
Nomura, N., Takada, M., Okada, H., Shinohara, Y., Nakajima-Kambe, T., Nakahara, T., Uchiyama, H., 2005. Identification and functional analysis of genes required for desulfurization of alkyl dibenzothiophenes of Mycobacterium sp. G3. J. Biosc. Bioeng. 100, 398-402.
Ohshiro, T., Hine, Y., Izumi, Y., 1994. Enzymatic desulfurization of dibenzothiophene by a cell-free system of Rhodococcus erythropolis D-1. FEMS Microbiol. Lett. 118, 341-344.
Ohshiro, T., Hirata, T., Hashimoto, I., Izumi, Y., 1996. Characterization of dibenzothiophene desulfurization reaction by whole cells of Rhodococcus erythropolis H-2 in the presence of hydrocarbon. J. Ferm. Bioeng. 82, 610-612.
Ohshiro, T., Hirata, T., Izumi, Y., 1995. Microbial desulfurization of dibenzothiophene in the presence of hydrocarbon. Appl. Microbiol. Biotechnol. 44, 249-252.
Ohshiro, T., Ishii, Y., Matsubara, T., Ueda, K., Izumi, Y., Kino, K., Kirimura, K., 2005. Dibenzothiophene desulfurizing enzymes from moderately thermophilic bacterium Bacillus subtilis WU-S2B: Purification, characterization and overexpression. J. Biosc. Bioeng. 100, 266- 273.
Ohshiro, T., Izumi, Y., 1999. Microbial desulfurization of organic sulfur compounds in petroleum. Biosci. Biotechnol. Biochem. 63, 1-9.
62 GENERAL INTRODUCTION
Ohshiro, T., Kojima, T., Torii, K., Kawasoe, H., Izumi, Y., 1999. Purification and characterization of dibenzothiophene (DBT) sulfone monooxygenase, an enzyme involved in DBT desulfurization, from Rhodococcus erythropolis D-1. J. Biosc. Bioeng. 88, 610-616.
Ohshiro, T., Suzuki, K., Izumi, Y., 1997. Dibenzothiophene (DBT) degrading enzyme responsible for the first step of DBT desulfurization by Rhodococcus erythropolis D-1: purification and characterization. J. Ferment. Bioeng. 83, 233-237.
Oldfield, C., Pogrebinsky, O., Simmonds, J., Olson, E.S., Kulpa, C.F., 1997. Elucidation of the metabolic pathway for dibenzothiophene desulphurization by Rhodococcus sp. strain IGTS8 (ATCC 53968). Microbiology. 143, 2961-2973.
Oldfield, C., Wood, N.T., Gilbert, S.C., Murray, F.D., Faure, F.R., 1998. Desulphurisation of benzothiophene and dibenzothiophene by actinomycete organisms belonging to the genus Rhodococcus, and related taxa. Antonie van Leeuwenhoek. 74, 119-132.
Olson, G.J., Kelly, R.M., 1986. Microbiological metal transformation: biotechnological applications and potential. Biotechnol. Prog. 2, 1-15.
Omori, T., Monna, L., Saiki, Y., Kodama, T., 1992. Desulfurization of dibenzothiophene by Corynebacterium sp. Strain SY1. Appl. Environ. Microbiol. 58, 911-915.
Onaka, T., Kobayashi, M., Ishii, Y., Konishi, J., Maruhashi, K., 2001a. Selective cleavage of the two C-S bonds in asymmetrically alkylated dibenzothiophenes by Rhodococcus erythropolis KA2-5- 1. J. Biosci. Bioeng. 92, 80-82.
Onaka, T., Konishi, J., Ishii, Y., Maruhashi, K., 2001b. Desulfurization characteristics of thermophilic Paenibacillus sp. strain A11-2 against asymmetrically alkylated dibenzothiophenes. J. Biosci. Bioeng. 92, 193-196.
Onodera-Yamada, K., Morimoto, M., Tani, Y., 2001. Degradation of dibenzothiophene by sulfate- reducing bacteria cultured in the presence of only nitrogen gas. J. Biosci. Bioeng. 91, 91-93.
Park, S.J., Lee, I.-S., Chang, Y.K., Lee, S.Y., 2003. Desulfurization of dibenzothiophene and diesel oil by metabolically engineered Escherichia coli. J. Microbiol. Biotechnol. 13, 578-583.
63 GENERAL INTRODUCTION
Piddington, C.S., Kovacevich, B.R., Rambosek, J., 1995. Sequence and molecular characterization of a DNA region encoding the dibenzothiophene desulfurization operon of Rhodococcus sp. strain IGTS8. Appl. Environ. Microbiol. 61, 468-475.
Prins, R., Egorova, A., Rothlisberger, A., Zhao, Y., Sivasankar, N., Kukula, P., 2006. Mechanisms of hydrodesulfurization and hydrodenitrogenation. Catal. Today. 111, 84-93.
Rambosek, J., Piddington, C.S., Kovacevich, B.R., Young, K.D., Denome, S.A., 1994. Recombinant DNA encoding a desulfurization biocatalyst. U.S. Patent 5356801.
Rashtchi, M., Mohebali, G.H., Akbarnejad, M.M., Towfighi, J., Rasekh, B., Keytash, A., 2006. Analysis of biodesulfurization of model oil system by the bacterium, strain RIPI-22. Biochem. Eng. J. 29, 169-173
Reichmuth, D.S., Hittle, J.L., Blanch, H.W., Keasling, J.D., 2000. Biodesulfurization of dibenzothiophene in Escherichia coli is enhanced by expression of a Vibrio harveyi oxidoreductase gene. Biotechnol. Bioeng. 67, 72-79.
Rhee, S., Chang, J.H., Chang, Y.K., Chang, H.N., 1998. Desulfurization of dibenzothiophene and diesel oils by a newly isolated Gordona strain, CYKS1. Appl. Environ. Microbiol. 64, 2327- 2331.
Santos, S.C.C., Alviano, D.S., Alviano, C.S., Goulart, F.R.V., Pádula, M., Leitão, A.C., Martins, O.B., Ribeiro, C.M.S., Sassaki, M.Y.M., Matta, C.P.S., Bevilaqua, J., Sebastian, G..V, Seldin, L., 2007. Comparative studies of phenotypic and genetic characteristics between two desulfurizing isolates of Rhodococcus erythropolis and the well-characterized R. erythropolis strain IGTS8. J. Ind. Microbiol. Biotechnol. 34, 423–431
Santos, S.C.C., Alviano, D.S., Alviano, C.S., Padula, M., Leitao, A.C., Martins, O.B., Ribeiro, C.M.S., Sassaki, M.Y.M., Matta, C.P.S., Bevilaqua, J., Sebastian, G..V, Seldin, L., 2006. Characterization of Gordonia sp strain F.5.25.8 capable of dibenzothiophene desulfurization and carbazole utilization. Appl. Microbiol. Biotechnol. 71, 355-362.
Schippers, A., Rohwerder, T., Sand, W., 1999. Intermediary sulfur compounds in pyrite oxidation: implications for bioleaching and biodepyritization of coal. Appl. Microbiol. Biotechnol. 52, 104- 110.
64 GENERAL INTRODUCTION
Schmidt, M., Siebert, W., Bagnall, K.W., 1973. “The chemistry of sulphur, selenium, tellurium and polonium”. Pergamon Texts in Inorganic Chemistry, vol.15, Pergamon Press, Oxford.
Serbolisca, L., de Ferra, F., Margarit, I., 1999. Manipulation of the DNA coding for the desulphurizing activity in a new isolate of Arthrobacter sp. Appl. Microbiol. Biotechnol. 52, 122-126.
Setti, L., Lanzarini, G., Pifferi, P.G., 1997. Whole cell biocatalysis for an oil desulfurization process. Fuel Process. Technol. 52, 145-153.
Setti, L., Rossi, M., Lanzarini, G., Pifferi, P.G., 1992. The effect of n-alkanes in the degradation of dibenzothiophene and of organic sulfur compounds in heavy oil by a Pseudomonas sp. Biotech. Lett. 14, 515-520.
Setti, L., Rossi, M., Lanzarini, G., Pifferi, P.G., 1993. Barrier and carrier effects of n-dodecane on the anaerobic degradation of benzothiophene by Desulfovibrio desulfuricans. Biotech. Lett. 15, 527- 530.
Shan, G.B., Xing, J.M., Zhang, H.Y., Liu, H.Z., 2005. Biodesulfurization of dibenzothiophene by microbial cells coated with magnetite nanoparticles. Appl. Environ. Microbiol. 71, 4497-4502.
Sohrabi, M, Jamshidi, A.M., Vahabzadeh, F., Bonakdarpour, B., 2006. Some aspects of microbial desulfurization of dibenzothiophene in a down flow jet loop bioreactor with coaxial draft tube. Afinidad. 63, 136-142.
Stackebrandt, E., Rainey, F. A., Ward-Rainey, N. L., 1997. Proposal for a new hierarchic classification system, Actinobacteria classis nov. Int J Syst Bacteriol 47, 479-491.
Stackebrandt, E., Smida, J., Collins, M. D., 1988. Evidence of phylogenetic heterogeneity within the genus Rhodococcus: revival of the genus Gordona (Tsukamura). J Gen Appl Microbiol 34, 341- 348.
Stoner, D.L., Wey, J.E., Barrett, K.B., Jolley, J.G., Wright, R.B., Dugan, P.R., 1990. Modification of water-soluble coal-derived products by dibenzothiophene-degrading microorganisms. Appl. Environ. Microbiol. 56, 2667-2676.
Swaty, T.E., 2005. Global refining industry trends: the present and future. Hydrocarbon. Processing. 84, 35-46.
65 GENERAL INTRODUCTION
Takada, M., Nomura, N., Okada, H., Nakajima-Kambe, T., Nakahara, T., Uchiyama, H., 2005. De- repression and comparison of oil-water separation activity of the dibenzothiophene desulfurizing bacterium, Mycobacterium sp. G3. Biotechnol. Lett. 27, 871-874.
Takeuchi, M., Hatano, K., 1998. Gordonia rhizosphera sp. nov. isolated from the mangrove rhizosphere. Int. J. Syst. Bacteriol. 48, 907-912.
Tanaka, Y., Onaka, T., Matsui, T., Maruhashi, K., Kurane, R., 2001. Desulfurization of benzothiophene by the gram-negative bacterium, Sinorhizobium sp. KT55. Curr. Microbiol. 43, 187-191.
Tsukamura, M., 1971. Proposal of a new genus, Gordona, for slightly acid-fast organisms occurring in sputa of patients with pulmonary disease and in soil. J Gen Microbiol 68, 15-26.
Van Afferden, M., Schacht, S., Klein, J., Trüper, H.G., 1990. Degradation of dibenzothiophene by Brevibacterium sp. DO. Arch. Microbiol. 153, 324-328.
Van Afferden, M., Tappe, D., Beyer, M., Trüper, H.G., Klein, J., 1993. Biochemical mechanisms for the desulfurization of coal-relevant organic sulfur compounds. Fuel. 72, 1635-1643.
Van Hamme, J.D., Singh, A., Ward, O.P., 2003. Recent advances in petroleum microbiology. Microbiol. Molec. Biol. Rev. 67, 503-549.
Wang, P., Krawiec, S., 1994. Desulfurization of dibenzothiophene to 2-hidroxybiphenyl by some newly isolated bacterial strains. Arch. Microbiol. 161, 266-271.
Wang, P., Krawiec, S., 1996. Kinetic analyses of desulfurization of dibenzothiophene by Rhodococcus erythropolis in batch and fed-batch cultures. Appl. Environ. Microbiol. 62, 1670- 1675.
Watkins, L.M., Rodriguez, R., Schneider, D., Broderick, R., Cruz, M., Chambers, R., Ruckman, E., Cody, M., Mrachko, G.T., 2003. Purification and characterization of the aromatic desulfinase, 2- (2'-hydroxyphenyl)benzenesulfinate desulfinase. Arch. Biochem. Biophys. 415, 14-23.
Xi, L., Squires, C.H., Monticello, D.J., Childs, J.D., 1997. A flavin reductase stimulates DszA and DszC proteins of Rhodococcus erythropolis IGTSB in vitro. Biochem. Biophys. Res. Comm. 230, 73-75.
66 GENERAL INTRODUCTION
Yang, J., Marison, I.W., 2005. Two-stage process design for the biodesulphurisation of a model diesel by a newly isolated Rhodococcus globerulus DAQ3. Biochem. Eng. J. 27, 77–82.
Yeong, K.H., Kim, T.S., Kim, B.H., 1990. Degradation of organic sulfur compounds and the reduction of dibenzothiophene to biphenyl and hydrogen sulfide. Biotechnol. Lett. 12, 761-764.
Zhang, Q., Tong, M.Y., Li, Y.S., Gao, H.J., Fang, X.C., 2007. Extensive desulfurization of diesel by Rhodococcus Erythropolis. Biotechnol. Lett. 29, 123–127.
67
Chapter 2
DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
DESULFURIZATION OF DIBENZOTHIOPHENE, BENZOTHIOPHENE AND OTHER THIOPHENE ANALOGUES BY A NEWLY ISOLATED BACTERIUM Gordonia alkanivorans STRAIN 1B
L. Alves, R. Salgueiro, C. Rodrigues, E. Mesquita, J. Matos, F.M. Gírio
INETI, Departamento de Biotecnologia, Estrada do Paço do Lumiar, 22, 1649-038 Lisboa, Portugal.
ABSTRACT - A novel bacterium, Gordonia alkanivorans strain 1B, was isolated from hydrocarbon-contaminated soil. Assessment of the biodegradation of distinct organic sulfur- compounds, such as dibenzothiophene (DBT), benzothiophene (BT), DBT sulfone, and alkylated tiophenic compounds, as the sole source of sulfur was investigated. G. alkanivorans strain 1B was able to remove selectively the sulfur from DBT while keeping intact the remaining carbon-carbon structure. Orthophenyl phenol (2-hydroxybiphenyl) was the only detected metabolic product. The bacterial desulfurization activity was repressed by sulfate. G. alkanivorans strain 1B consumed 310 µM DBT after 120 h of cultivation, corresponding to a specific desulfurization rate of 1.03 µmol g-1 of dry cells h-1. When an equimolar mixture of DBT/BT was used as a source of sulfur in the growth medium, G. alkanivorans strain 1B assimilated both compounds in a sequential manner, with BT as the preferred source of sulfur. Only when BT concentration was decreased to a very low level was DBT utilized as the source of sulfur for bacterial growth. The specific desulfurization overall rates of BT and DBT obtained were 0.954 and 0.813 µmol g-1 of dry cells h-1, respectively. The newly isolated G. alkanivorans strain 1B has good potential for application in the biodesulfurization of fossil fuels.
Keywords: Biodesulfurization, Dibenzothiophene, Benzothiophene, Sulfur, Gordonia alkanivorans.
71 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
INTRODUCTION
The increasing utilization of fossil fuels owing to the demand of developed countries has resulted in sulfur oxide emission into the atmosphere being one of the main contributors to air pollution. These oxides are primarily responsible for acid rain owing to their reaction with water in the atmosphere, which results in the formation of sulfuric acid (Reichmuth et al., 2000). Additionally, sulfur oxides at low concentration in the atmosphere provoke irritation in the human respiratory system and damage plants (Schmidt et al., 1973). This problem could be much more severe in the future; the estimated average sulfur content in the world’s crude oil reserves is expected to increase 12% from 1990 to 2010 (Monticello, 1998).
Consequently, developed countries are imposing increasingly stringent restrictions on the maximum sulfur concentration present in fossil fuels in order to control the emission of sulfur (Gray et al., 2003). In the European Union, petrochemical plants are mandatory in order to produce fuels of lower sulfur content by 2005 (maximum allowable level of 50 ppm) than the current content (maximum allowable level of 150 ppm). The current physicochemical process used in most refineries, so-called hydrodesulfurization, has technical limitations. It works inefficiently for removing the sulfur from organic aromatic compounds, such as benzothiophene (BT), dibenzothiophene (DBT), and their derivatives (Tanaka et al., 2002). Consequently, hydrodesulfurization- refractory sulfur compounds from middle and heavy distillate range fuels pose a significant barrier to reaching very low sulfur levels (Grossman et al., 2001).
In the last two decades, an alternative/complementary biologic process for oil desulfurization involving bacteria with the ability to use either BT or DBT (model compounds) as the only source of sulfur was described (Ohshiro & Izumi, 1999). However, most of the bacteria that metabolize DBT did not have the ability to degrade BT and vice versa (Kirimura et al., 2002). In Rhodococcus erythropolis IGTS8, desulfurization of DBT occurs through the so-called 4S metabolic pathway, producing 2-hydroxybiphenyl (2-HBP)
72 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
as an end product (Gallagher et al., 1993; Oldfield et al., 1997). On the other hand, the desulfurization of BT by Gordonia sp. strain 213E was reported to yield 2-(2'- hydroxyphenyl)ethan 1-al (Gilbert et al., 1998). Both metabolic pathways have the advantage of keeping the C-C structure of these sulfurcontaining aromatic hydrocarbons intact, maintaining their combustion value. Other DBT-catabolic pathways reported in several other bacteria are not as attractive for industrial purposes because they either mineralize DBT to CO2 (Van Afferden et al., 1990) or partially destroy the DBT carbon structure (Kodama et al., 1973).
From a practical point of view, a bacterial strain that displays the ability to desulfurize both BT and DBT would be ideal. So far, this ability has been reported only in the bacteria Paenibacillus sp. strain A11-2 (Konishi et al., 2000), Mycobacterium phlei GTIS10 (Kayser et al., 2002), Gordonia strain CYKS1 (Rhee et al., 1998), Nocardia sp. strain CYKS2 (Chang et al., 1998), and Rhodococcus spp. strain KT462 (Tanaka et al., 2002). However, all of these studies reported only the single utilization of DBT or BT as the sole source of sulfur in the bacterial growth medium.
In this article, we describe the isolation and characterization of a novel strain, Gordonia alkanivorans 1B, toward the utilization of several inorganic and organic sources of sulfur for growth. We also discuss the kinetics of DBT and BT degradation by G. alkanivorans strain 1B during a mixed sulfur source growth.
MATERIALS AND METHODS
Chemicals
DBT (99%) was obtained from Acros Organics, DBT sulfone (97%) and BT (95%) were from Aldrich, 2-HBP was from Sigma, DMF was from Riedelde Haën, and anthracene and
73 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
ethyl acetate were from Merck. All other reagents were of the highest grade commercially available.
Culture media and growth conditions
Sulfur-free mineral (SFM) culture medium contained 1.22 g of NH4Cl, 2.5 g of KH2PO4, 2.5 g of Na2HPO4·2H2O, 0.17 g of MgCl2·6H2O, and 1 liter of milli-Q water. To the SFM medium was added 0.5 ml of a trace elements solution without sulfur containing 25 g l-1 of -1 -1 -1 -1 EDTA, 2.14 g l of ZnCl2, 2.5 g l of MnCl2·4H2O, 0.3 g l of CoCl2·6H2O, 0.2 g l of -1 -1 -1 CuCl2·2H2O, 0.4 g l of NaMoO4·2H2O, 4.5 g l of CaCl2·2H2O, 2.9 g l of FeCl3·6H2O, -1 -1 -1 1.0 g l of H3BO3, and 0.1 g l of KI. For agar medium, 15 g l of agar was added. Unless otherwise indicated, filter-sterilized glucose (10 g l-1) was used as the only carbon source. DBT, BT, and the other hydrocarbons were dissolved in dimethylformamide (DMF) and added to the sterilized medium. All bacterial cultures were performed in shake flasks at 30 °C, pH 6.9–7.0, with 150-rpm shaking.
Bacterial isolation, identification, and maintenance
G. alkanivorans strain 1B was isolated from soil samples harvested from hydrocarbon- contaminated grounds at EXPO-98 area (Lisbon, Portugal). Soil samples (5 g) were suspended in sterilized SFM medium (100 ml) and supplemented with 5 g l-1 of yeast extract and 0.2 mM DBT. The slurry was filtered through sterilized filter paper (Whatman no. 42) to remove the major solids. Enrichment growth occurred in 1 week to increase the number of bacterial cells and was used to seed a selective medium (SFM without yeast extract) containing DBT as the only S-source. From this growth medium, a serial dilution method was used for bacterial isolation. The isolated bacterial strain that showed a higher ability to grow in DBT was selected for further studies. After a classic microbiologic characterization based on colony morphology techniques, the isolate 1B was sent to Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ), Germany, for biochemical, physiologic, and chemotaxonomic properties and for a comparison of the first 500 bases
74 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
among the 16S rDNA of strain 1B and other 16S rDNA from the DSMZ bacterial database. The bacterium G. alkanivorans strain 1B was routinely maintained on agar SFM medium slants at 4 °C in the dark and was transferred into fresh basal medium once a month. For long-term storage, stock cultures on liquid SFM were stored at –20 °C medium containing 50% (w/v) glycerol.
Substrate utilization
To check the ability of G. alkanivorans strain 1B to use several sulfur and carbon compounds, two independent sets of experiments of bacterial growth were performed. In the first set of experiments, G. alkanivorans strain 1B was cultured in SFM medium supplemented with different sulfur compounds (1 mM) using glucose (5 g l-1) as the carbon source. In the second set of experiments, different carbon compounds (1 g l-1) were used and DBT (200 µM) served as the sole sulfur source.
Analytical methods
Cell growth was monitored by measuring the optical density (OD) of culture at 600 nm. Dry cell weight (DCW) was determined using 0.22-µm cellulose acetate membranes after 18 h at 100 °C. Five hundred microliters of the culture broth was acidified with 25 µl of 4 M HCl, and the organic metabolites were extracted with 2 ml of ethyl acetate during a 5-min vortex. After 1 h, 5 µl of the organic fraction was analyzed to detect and quantify DBT, BT, and other metabolic products using a gas chromatograph (Model CP 9001; Chrompack, Middelburg, The Netherlands) equipped with a flame ionization detector. Chromatography was accomplished over 40 min by using an oven temperature of 120 °C for 5 min followed by a 4 °C min-1 rise up to 250 °C and held for 1 min at this temperature. The injector and detector temperatures were set to 250 and 335 °C, respectively. The carrier gas used was nitrogen. In all gas chromatography (GC) measurements, a calibration curve between a known concentration of anthracene (standard) and different concentrations of 2-HBP, DBT, and BT was used. 2-HBP, DBT, and anthracene were separated using a Chromosorb WAW-
75 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
DMCS column (80–100 mesh) with retention times of 8.2, 14.2, and 15.1 min, respectively. BT and anthracene were separated using a Chromosorb WHP column (100–120 mesh) with retention times of 4.2 and 20.2 min, respectively.
RESULTS AND DISCUSSION
Taxonomic characterization of strain 1B
The screening of novel bacteria from soil samples contaminated with hydrocarbons allowed the isolation of several bacterial strains. Strain 1B was selected for further studies because it showed better growth when cultured using DBT as the only source of sulfur (data not shown). Strain 1B was an aerobic, Gram-positive, catalase-positive, oxidase-negative, and pink/orange-pigmented bacterium. The cells were shown to be short branched hyphae, which disintegrated to rods and coccus-like elements when visualized by phase contrast microscopy. Nonmotile cells generally occurred in groups but sometimes singly. This morphologic appearance fitted the original description of G. alkanivorans (Kummer et al., 1998). Moreover, these data were later confirmed by the DSMZ German culture collection, showing that the chemical composition of the bacterial cell wall was based on the meso form of diaminopimelic acid and the nocardomycolic acid. Additional analysis of the mycolic acids by high-temperature GC showed that the strain synthesized a homologous series of mycolic acids ranging from C54 to C60. The fatty acid pattern revealed unbranched, saturated, and unsaturated fatty acid plus tuberculostearic acid. A 100% sequence similarity of the first 500 bases of the 16S rDNA compared to the strain G. alkanivorans DSM 44369T was also obtained.
Substrate utilization
Considering the complex composition of fossil fuels in terms of sulfur compounds, it is advantageous that a bacterial strain can desulfurize most of those compounds. The ability of
76 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
G. alkanivorans strain 1B to use several sources of carbon and sulfur was investigated. Table 1 presents the results of bacterial growth and the pH of the culture medium after 5 d of cultivation as a function of different sources of carbon.
Table 1. Substrate utilization by G. alkanivorans strain 1B for various sulfur compoundsa.
Substrate Growth (OD600) Final pH
Control 0.23 6.83 Maltose 0.25 6.57 Arabinose 0.21 6.60 Galactose 0.21 6.67 Mannitol 1.08 6.60 Mannose 0.29 6.56 Glucose 1.88 6.37 Sucrose 1.37 6.64 2-HBP 0.26 6.95 Anthrone 0.57 6.92 Benzoate 0.21 6.94 n-Hexadecane 1.18 6.18 Naphthalene 0.23 6.85 Anthracene 0.28 6.93 Tetrahydrofuran 0.46 6.58 DBT sulfone 0.33 6.93
aThe source of carbon was glucose (5 g l-1). The data were obtained after 5 d of culture and are the age values obtained from triplicates, with an SD <5%. The control assay was carried out from a growth medium containing glucose as the source of carbon and with no source of sulfur added.
When DBT was used as the sole carbon and sulfur source (control assay), no significant bacterial growth was detected. Glucose, sucrose, n-hexadecane, and mannitol were in a
77 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
decreasing order the most suitable sources of carbon to support bacterial growth with OD values of >1 unit. Strain 1B showed poor growth on tetrahydrofuran and anthrone, and for the other sources of carbon tested, including DBT sulfone and 2-HBP, no bacterial growth was observed.
Table 2 displays the effect of different sources of sulfur on G. alkanivorans 1B growth and the pH of the culture medium after 5 d of cultivation.
Table 2. Substrate utilization by G. alkanivorans Strain 1B for various sulfur compoundsa.
Substrate Growth (OD600) Final pH
Control 0.27 6.80 DBT 1.18 6.66 DBT sulfone 1.64 6.66 Benzothiophene 0.91 6.67 Elemental sulfur 4.61 4.35 2-Methylthiophene 1.53 6.70 2-Mercaptoethanol 1.76 5.33 Potassium sulfate 5.97 4.49 Sodium sulfide 6.13 5.63 Sodium sulfite 5.75 4.49
aThe source of carbon was glucose (5 g l-1). The data were obtained after 5 d of culture and are the average values obtained from triplicates, with an SD <5%. The control assay was carried out from a growth medium containing glucose as the source of carbon and with no source of sulfur added.
In the control assay, in which no source of sulfur was added to the glucose mineral medium, no significant growth was obtained. Conversely, G. alkanivorans strain 1B showed vigorous growth (>4.5 OD units) on sulfate, sulfide, sulfite, and elemental sulfur and significant growth (>1 OD unit) on DBT, DBT sulfone, BT, 2-methylthiophene, and 2-
78 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
mercaptoethanol. These results clearly demonstrated that G. alkanivorans strain 1B utilized different sources of sulfur-containing aromatic hydrocarbons, including alkylated thiophene, BT, and DBT, suggesting its ability to decrease the sulfur content of fossil fuels.
Metabolic pathway of DBT
In most biodesulfurization studies, DBT has been used as the representative molecule (model) of the thiophenic compounds present in the fossil fuels. When considering a practical application of fossil fuel desulfurization, it is important that the biocatalyst be able to use only DBT as the source of sulfur, to avoid weakening of the combustion value of fossil fuels.
In this context, the metabolic pathway of DBT used by G. alkanivorans strain 1B was studied. Desulfurization of DBT by strain 1B was repressed in the presence of sulfate in the culture medium (data not shown). This repression effect of desulfurization by sulfate has been described to be characteristic of the 4S metabolic pathway (Ohshiro et al., 1996). Further chromatographic analysis for DBT-derived metabolites revealed that 2-HBP was the only metabolite detected. Moreover, the bacterium was able to utilize DBT sulfone as the sole source of sulfur, with the end product still being 2-HBP (data not shown). Regarding the use of these thiophenic compounds as potential carbon sources for G. alkanivorans strain 1B, Table 1 shows that this bacterium was not able to use DBT, DBT sulfone, or 2- HBP as the sole source of carbon, which excludes the alternative Van Afferden (Van Afferden et al., 1990) and Kodama (Kodama et al., 1973) pathways.
Thus, desulfurization of DBT by G. alkanivorans strain 1B can be assumed to proceed as follows: DBT → DBT sulfone → 2-HBP + sulfate. Examination of the described bacterial DBT degradation pathways (Gallagher et al., 1993; Van Afferden et al., 1990; Kodama et al., 1973) reveals that the 4S pathway is the only known bacterial pathway where 2-HBP is produced as the end product of the metabolic pathway.
79 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
Therefore, these results strongly suggest that G. alkanivorans strain 1B uses the 4S metabolic pathway for desulfurization of DBT. This proposal is in agreement with the proposed DBT metabolism for other strains of Gordonia spp. (Rhee et al., 1998; Finkel’shtein et al., 1999).
Kinetics of consumption of DBT by strain 1B
Figure 1 displays the kinetics of the consumption of DBT by G. alkanivorans strain 1B. The consumption of DBT was observed after the first 24 h of culture, but the accumulation of 2- HBP in cultivation broth was observed only after 48 h, which might suggest the existence of a concentration- dependent efflux system.
500 7
6 400 5 M) ) µ 300 4 -1
200 3 DCW (gl
DBT; 2-HBP ( 2 100 1
0 0 0 24 48 72 96 120 144 168 192
Time (hours)
Fig. 1. Time-course of desulfurization of DBT by G. alkanivorans strain 1B. The bacterium was cultivated at 30 °C, pH 7.0, in SFM medium with 10 g l-1 of glucose and 500 µM DBT. () DCW; (c) DBT; ( ) 2-HBP.
80 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
The concentration of DBT decreased from 478 to 168 µM after 120 h of culture, corresponding to a specific desulfurization rate of 1.03 µmol of DBT g-1 of dry cell h-1. This rate is similar to that obtained with another Gordonia sp. strain, CYKS1 (0.917 µmol of DBT g-1 of dry cell h-1), which reduced the concentration of DBT from 320 to 50 µM after 120 h of bacterial cultivation (Rhee et al., 1998).
The maximal extracellular concentration of 2-HBP was about 120 µM, which is only 27% of the consumed DBT (450 µM). This nonstoichiometric accumulation of 2-HBP had also been observed in Bacillus subtilis strain WU-S2B (Kirimura et al., 2001) and in Nocardia sp. Strain CYKS2 (Chang et al., 1998). This fact was probably owing to a transient accumulation of 2-HBP (up to a certain threshold) inside the cell during its formation. On the other hand, Wang & Krawiec (1994) suggested that the difference might be owing to the volatile characteristics of 2-HBP.
Kinetics of Single and Mixed DBT/BT Consumption by Strain 1B
G. alkanivorans strain 1B utilizes both BT and DBT as a single source of sulfur, with DBT the preferred sulfur (Table 2). A Nocardia sp. strain (CYKS2) and a Gordonia sp. strain (CYKS1) that utilize DBT preferably to BT have also been reported (Rhee et al., 1998; Chang et al., 1998). Conversely, other bacterial strains have been described to show a faster metabolism of BT compared with DBT (Tanaka et al., 2002; Gallagher et al., 1993). However, as far as we know, there are not reported data about the bacterial degradation kinetics of mixed DBT/BT sulfur sources.
Figure 2 shows the desulfurizing activity of G. alkanivorans strain 1B grown on glucose mineral medium containing DBT/BT as mixed source of sulfur under equimolar concentrations. The preferred source of sulfur was BT, with DBT used only after 96 h, when the concentration of BT became residual. This suggests the existence of a nonspecific transport system able to uptake both thiophenic compounds, with the higher affinity for BT.
81 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
For this mixed DBT/BT-containing growth medium, the specific desulfurization rates of BT and DBT were 0.954 and 0.813 µmol of DBT g-1 of dry cell h-1, respectively. One hundred percent of both BT and DBT in the culture medium was consumed after 144 and 192 h, respectively. The strains Sinorhizobium sp. KT55 (Tanaka et al., 2001) and Rhodococcus sp. WU-K2R growing on BT as the single source of sulfur only degraded 71 and 59%, respectively, of the initial concentration (100 and 270 µM, respectively) after 5 d of culture (Kirimura et al., 2002). The strain WU-K2R also preferentially degraded BT for a BT/naphthothiophene mixture.
280 5
240 4 200 ) M) -1
µ 3 160
120 2 DCW (g l DBT; BT DBT; ( 80 1 40
0 0 0 24 48 72 96 120 144 168 192
Time (hours)
Fig. 2. Time-course of desulfurization of DBT and BT by G. alkanivorans strain 1B. Both hydrocarbons were added simultaneously to the culture medium in a concentration of 250 µM each. () DCW; (c) DBT; (z) BT.
It was previously described that BT cannot be desulfurized via the DBT-specific pathway and that DBT cannot be desulfurized through the BT-specific pathway for many bacterial species (Kirimura et al., 2002; Gilbert et al., 1998) despite the fact that their metabolic
82 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
pathways apparently are rather similar (Kobayashi et al., 2000). This means that microbial fuel desulfurization technology might need to employ different bacterial systems in a complementary basis for enhancement of the desulfurization rate of the wide range of sulfur sources available in fuel oil.
However, the ability of G. alkanivorans strain 1B to desulfurize DBT, BT, and other sulfur organic compounds suggests the usefulness of this strain for further genetic improvement of its desulfurization rate.
ACKNOWLEDGMENTS
We thank Belarmino Barata (Faculty of Sciences, University of Lisbon) for the gift of soil samples used for the isolation of G. alkanivorans strain 1B. This work was supported by contracts PRAXIS/PBICT/BIO/05/95 and INCO-COPERNICUS-Nº IC15-CT 96-0716.
83 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
REFERENCES
Chang, J.H., Rhee, S.-K., Chang, Y.K., Chang, H.N., 1998. Desulfurization of diesel oils by a newly isolated dibenzothiophene-degrading Nocardia sp. Strain CYKS2. Biotechnol. Prog. 14, 851-855.
Finkel’shtein, Z.I., Baskunov, B.P., Golovlev, E.L., Golovleva, L.A., 1999. Desulfurization of 4,6- dimethyldibenzothiophene and dibenzothiophene by Gordona aichiensis 51. Microbiology. 68, 154-157.
Gallagher, J.R., Olson, E.S., Stanley, D.C., 1993. Microbial desulfurization of dibenzothiophene - a sulfur-specific pathway. FEMS Microbiol. Lett. 107, 31-36.
Gilbert, S.C., Morton, J., Buchanan, S., Oldfield, C., McRoberts, A., 1998. Isolation of a unique benzothiophene-desulfurizing bacterium, Gordona sp. strain 213E (NCIMB 40816), and characterization of the desulphurization pathway. Microbiology. 144, 2545-2553.
Gray, K.A., Mrachko, G.T., Squires, C.H., 2003. Biodesulfurization of fossil fuels Curr. Opin. Microbiol. 6, 229-235.
Grossman, M.J., Lee, M.K., Prince, R.C., Minak-Bernero, V., George, G.N., Pickering, I.J., 2001. Deep desulfurization of extensively hydrodesulfurized middle distillate oil by Rhodococcus sp. strain ECRD-1. Appl. Environ. Microbiol. 67, 1949-1952.
Kayser, K.J., Cleveland, L., Park, H.-S., Kwak, J.-H., Kolhatkar, A., Kilbane II, J.J., 2002. Isolation and characterization of a moderate thermophile, Mycobacterium phlei GTIS10, capable of dibenzothiophene desulfurization. Appl. Microbiol. Biotechnol. 59, 737-745.
Kirimura, K., Furuya, T., Nishii, Y., Ishii, Y., Kino, K., Usami, S., 2001. Biodesulfurization of dibenzothiophene and its derivatives through the selective cleavage of carbon-sulfur bonds by a moderately thermophilic bacterium Bacillus subtilis WU-S2B. J. Biosci. Bioeng. 91, 262-266.
Kirimura, K., Furuya, T., Sato, R., Ishii, Y., Kino, K., Usami, S., 2002. Biodesulfurization of naphthothiophene and benzothiophene through selective cleavage of carbon-sulfur bonds by Rhodococcus sp. strain WU-K2R. Appl. Environ. Microbiol. 68, 3867-3872.
84 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
Kobayashi, M., Toshimitsu, O., Onaka, T., Ishii, Y., Konishi, J., Takaki, M., Okada, H., Ohta, Y., Koizumi, K., Susuki, M., 2000. Desulfurization of alkylated forms of both dibenzothiophene and benzothiophene by a single bacterial strain. FEMS Microbiol. Lett. 187,123-126.
Kodama, K., Umehara, K., Shimizu, K., Nakatani, S., Monioda, Y., Yamada, K., 1973. Identification of microbial products from dibenzothiophene and its proposed oxidation pathway. Agric. Biol. Chem. 37, 45-50.
Konishi, J., Onaka, T., Ishii, Y., Susuki, M., 2000. Demonstration of the carbon-sulfur bond targeted desulfurization of benzothiophene by thermophilic Paenibacillus sp. Strain A11-2 capable of desulfurizing dibenzothiophene. FEMS Microbiol. Lett. 187, 151-154.
Kummer, C., Schumann, P., Stackebrandt, E., 1999. Gordonia alkanivorans sp. nov., isolated from tar-contaminated soil. Int. J. Syst. Bacteriol. 49, 1513-1522.
Monticello, D.J., 1998. Riding the fossil fuel biodesulfurization wave. Chemtech. 28, 38-45.
Ohshiro, T., Izumi, Y., 1999. Microbial desulfurization of organic sulfur compounds in petroleum. Biosci. Biotechnol. Biochem. 63, 1-9.
Ohshiro, T., Suzuki, K., Izumi, Y., 1996. Regulation of dibenzothiophene degrading enzyme activity of Rhodococcus erythropolis D-1. J. Ferment. Bioeng. 81, 121-124.
Oldfield, C., Pogrebinsky, O., Simmonds, J., Olson, E.S., Kulpa, C.F., 1997. Elucidation of the metabolic pathway for dibenzothiophene desulphurization by Rhodococcus sp. strain IGTS8 (ATCC 53968). Microbiology. 143, 2961-2973.
Reichmuth, D.S., Hittle, J.L., Blanch, H.W., Keasling, J.D., 2000. Biodesulfurization of dibenzothiophene in Escherichia coli is enhanced by expression of a Vibrio harveyi oxidoreductase gene. Biotechnol. Bioeng. 67, 72-79.
Rhee, S., Chang, J.H., Chang, Y.K., Chang, H.N., 1998. Desulfurization of dibenzothiophene and diesel oils by a newly isolated Gordona strain, CYKS1. Appl. Environ. Microbiol. 64, 2327-2331.
Schmidt, M., Siebert, W., Bagnall, K.W., 1973. “The chemistry of sulphur, selenium, tellurium and polonium”. Pergamon Texts in Inorganic Chemistry, vol.15, Pergamon Press, Oxford.
85 DBT DESULFURIZATION BY G. alkanivorans STRAIN 1B
Tanaka, Y., Matsui, T., Konishi, J., Maruhashi, K., Kurane, R., 2002. Biodesulfurization of benzothiophene and dibenzothiophene by a newly isolated Rhodococcus strain. Appl. Microbiol. Biotechnol. 59, 325-328.
Tanaka, Y., Onaka, T., Matsui, T., Maruhashi, K., Kurane, R., 2001. Desulfurization of benzothiophene by the gram-negative bacterium, Sinorhizobium sp. KT55. Curr. Microbiol. 43, 187-191.
Van Afferden, M., Schacht, S., Klein, J., Trüper, H.G., 1990. Degradation of dibenzothiophene by Brevibacterium sp. DO. Arch. Microbiol. 153, 324-328.
Wang, P., Krawiec, S., 1994. Desulfurization of dibenzothiophene to 2-hydroxybiphenyl by some newly isolated bacterial strains. Arch. Microbiol. 161, 266-271.
86
Chapter 3
DESULFURIZATION GENES FROM G. alkanivorans
SEQUENCING, CLONING AND EXPRESSION OF THE dsz GENES REQUIRED FOR DIBENZOTHIOPHENE SULFONE DESULFURIZATION FROM Gordonia alkanivorans STRAIN 1B
L. Alves 1, M. Melo1, D. Mendonça1, F. Simões1, J. Matos1, R. Tenreiro2, F.M. Gírio1
1 INETI, Departamento de Biotecnologia, Estrada do Paço do Lumiar, 22, 1649-038 Lisboa, Portugal. 2 FCUL, Departamento de Biologia Vegetal/ Centro de Genética e Biologia Molecular, Campo Grande, 1749-016 Lisboa, Portugal.
ABSTRACT - Biological desulfurization of fossil fuels may offer an alternative process to reduce sulfur dioxide emissions that cause environmental pollution. Gordonia alkanivorans strain 1B is able to convert dibenzothiophene sulfone (DBTS) to 2-hydroxybiphenyl (2- HBP), the final product of 4S pathway. G. alkanivorans genes that code for the enzymes involved in this degrading pathway were PCR amplified using homologous primers based on known sequences from Rhodococcus erythropolis IGTS8. Amplified fragments were further cloned and sequenced. Strain 1B desulfurization genes (dsz) were identified and compared with previously described bacterial genes from other strains. Three open reading frames were identified (dszA, dszB and dszC) and have shown high similarity when compared to those from R. erythropolis IGTS8 (88% for dszA, 88% for dszB and 90% for dszC). G. alkanivorans dszAB genes were further expressed in Escherichia coli. This recombinant strain was able to grow in a culture medium containing dibenzothiophene sulfone (DBTS) as the only sulfur source desulfurizing the same amount of DBTS (0.2 mM) to 2-HBP but 4.5-fold faster than strain 1B. In addition, the recombinant strain could also desulfurize DBTS in LB medium containing other sulfur compounds such as sulfates, showing no sulfate repression of the dszAB genes expression.
Keywords: Biodesulfurization, Dibenzothiophene Sulfone, dsz genes, Gordonia alkanivorans, Escherichia coli, Recombinant bacteria.
89 DESULFURIZATION GENES FROM G. alkanivorans
INTRODUCTION
The combustion of petroleum distillates produces sulfur oxides that contribute to air pollution. In addition, the decreased availability of low sulfur crude oils has resulted in a need to utilize higher sulfur crude oil (Monticello, 1998). Regulatory agencies throughout the world have recognized these problems producing regulations limiting both the sulfur emissions from power plants and the level of sulfur allowed in transportation fuels (Gupta et al., 2005).
Nowadays the petroleum industry treats the crude oil by hydrodesulfurization using extremely high temperature and pressure conditions. Microbiological desulfurization offers the potential of a complementary method for lowering the sulfur content of petroleum products (Folsom et al., 1999). There are many desulfurizing bacteria which has shown the ability to desulfurize DBT producing 2-hydroxybiphenyl (2-HBP) using the specific sulfur pathway known as “4S pathway”. Several model compounds such as DBT, DBTS, and simple mononuclear thiophene derivatives can be used to characterize organic sulfur in coal, coal tars, and crude oils (Constanti et al., 1996).
The dsz genes of Rhodococcus erythropolis strain IGTS8 were the first to be characterized (Denome et al., 1994). These genes are organized in an operon that comprises three open reading frames, transcribed in the same direction coding for two monooxygenases (dszC and dszA) and a desulfinase (dszB). The dsz operon of Mycobacterium phlei GTIS10 (Kayser et al., 2002) and Arthrobacter sp. DS7 (Serbolisca et al., 1999) were cloned and sequenced and found to be identical to that of R. erythropolis IGTS8. However, this homology of the dsz nucleotide sequence is not observed for other bacterial species (Darzins & Mrachko, 2000; Ishii et al., 2000; Park et al., 2003; Kirimura et al., 2004).
The dsz promoter has also been characterized and it was found that the dsz genes expression is strongly repressed by sulfate or other sulfur compounds even in the presence of DBT (Li et al., 1996; Denis-Larose et al., 1997). Gallardo et al. (1997) reported one of the first
90 DESULFURIZATION GENES FROM G. alkanivorans
studies of engineering R. erythropolis IGTS8 dsz genes under the control of heterologous broad-host-range regulatory control to alleviate the mechanism of sulfur repression. Genetic engineering studies are important to obtain recombinant bacterial strains able to rapidly desulfurize fossil fuels using cheaper culture media containing alternative carbon sources even in the presence of sulfates and other sulfur compounds. Indeed, it has been reported that some recombinant bacteria are better desulfurizing strains than the classical wild-type R. erythropolis strain IGTS8 (Gallardo et al., 1997; Matsui et al., 2001a). To achieve a future industrial application it is necessary to obtain new recombinant biocatalysts with an ultra-high desulfurization activity (Ohshiro & Izumi, 1999).
The ability to use several compounds as source of sulfur including DBT and DBTS which are desulfurized to 2-HBP by Gordonia alkanivorans strain 1B was previously studied (Alves et al., 2005). Members of the genus Gordonia are chemoorganotrophs that exhibit a great number of metabolic pathways, which translate a richness of secondary metabolic ability that enables these microorganisms to use a significant range of organic compounds as sources of carbon, energy and micronutrients (Oldfield et al., 1998). The potential of this group of microorganisms for the bioremediation and biodegradation of toxic and/or xenobiotic compounds is still high, and despite the high costs and economic unfeasibility of many processes, there is continuing industrial interest, as indicated by the increasing number of patents linked to the genus Gordonia (Arenskötter et al., 2004).
In this paper, it is described for the first time the dsz gene sequences from the bacterial species G. alkanivorans strain 1B and they were compared with the already known dsz sequences for other bacterial species.
91 DESULFURIZATION GENES FROM G. alkanivorans
MATERIALS AND METHODS
Bacterial strains and growth conditions
G. alkanivorans strain 1B was previously isolated from oil-contaminated soil (Alves et al., 2005). Escherichia coli DH5α and E. coli SG13009 from Qiagen (Hilden, Germany) were used as host strains for general cloning studies. Strain 1B was cultured in sulfur-free mineral
(SFM) medium containing 1.22 g NH4Cl, 2.5 g KH2PO4, 2.5 g Na2HPO4·2H2O, 0.17 g
MgCl2·6H2O per liter of milli-Q water. This medium was supplemented with 0.5 ml of a trace elements solution containing (l): 25 g EDTA, 2.14 g ZnCl2, 2.5 g MnCl2·4H2O, 0.3 g
CoCl2·6H2O, 0.2 g CuCl2·2H2O, 0.4 g NaMoO4·2H2O, 4.5 g CaCl2·2H2O, 2.9 g FeCl3·6H2O, −1 1.0 g H3BO3, 0.1 g KI. Filter sterilized glucose (10 g l ) was used as the only carbon source. For E. coli DH5α the cells were cultured either in the SFM medium supplemented with 0.01% casamino acids and thiamine (5 mg ml−1) or in Luria–Bertani broth (LB) consisting on 1% bactotryptone, 0.5% yeast extract, and 0.5% NaCl. DBTS dissolved in dimethylformamide were added to the sterilized culture media as sulfur source. All G. alkanivorans and E. coli liquid cultivations were performed in shake-flasks under 150 rpm shaking orbital agitation at 30 ºC and 37 ºC, respectively, and pH 7.0.
DNA extraction
G. alkanivorans strain 1B was grown on 150 ml of SFM medium, containing glucose (1%) and DBT (500 µM). Total DNA was extracted using standard procedures (Sambrook et al., 1989) at an early exponential state.
PCR amplification and DNA sequencing of the dsz genes
Desulfurization (dsz) genes of G. alkanivorans strain 1B were PCR amplified using homologous primers based on known sequences from R. erythropolis strain IGTS8 (GenBank accession number U08850), as well as primers based on known sequences from
92 DESULFURIZATION GENES FROM G. alkanivorans
G. alkanivorans strain 1B, that are listed on table 1. The oligonucleotides were purchased from MWG (Ebersberg, Germany). Primers AF and BR were used to amplify dszA and dszB genes of strain 1B, respectively. Primers BCF and CR, were used to amplify dszC gene. The dsz genes sequence of strain 1B was determined by primer walking technique. The flanks of the dsz genes were sequenced using a Universal GenomeWalker™ Kit (Clontech Laboratories, Palo Alto, CA, USA). The protocol was followed as recommended by manufacturer except that SmaI (MBI Fermentas, Vilnius, Lithuania) was also used for the digestion of G. alkanivorans strain 1B total DNA.
Table 1. List of primers used in this study.
Resulting PCR Primer Primer Sequence products
AF 5’- CGCGATGACTCAACAACGAC -3’
BR 5’- CTATCGGTGGCGATTGAGGC -3’ dszAB
FIMCR 5’- AGATCCTCAGGAGGTGAAGC -3’ dszABC
BCF 5’- GCGCCGACTTCCAGCAGCG -3’ dszC 5’- AGATCCTCAGGAGGTGAAGC -3’ CR
ApQE30F 5’- CCGGTACCGCTCAACGGCG -3’ dszAB
BpQE30F 5’- CCGGTACCGCAGGCCGCCTCAGCC -3’ dszB
BpQE30R 5’- CCAAGCTTCTATCGGTGACGG -3’ dszAB / dszB
ApUC19F 5’- CCTTAAGCTTAGAGGACACATACG -3’ dszAB CpUC19R 5’- CCGGTACCCTAGGAGGTGAAGC -3’
AF, BR and FIMCR are R. erythropolis strain IGTS8 dsz sequence specific; the other primers are strain 1B dsz sequence specific. Primers ending with F are forward and with R are reverse. KpnI restriction sites are underlined; HindIII restriction sites are in bold.
93 DESULFURIZATION GENES FROM G. alkanivorans
All PCR amplifications were performed using TaQ DNA polimerase (MBI Fermentas, Vilnius, Lithuania), according to manufacturer’s instructions. The amplification products were analyzed on 1% agarose gel electrophoresis. PCR products of single or multiple bands were purified for further DNA sequencing using the GFX™ PCR DNA and Gel Band Purification Kit (Amersham Biosciences, Piscataway, NJ, USA) and sequenced using PCR primers. Some bands were also cloned on pTZ57R/T vector with Instant T/Aclone™ PCR Product Cloning Kit (MBI Fermentas). Vectors were extracted using Rapid Plasmid Miniprep Kit (Qiagen) and sequenced with vector specific universal primers. The fluorescence-labeled dideoxynucleotide technology was used for DNA sequencing reactions with the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction v2.0 Kit (Applied Biosystems, Foster City, CA, USA). The sequencing fragments were separated using an Applied Biosystems automated DNA sequencer model ABI PRISM 310 Genetic Analyzer.
The sequence analysis was performed using the blasts programs of the National Center for Biotechnology Information. Sequences manipulation and editing were performed using the BioEdit graphic interface. dszAB genes expression in E. coli strains
Vector construction
The plasmid pUC19 (MBI Fermentas) was used to clone strain 1B dszAB genes under the control of lacZ promoter. The expression vector pQE30 was obtain from Qiagen (Hilden, Germany) and was used to over-express dszA and dszB. The nucleotide fragments containing dsz genes were obtained using primers containing either HindIII or KpnI restriction sites (Table1). The restriction endonucleases digestions of both PCR products and vectors were performed according the manufacturer’s protocol.
94 DESULFURIZATION GENES FROM G. alkanivorans
Induction of DszAB proteins
E. coli SG13009 cells were used as the host for pQE30 with dszAB or dszB and were grown overnight at 37 ºC, 200 rpm, on Luria-Bertani (LB) medium supplemented with ampicillin (100 µg ml-1) and kanamicin (25 µg ml-1). Cells were diluted 1:10 in fresh LB-ampicilin- kanamicin medium; 1 mM of IPTG was added to the culture medium when the absorbance reached 0.5. Aliquots of the culture broth were taken before IPTG induction and 2 and 4 h after the induction. Negative controls of E. coli strain SG13009 carrying pQE30 plasmid were also grown under the same conditions. Proteins from the whole cells were separated on a 12.5% SDS-PAGE using the standard protocols.
Desulfurization Assays
E. coli DH5α cells were used as the host for pUC19 carrying dszAB genes and were tested for their ability to desulfurize DBTS inoculating SFM medium supplemented as described by Denome et al. (1994) or LB medium, both containing ampicillin (100 µg ml-1) and 200 µM DBTS. Aliquots from the culture medium were taken periodically in intervals of 1 h. Absorbance was measured, and the desulfurization of DBTS and the production of 2- HBP were assayed by gas-chromatography (GC).
Analytical methods
The cell growth was monitored measuring the absorbance of the culture broth samples at 600 nm. Dry cell weight (DCW) was determined using 0.22 µm cellulose acetate membranes after 18 h at 100 ºC. A 500 µl sample of the culture broth was acidified with 25 µl HCl 4 M and the organic metabolites were extracted with 2 ml ethyl acetate during 5 min in vortex. After 1 h, 5 µl of the organic fraction was analyzed using a gas chromatograph (Model CP 9001, Chrompack, Middelburg, The Netherlands) equipped with a flame ionization detector. The columns used were Chromosorb WAW–DMCS (80 – 100 mesh) and Chromosorb WHP (100-120 mesh). The chromatography was performed as
95 DESULFURIZATION GENES FROM G. alkanivorans
described by Alves et al. (2005). In all GC measurements, anthracene was used as internal standard to minimize variation.
Accession number
The nucleotide sequence data for the desulfurization genes of G. alkanivorans strain 1B were submitted to the GenBank nucleotide sequence databases under the accession number AY678116.
RESULTS AND DISCUSSION dszABC genes sequence analysis
Polymerase chain reaction (PCR) experiments were performed with oligonucleotides based on both nucleotide sequence of R. erythropolis strain IGTS8 and known sequences from G. alkanivorans strain 1B (Table 1). The nucleotide fragments obtained using the primers designed on dsz genes of strain IGTS8 allowed to know some sequences of dsz genes from strain 1B. These sequences were used to construct other primers that are specific for G. alkanivorans. The dszABC genes were PCR amplified and sequenced.
The results obtained allowed the sequencing of 4581 bp of strain 1B dsz operon. The comparison of dsz sequences of G. alkanivorans strain 1B with dsz sequences of R. erythropolis strain IGTS8 (Denome et al., 1994; Piddington et al., 1995) showed an overall sequence similarity of about 89%. The physical arrangement of strain 1B dsz genes presented some differences when compared to strain IGTS8 (Fig. 1). The dsz genes of strain 1B have 1425 bp (dszA), 1098 bp (dszB), and 1251 bp (dszC) including in all cases the stop codon. The dszA of strain 1B was the most divergent from the 3 genes containing 63 additional bp as compared with dszA of strain IGTS8. This potentially novel member group of the dszA gene family has been already reported (Duarte et al., 2001). These authors
96 DESULFURIZATION GENES FROM G. alkanivorans
described evidences for the existence of uncultured organisms with divergent dszA gene sequences. Recently a dszA gene of other Gordonia strain was described with the same size than of strain 1B dszA (Park et al., 2003).
64 bp overlapping
dszA dszB dszC 1422 bp 1095 bp 1248 bp G. alkanivorans strain 1B
Conserved promotor 13 bp region
dszA dszB dszC 1359 bp 1095 bp 1251 bp R. erythropolis strain IGTS8
ATGA
Fig. 1. Comparison of dszABC operons of G. alkanivorans strain 1B and R. erythropolis strain IGTS8.
There is a 4-bp overlap between the initiation and stop codons of dszA and dszB in strain IGTS8 (Denome et al., 1994; Piddington et al., 1995), in Bacillus subtilis WUS2B (Kirimura et al., 2004) and in Paenibacillus sp. A11-2 (Ishii et al., 2000). In G. alkanivorans strain 1B there is a 64 bp overlap between dszA and dszB. Gene partial overlapping was already described in prokaryotes, as the example of genes overlap in paa cluster E. coli W involved in the aerobic catabolism of phenylacetic acid (Ferrández et al., 1998).
97 DESULFURIZATION GENES FROM G. alkanivorans
There is a 13 bp gap between dszB and dszC similar to strain IGTS8. On the other hand dszC has one less codon in strain 1B when compared to that of strain IGTS8. It was found in dsz genes of strain 1B potential ribosome biding sites (rbs) with an AGGA sequence upstream each initiation codon (ATG) as described in strain IGTS8 dsz sequence (Denome et al., 1994). A similar physical arrangement of the dsz genes was described in G. nitida strain CYKS1 (Park et al., 2003). Unfortunately the authors do not present the dsz nucleotide sequence or its accession number of Genbank database. However, in this database the complete dsz nucleotide sequences of both Gordonia sp. strain CYKS2 and G. nitida are described with the accession numbers AY396519 and AY714057, respectively.
The comparison between the dsz sequences of these two Gordonia strains and strain 1B showed an overall sequence similarity 99.6% and 88%, respectively. The identity of the gene sequences of strain 1B with strain IGTS8 and Gordonia sp. strain CYKS2 is shown in Table 2. As expected, strain 1B showed a higher identity in terms of dsz gene sequences with strain CYKS2 than with strain IGTS8. The identity of dsz gene sequences with strain IGTS8 it is between 88% and 90%.
Table 2. Identity values for dsz gene sequences and Dsz protein sequences (based on the dsz aminoacid sequences deducted from the nucleotide sequences) of Gordonia alkanivorans strain 1B compared to Rhodococcus erythropolis strain IGTS8 or Gordonia sp. strain CYKS2.
Identity with IGTS8 Identity with Gordonia sp. Genes/ Proteins sequences (%) strain CYKS2 (%)
dszA / DszA 88.0 / 91.2 99.2 / 99.4
dszB / DszB 88.0 / 86.9 99.7 / 100.0
dszC / DszC 90.0 / 91.1 99.8 / 99.5
98 DESULFURIZATION GENES FROM G. alkanivorans
The results of genome walking revealed around 500 bp 5’ and 500 bp 3’ of the strain 1B dsz genes. These sequences are conserved to strain IGTS8 117 bp upstream the ATG of dszA and 94 bp downstream the stop codon of dszC. It was interesting to find out that the promoter region is similar to strain IGTS8 and the sequence is not conserved upstream this region, where an activator biding site was described for strain IGTS8 (Li et al., 1996). This may suggest that in G. alkanivorans strain 1B there is no such activator protein.
Sulfur repression of strain 1B dsz activity was found in a similar way as described before for strain IGTS8 (Li et al., 1996). In the latter strain the promoter and operator are overlapped. Since that is a conserved region, it is probable that there is also an overlapped operator in strain 1B. However, the region of a potential hairpin structure described as a part of the operator in strain IGTS8 (Li et al., 1996), is not totally conserved in strain 1B and may not be formed since its free energy is considerably higher. Thus, these results allowed the conclusion that the dsz genes of G. alkanivorans strain 1B are expressed under a coordinated expression unit, i.e. an operon.
Amino acid sequence analysis
The identity of strain 1B Dsz protein sequences with strain IGTS8 and Gordonia sp. strain CYKS2 is shown in Table 2. Strain 1B showed a higher identity in terms of Dsz protein sequences with strain CYKS2 than with strain IGTS8. The identity of Dsz protein sequences with strain IGTS8 it is between 87 - 91%.
DszA of strain 1B have 21 more amino acids at the end of the protein sequence comparing to that of strain IGTS8. These two proteins differ in 40 amino acids corresponding to 91.2% of identity. The results of Konishi & Maruhashi (2003) clearly indicated that residue 345 of DszA of strain IGTS8 is involved in determining the specificity of C-S bond cleavage. Similarly to this strain, DszA of strain 1B have the same amino acid in position 345 (glutamine), as well as 12 and 7 conserved residues upstream and downstream of residue 345, respectively. DszB of strains 1B and IGTS8 have the same number of amino acids
99 DESULFURIZATION GENES FROM G. alkanivorans
(365) with 48 different that corresponds to 86.9% identity. The only cysteine residue (position 26) is involved in the DszB activity of R. erythropolis KA2-5-1 (Nakayama et al., 2002). Also DszB of strain 1B contains only one cysteine residue in its structure at the position 27 as observed for DszB of strain IGTS8.
DszC has one less amino acid (lack of one alanine in position 16) than the corresponding protein of strain IGTS8 and 37 different amino acids that correspond to 91.1% identity. The work of Arensdorf et al. (2002) has shown that amino acid 261 of DszC is important for DBT monooxygenase activity. Strain 1B also has a valine in that position as well as 6 and 27 conserved amino acids upstream and downstream of position 261, respectively.
The identity of Dsz protein sequences with strain CYKS2 is between 99 - 100%. DszB from these two strains has 100% identity while DszA have 3 aa changes (Gly16Val, Asn17Lys and Ser31Trp) and DszC have 2 aa changes (Ala40Thr and Val46Ile). Interestingly, DszA of strains IGTS8 and 1B have the same residues in positions 16, 17 and 31. dszAB genes expression in E.coli
Plasmids pQEdszAB and pQEdszB were obtained and expressed in E. coli SG13009 cells under IPTG induction. SDS-PAGE revealed the overexpression of just DszA for pQEdszAB, and DszB for pQEdszB (Fig. 2). The molecular masses were calculated to be 50.4 kDa for DszA and 39.0 kDa for DszB. There is a good correlation between these values and those obtained from the deduced aminoacid sequences, which were 51.9 kDa and 39.2 kDa for DszA and DszB, respectively. The deduced molecular mass for DszC of strain 1B was 45 kDa.
Using primers ApUC19F and CpUC19R (Table 1) a fragment around 4000 bp was amplified, comprising dszAB genes and the rbs upstream dszA. A larger fragment was used to prevent the same problem that occurred with plasmid pQEdszAB. This fragment was kpnI/HindIII digested and linked to pre-digested pUC19. In order to reduce the inhibition
100 DESULFURIZATION GENES FROM G. alkanivorans
effect of sulfate and other sulfur sources, the promoterless fragment of the dsz operon was put under the control of lacZ promoter. Some recombinant E. coli strains containing the plasmid pUCdszAB were further obtained showing the ability to desulfurize DBTS producing 2-HBP.
kDa M C dsz dsz
9 6
4 50 kDa
39 kDa 3
Fig. 2. Coomassie-stained SDS-polyacrylamide 2 gels showing overexpression of the DszA and DszB proteins. Lane 1, molecular weight markers; lane 2, control – strain DH5α without dszAB; lane 3, DszA (~ 50 kDa); lane 4, DszB (~ 39 kDa).
Desulfurization of DBTS by recombinant E.coli
The recombinant E. coli strain DH5α was grown in SFM medium (sulfate free) and LB (with sulfate) with 0.2 mM DBTS producing 2-HBP, which is accumulated in the culture supernatant. E. coli strain DH5α without dsz genes was used as a control, showing no ability to desulfurize DBTS. Recombinant E. coli and Pseudomonas strains were already obtained in several studies (Denome et al., 1994; Ishii et al., 2000; Park et al., 2003; Kirimura et al.,
101 DESULFURIZATION GENES FROM G. alkanivorans
2004; Gallardo et al., 1997; Watanabe et al., 2002). The recombinant strain used in this study was also able to grow in SFM medium with DBTS as the only sulfur source. After 30 h of growth the absorbance was 1.03 and it was detected 47 µM of 2-HBP corresponding to a 24% conversion of DBTS to 2-HBP. A recombinant E. coli W3110 was described carrying the dszABC genes of G. nitida that converted 8.7% and 11.4% of DBT to 2-HBP after 48 h and 60 h, respectively (Park et al., 2003).
150 3
2.5 120 )
2 0
M) 90 µ 1.5 60 2-HBP ( 2-HBP 1 (OD/OD Ln
30 0.5
0 0 0 25 50 75 100 125 150 175 200
Time (h)
Fig. 3. Time courses of 2-HBP production and cell growth by growing cells of recombinant E. coli strain DH5α and wild G. alkanivorans strain 1B. Strain DH5α was grown in LB medium with 100 µg ml-1 ampicilin, 0.2% (v/v) glycerol and strain 1B was grown in SFM medium with 1% (w/v) glucose. Both strains were grown at 30 ºC and 200 rpm with 200 µM DBTS. y - DH5α growth; - HBP production by DH5α; ■ - 1B growth; S - HBP production by 1B.
Figure 3 shows the time course of 2-HBP production by the recombinant E. coli DH5α and the wild strain G. alkanivorans 1B growing in medium containing DBTS. Using the lacZ promoter to control the dszAB expression, the desulfurization activity was not repressed in the presence of other sulfur sources (medium LB) as occurs in G. alkanivorans strain 1B
102 DESULFURIZATION GENES FROM G. alkanivorans
(data not shown). Similar results were already achieved for other recombinant bacteria (Park et al., 2003; Piddington et al., 1995; Matsui et al., 2002). Although the 0.2 mM DBTS in the culture medium was completely depleted by the recombinant E. coli DH5α after 31 h of growth (data not shown), only 0.105 mΜ 2-HBP was detected corresponding to 52.5% of the initial DBTS concentration. For G. alkanivorans strain 1B all DBTS was metabolized after 167 h of cultivation (data not shown), but only 0.138 mM 2-HBP was detected (69%).
These non-stoichiometric values for the bioconversion DBTS/2-HBP were also described previously for a recombinant Rhodococcus strain (Matsui et al., 2001). According to these authors the rest of the 2-HBP produced may have remained inside the cells. The maximum specific desulfurization activity achieved for strain DH5α and strain 1B was 7.1 µmol 2- HBP g-1 (DCW) h-1 and 1.59 µmol 2-HBP g-1(DCW) h-1, respectively and the maximum specific growth was 0.20 h-1 and 0.0192 h-1, respectively. The recombinant strain desulfurizes the same amount of DBTS 4.5-fold faster than strain 1B. Accumulation of 2- HBP follows a direct relationship with growth and this suggests that in stationary growth phase cells are not able to desulfurize DBTS. Gallardo et al. (1997) also described a faster metabolism of DBT using a recombinant Pseudomonas strain: while this strain transformed 95% of the DBT at 24 h of incubation, only 18% of the DBT was transformed by R. erythropolis strain IGTS8.
In this work, new desulfurization genes from G. alkanivorans strain 1B were described and successfully expressed into E. coli DH5α. The dsz operon of G. alkanivorans codes the enzymes responsible of the well-known desulfurization 4S metabolic pathway which confirms our previous physiological data (Alves et al., 2005). So, this bacterium has potential to be used in biodesulfurization studies of fossil fuels since it not decrease the carbon content of desulfurized compounds. Moreover, the expression of the dsz genes in the presence of inorganic sulfur strongly contributes for further use of alternative culture media containing cheaper carbon sources, like those obtained from lignocellulosic or agro- industrial by-products.
103 DESULFURIZATION GENES FROM G. alkanivorans
REFERENCES
Alves, L., Salgueiro, R., Rodrigues, C., Mesquita, E., Matos, J., Gírio, F.M., 2005. Desulfurization of dibenzothiophene, benzothiophene and other thiophene analogues by a newly isolated bacterium, Gordonia alkanivorans strain 1B. Appl. Biochem. Biotechnol. 120, 199-208.
Arensdorf, J.J., Loomis, A.K., DiGrazia, P.M., Monticello, D.J., Pienkos, P.T., 2002. Chemostat approach for the directed evolution of biodesulfurization gain-of-function mutants. Appl. Environ. Microbiol. 68, 691-698.
Arenskötter, M., Bröker, D., Steinbüchel, A., 2004. Biology of the metabolically diverse genus Gordonia. Appl. Environ. Microbiol. 70, 3195-3204.
Constanti, M., Giralt, J., Bordons, A., 1996. Degradation and desulfurization of dibenzothiophene sulfone and other sulfur compounds by Agrobacterium MC501 and a mixed culture. Enzyme Microb. Technol. 19, 214–219.
Darzins, A., Mrachko, G.T., 2000. Sphingomonas biodesulfurization catalyst. U.S. Patent nº 6133016.
Denis-Larose, C., Labbé, D., Bergeron, H., Jones, A.M., Greer, C.W., Al-Hawari, J., Grossman, M.J., Sankey, B.M., Lau, P.C.K., 1997. Conservation of plasmid-encoded dibenzothiophene desulfurization genes in several Rhodococci. Appl. Environ. Microbiol. 63, 2915-2919.
Denome, S.A., Oldfield, C., Nash, L.J., Young, K.D., 1994. Characterization of desulfurization genes from Rhodococcus sp. strain IGTS8. J. Bacteriol. 176, 6707-6716.
Duarte, G.F., Rosado, A.S., Seldin, L., Araujo, W., Van Elsas, J.D., 2001. Analysis of bacterial community structure in sulfurous-oil-containing soils and detection of species carrying dibenzothiophene desulfurization (dsz) genes. Appl. Environ. Microbiol. 67, 1052-1062.
Ferrandez, A., Minambres, B., Garcia, B., Olivera, E.R., Luengo, J.M., Garcia, J.L., Diaz, E., 1998. Catabolism of phenylacetic acid in Escherichia coli - Characterization of a new aerobic hybrid pathway. J. Biol. Chem. 273, 25974-25986.
Folsom, B.R., Schieche, D.R., DiGrazia, P.M., Werner, J., Palmer, S., 1999. Microbial desulfurization of alkylated dibenzothiophenes from a hydrodesulfurized middle distillate by Rhodococcus erythropolis I-19. Appl. Environ. Microbiol. 65, 4967-4972.
104 DESULFURIZATION GENES FROM G. alkanivorans
Gallardo, M.E., Ferrandez, A., De Lorenzo, V., Garcia, J.L., Diaz, E., 1997. Designing recombinant Pseudomonas strains to enhance biodesulfurization. J. Bacteriol. 179, 7156-7160.
Gupta, N., Roychoudhury, P.K., Deb, J.K., 2005. Biotechnology of desulfurization of diesel: prospects and challenges. Appl. Microbiol. Biotechnol. 66, 356-366.
Ishii, Y., Konishi, J., Suzuki, M., Maruhashi, K., 2000. Cloning and expression of the gene encoding the thermophilic NAD(P)H-FMN oxidoreductase coupling with the desulfurization enzymes from Paenibacillus sp. A11-2. J. Biosci. Bioeng. 90, 591-599.
Kayser, K.J., Cleveland, L., Park, H.-S., Kwak, J.-H., Kolhatkar, A., Kilbane II, J.J., 2002. Isolation and characterization of a moderate thermophile, Mycobacterium phlei GTIS10, capable of dibenzothiophene desulfurization. Appl. Microbiol. Biotechnol. 59, 737-745.
Kirimura, K., Harada, K., Iwasawa, H., Tanaka, T., Iwasaki, Y., Furuya, T., Ishii, Y., Kino, K., 2004. Identification and functional analysis of the genes encoding dibenzothiophene-desulfurizing enzymes from thermophilic bacteria. Appl. Microbiol. Biotechnol. 65, 703–713.
Konishi, J., Maruhashi, K., 2003. Residue 345 of dibenzothiophene (DBT) sulfone monooxygenase is involved in C–S bond cleavage specificity of alkylated DBT sulfones. Biotechnol. Lett. 25, 1199– 202.
Li, M.Z., Squires, C.H., Monticello, D.J., Childs, J.D., 1996. Genetic analysis of the dsz promoter and associated regulatory regions of Rhodococcus erythropolis IGTS8. J. Bacteriol. 178, 6409- 6418.
Matsui, T., Hirasawa, K., Koizumi, K., Maruhashi, K., Kurane, R., 2001a. Optimization of the copy number of dibenzothiophene desulfurizing genes to increase the desulfurization activity of recombinant Rhodococcus sp. Biotechnol. Lett. 23, 1715-1718.
Matsui, T., Hirisawa, K., Konishi, J., Tanaka, Y., Maruhashi, K., Kurane, R., 2001. Microbial desulfurization of alkylated dibenzothiophene and alkylated benzothiophene by recombinant Rhodococcus sp. strain T09. Appl. Microbiol. Biotechnol. 56, 196-200.
Matsui, T., Noda, K., Tanaka, Y., Maruhashi, K., Kurane, R., 2002. Recombinant Rhodococcus sp. strain T09 can desulfurize DBT in the presence of inorganic sulphate. Current Microbiol. 45, 240- 244.
105 DESULFURIZATION GENES FROM G. alkanivorans
Monticello, D.J., 1998. Riding the fossil fuel biodesulfurization wave. Chemtech. 28, 38-45.
Nakayama, N., Matsubara, T., Ohshiro, T., Moroto, Y., Kawata, Y., Koizumi, K., Hirakawa, Y., Suzuki, M., Maruhashi, K., Izumi, Y., Kurane, R., 2002. A novel enzyme, 2 ’-hydroxybiphenyl-2- sulfinate desulfinase (DszB), from a dibenzothiophene-desulfurizing bacterium Rhodococcus erythropolis KA2-5-1: gene overexpression and enzyme characterization. Biochim. Biophys. Acta. 1598, 122–30.
Ohshiro, T., Izumi, Y., 1999. Microbial desulfurization of organic sulfur compounds in petroleum. Biosci. Biotechnol. Biochem. 63, 1-9.
Oldfield, C., Wood, N.T., Gilbert, S.C., Murray, F.D., Faure, F.R., 1998. Desulphurisation of benzothiophene and dibenzothiophene by actinomycete organisms belonging to the genus Rhodococcus, and related taxa. Antonie van Leeuwenhoek. 74, 119-132.
Park, S.J., Lee, I.-S., Chang, Y.K., Lee, S.Y., 2003. Desulfurization of dibenzothiophene and diesel oil by metabolically engineered Escherichia coli. J. Microbiol. Biotechnol. 13, 578-583.
Piddington, C.S., Kovacevich, B.R., Rambosek, J., 1995. Sequence and molecular characterization of a DNA region encoding the dibenzothiophene desulfurization operon of Rhodococcus sp. strain IGTS8. Appl. Environ. Microbiol. 61, 468-475.
Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Serbolisca, L., de Ferra, F., Margarit, I., 1999. Manipulation of the DNA coding for the desulphurizing activity in a new isolate of Arthrobacter sp. Appl. Microbiol. Biotechnol. 52, 122- 126.
Watanabe, K., Noda, K., Ohta, Y., Maruhashi, K., 2002. Desulfurization of light gas oil by a novel recombinant strain from Pseudomonas aeruginosa. Biotechnol. Lett. 24, 897-903.
106
Chapter 4
ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
EFFECT OF ZINC AND OTHER METAL IONS ON THE PERFORMANCE OF DIBENZOTHIOPHENE DESULFURIZATION BY Gordonia alkanivorans STRAIN 1B
L. Alves 1, J. Matos1, R. Tenreiro2, F.M. Gírio1
1INETI, Departamento de Biotecnologia, Estrada do Paço do Lumiar, 22, 1649-038 Lisboa, Portugal. 2Instituto de Ciência Aplicada e Tecnologia - Centro de Genética e Biologia Molecular, Edifício ICAT, Campus da FCUL, Campo Grande, 1749-016 Lisboa, Portugal.
ABSTRACT - Biodesulfurization of fossil fuels may offer an alternative process to obtain more environmental friendly fuels. Gordonia alkanivorans strain 1B is able to convert dibenzothiophene (DBT) to 2-hydroxybiphenyl (2-HBP), the final product of 4S pathway. The composition of the culture medium used to cultivate desulfurizing bacteria is an important factor, because some of the metal ions can be necessary as enzymatic cofactors. In this work, the effect of several metal ions in the growth and DBT desulfurization by G. alkanivorans was studied. From all the metal ions used, only the absence of Zinc has affected significantly the cell growth and the desulfurization rate. By increasing the concentration of Zn from 1 to 10 mg l-1, 2-HBP productivity was improved by 26%. The absence of Zn2+, when sulfate was also used as the only sulfur source, did not cause any difference in the bacterial growth. Resting cells grown in the presence of Zn2+ presented a 2- HBP specific productivity of 2.29 µmol g-1 (DCW) h-1, 7.6-fold higher than the specific productivity obtained by resting cells grown in the absence of Zn2+ (0.30 µmol g-1 (DCW) h-1). These results suggest that Zinc might be an important cofactor for one or more desulfurizing enzymes of strain 1B.
Keywords: Dibenzothiophene, Biodesulfurization, Gordonia alkanivorans, Zinc, enzyme cofactors.
110 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
INTRODUCTION
Sulfur is one of the major contaminants present in a large number of hydrocarbons of fossil fuels. For the becoming years, treating crude oil with high sulfur content is the reality for refineries, while at the same time they will face increasing pressure to reduce carbon dioxide emissions and make ultra-low sulfur transportation fuels. About 72% of oil is refined into transportation fuels in the US and this yield will be increasingly hard to obtain as the quantity of crude declines (Kilbane, 2006). Currently, refineries use a physical-chemical process called hydrodesulfurization to remove sulfur. This process is very expensive due to the use of hydrogen gas and high temperature and pressure. With increasing restrictions on the sulfur emissions from fossil fuels, it is of great importance to investigate alternative methods to remove this pollutant from oil compounds. Biodesulfurization (BDS) of fossil fuels may offer an alternative process to obtain more environmental friendly fuels and, in the last 15 years, several studies in the field of BDS were reported, mostly involving bacterial strains from the genus Rhodococcus.
The impetus of biodesulfurization was the discovery of a highly specific 4S metabolic pathway in the organism Rhodococcus erythropolis IGTS8 (Gray et al., 2003). This metabolic pathway integrates two monoxygenases, one desulfinase and one flavin reductase. More recently, other bacterial genera were reported with the ability to use the 4S pathway to desulfurize DBT (Alves et al., 2007; Konishi et al., 2000; Lu et al., 1999). However, the biodesulfurization studies reported for these new genera are very few compared with the knowledge obtained during the last decade for the well studied Rhodococcus sp. Although the non-Rhodococcus genera have also a good potential to be used in the biodesulfurization of fossil fuels, the optimization of the cultivation conditions, as well as using molecular tools to obtain better biocatalysts, are necessary.
In fact, the key research needs to upgrade fossil fuels are the development of desulfurization biocatalysts with higher specific activity, broader substrate range and higher thermal tolerance (Kilbane, 2006). The composition of the culture medium is an important parameter
111 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
to grow desulfurizing bacteria, because some of the metal ions can be necessary as enzymatic cofactors. Ohshiro et al. (1999) reported that the DszA activity from R. erythropolis D-1 was inhibited by EDTA, suggesting that a metal might be involved in its activity. Molecular characterization of other bacterial monooxygenases revealed that some metal ions such as zinc, copper and iron are present in the active site of the enzymes (Lieberman & Rozenzweig, 2005; Yoshizawa & Yumura, 2003; Yu, et al., 2003).
Metal ions, as part of metalloenzymes, are essential for numerous biocatalytic processes. In many cases, metalloenzymes require specific metal ions to achieve catalytic functionality, e.g. zinc for hydrolytic activities or iron for redox proteins (Schilling et al., 2005). Zinc is an integral component of a large number and variety of proteins involved in a multiplicity of vital processes accounting for its key role in metabolism, transmission of the genetic message, growth and development. The chemically stable but stereo chemically flexible, nontoxic nature of zinc, combined with its amphoteric properties, is the basis for the biochemical organization of a series of zinc binding motifs critical to life and its perpetuation (Vallee & Auld, 1993). More than 300 zinc enzymes covering all six classes of enzymes have been discovered, and in most cases it is an essential cofactor for the observed biological function of these metalloenzymes (McCall et al., 2000).
In this study, the effect of the absence of several metal ions in the culture medium, particularly zinc ion, was examined for growth and dibenzothiophene desulfurization by G. alkanivorans strain 1B.
112 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
MATERIALS AND METHODS
Bacterial strain and growth conditions
Gordonia alkanivorans strain 1B was previously isolated in our laboratory from oil- contaminated soil (Alves et al., 2005). This bacterium was cultured in sulfur-free mineral
(SFM) medium containing 1.22 g NH4Cl, 2.5 g KH2PO4, 2.5 g Na2HPO4.2H2O, 0.17 g
MgCl2.6H2O per liter of milli-Q water. This medium was supplemented with 0.5 ml of trace elements containing (per liter): 25 g EDTA, 2.14 g ZnCl2, 2.5 g MnCl2.4H2O, 0.3 g
CoCl2.6H2O, 0.2 g CuCl2.2H2O, 0.4 g NaMoO4.2H2O, 4.5 g CaCl2.2H2O, 2.9 g FeCl3.6H2O,
1.0g H3BO3, 0.1 g KI. For the assays in which a single metal ion was removed, the culture medium was prepared adding the metal compounds separately instead of adding the trace elements solution. Filter sterilized glucose (10 g l-1) was used as the only carbon source. 500 µM DBT or DBT sulfone (both dissolved in dimethylformamide) or sulfate was added as sulfur source to the sterilized culture media. All G. alkanivorans cultivations were performed in shake-flasks at 30 ºC, pH 7.5, and 150 rpm shaking, at least in duplicates.
Resting cells
In order to investigate the effect of Zn2+ ion in desulfurization activity of resting cells of G. alkanivorans strain 1B, 500 ml of culture broth (SFM medium supplemented with 10 g l-1 glucose and 150 µM DBT), with and without zinc, was taken at a late growth phase (7 days after inoculation). Cells were harvested by centrifugation at 7500 × g and 4 ºC for 10 min. The harvested cells were washed twice with a saline solution (0.85% NaCl) and kept in 0.1 M phosphate solution (pH 7.0) at 4 ºC. Some of the resting cells pre-grown without zinc were incubated overnight with 1 mg l-1 zinc and then used in a desulfurization assay.
113 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
Analytical methods
The cell growth was monitored by measuring the absorbance of the culture broth samples at 600 nm. Dry cell weight (DCW) was determined using 0.22 µm cellulose acetate membranes after 18 h at 100 ºC. For gas chromatography (GC) organic metabolites quantification, 750 µl of the culture broth was acidified with 25 µl HCl 4 M and the metabolites were extracted with 1.5 ml ethyl acetate during 5 min in vortex. After 1 h, 5 µl of the organic fraction was analyzed using a gas chromatograph (Model CP 9001, Chrompack, Middelburg, The Netherlands) equipped with a flame ionization detector. The columns used were Chromosorb WAW–DMCS (80 – 100 mesh) and Chromosorb WHP (100-120 mesh). The chromatography was accomplished over 40 min, by using an oven temperature of 120 ºC for 5 min, followed by a 4 ºC min-1 rise up to 250 ºC and held for 1 min at this temperature. The injector and detector temperatures were set to 250 ºC and 335 ºC, respectively. Carrier gas used was nitrogen. In all GC measurements, anthracene was used as an internal standard to minimize the variation.
Chemicals
DBT (99%) was obtained from Acros Organics, DBT sulfone (97%) from Aldrich, 2-HBP from Sigma, DMF from Riedel-deHaën, anthracene and ethyl acetate from Merck. All other materials were of the highest purity commercially available and were used without further purification.
RESULTS AND DISCUSSION
The effect of the absence of several metal ions in the culture medium on growth of Gordonia alkanivorans strain 1B was investigated. Some metal ions could be necessary to the structure of the Dsz enzymes, and consequently to their activity. The bacterial strain was
114 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
grown in SFM media supplemented with different metal solutions. The metal ions studied 2+ 2+ 3+ 2+ 2- 2+ 2+ were Ca , Co , Fe , Mn , MoO4 , Cu and Zn (Table 1).
Table 1. Effect of removal of some metal ions on G. alkanivorans strain 1B growth and 2-HBP production after 7 days of culture.
-1 Metal ion Abs600 Final pH DCW (g l ) 2-HBP (µM)
Negative control 9.29 5.69 3.6 133.2 Positive control 2.66 6.64 1.3 91.1 - Fe3+ 6.61 6.06 2.2 159.9 - Co2+ 9.30 5.76 3.3 131.1 - Cu2+ 9.35 6.02 3.4 97.8 - Ca2+ 8.99 6.02 3.4 107.6 - Mn2+ 7.92 5.99 2.7 153.0 2- - MoO4 10.25 5.89 3.7 115.4 - Zn2+ 3.60 6.58 1.4 97.2
Note: G. alkanivorans was grown in SFM medium supplemented with 1 g l-1 glucose as carbon source and 250 µM DBT as sulfur source. As positive and negative controls, the culture medium contained all and none of the metal ions tested, respectively.
As positive and negative controls, the culture medium contained none and all of the metal ions tested, respectively. The absence of all metal ions causes a decrease in biomass and 2-HBP production of about 64% and 32%, respectively. In addition, the absence of Fe3+ and Mn2+ enhanced DBT desulfurization, while Co2+ had no effect. Furthermore, a 2+ 2- decrease in 2-HBP production was detected in the absence of Ca and MoO4 . The most significant decrease in 2-HBP production was detected in the absence of Zn2+ and Cu2+. However, the absence of Cu2+ did not significantly decrease the production of biomass, comparing to the value obtained without any metal ions. The effect of zinc in both growth and 2-HBP production suggests that this metal might have an important role in the
115 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
metabolism of G. alkanivorans strain 1B. Thus, zinc was selected for further growth and desulfurization studies by strain 1B.
3
2 A
600 nm 600 1
OD Ln 0
-1
300
250 B
200 M)
µ 150
( 2-HBP 100
50
0 024681012
Time (days)
Fig. 1. Time course of (A) growth and (B) 2-HBP production by G. alkanivorans strain 1B. The strain was grown at 30 ºC and 150 rpm, in SFM medium with DBT as sulfur -1 2+ -1 2+ source and several zinc concentrations: × - 25 mg l Zn ; U - 10 mg l Zn ; - 1 mg l-1 Zn2+; - without Zn2+. The results presented are the mean values of duplicates with a standard deviation less than 5%.
116 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
The effect of Zn2+ concentration on strain 1B growth and 2-HBP production (Fig 1) was investigated. The increase of the concentration of this metal ion in the culture medium enabled both a higher bacterial growth (Fig 1A) and a higher 2-HBP production (Fig 1B). 10 and 25 mg l-1 Zn2+ in the culture medium showed a greater effect in the 2-HBP -1 2+ -1 production than 1 mg l Zn . The addition of 1 mg l zinc has increased both µmax by 32%, from 0.019 h-1 to 0.025 h-1, and DCW around 2-fold (data not shown), comparing with the absence of zinc. Further increases of Zn2+ did not affect the growth rate. The 2- HBP productivity after 10 days of culture (Fig. 1B) in the absence of zinc was only 0.51 µM h-1.
The addition of 1, 10 and 25 mg l-1 Zn2+ has promoted a 2-HBP productivity of 0.91, 1.15 and 1.09 µM h-1, corresponding to an increase of about 78.4, 125.5 and 113.7%, respectively. A slight decrease in 2-HBP productivity was observed with 25 mg l-1 Zn2+, as compared to the performance of G. alkanivorans with 10 mg l-1 Zn2+, possibly due to an inhibitory effect of this Zn2+ concentration in the culture medium. Although zinc is an essential trace element used in the majority of the bacterial culture media, it is known to be a potent inhibitor of the respiratory electron transport system (Beard et al., 1995).
Previous studies in our lab, using G. alkanivorans strain 1B (Alves et al., 2005), have shown a 2-HBP specific productivity of 1.03 µmol g-1(DCW) h-1 when cultured with 0.5 mg l-1 Zn2+, 20% higher than the specific productivity obtained without zinc in the present work (data not shown). In addition, the same effect was observed in the presence and absence of Zn when DBT sulfone was used instead DBT (data not shown).
Moreover, the effect of zinc absence was analyzed during growth of strain 1B using either DBT or sulfate as the only sulfur source (Fig. 2). When G. alkanivorans strain 1B was grown in sulfate containing SFM, no difference was observed in the bacterial growth in the presence or absence of Zn2+. In contrast, when DBT was used as the only sulfur source, G. alkanivorans strain 1B presented a higher growth in the presence of the zinc ion. This difference was only significant after three days of culture, probably due to a
117 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
residual concentration of zinc present in the culture medium. After this time, the bacterial culture showed an increased growth with zinc of 38% after 7 days, as compared to growth without Zn2+. As expected, strain 1B grows better in the culture media with sulfate compared with the one with DBT, since sulfate is more easily assimilated as sulfur source. From these data it is suggested that zinc might have a key physiological role in the metabolism of DBT desulfurization.
3.5
3
2.5 ) 0 2
1.5 Ln (OD/OD Ln 1
0.5
0 012345678
Time (days)
Fig. 2. Time course of growth of G. alkanivorans strain 1B. The strain was grown at 30 ºC and 150 rpm, in SFM medium with DBT or sulfate as sulfur source and with or 2+ 2+ 2+ without zinc: Q - Sulfate without Zn ; S - Sulfate with Zn ; - DBT with Zn ; ♦ - DBT without Zn2+.
To confirm these results, some assays were performed with resting cells pre-grown in the presence or absence of zinc. The data obtained confirmed the importance of Zn2+ for DBT desulfurization (Fig. 3). The resting cells pre-grown in the presence of Zn2+ displayed a 2- HBP specific productivity of 2.29 µmol g-1 (DCW) h-1, 7.6-fold higher than the specific productivity obtained by resting cells pre-grown in the absence of Zn2+ (0.30 µmol g-1 (DCW) h-1).
118 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
An additional assay was carried out using G. alkanivorans resting cells pre-grown in the absence of zinc after overnight incubation with 1 mg l-1 Zn2+ (Fig 3).
12
10 )
8 DCW ( -1
) g 6 2-HBP 4 mol ( µ 2
0 01234567
Time (h)
Fig. 3. Time course of 2-HBP production by resting cells of G. alkanivorans strain 1B: S - Resting cells pre-grown in the absence of zinc; - Resting cells pre-grown in the 2+ presence of zinc; Q - Resting cells after overnight incubation with Zn .
The results obtained showed an increase of 70% on 2-HBP production by the resting cells incubated with zinc. However, the value of 2-HBP production obtained is still much smaller than the one obtained using resting cells pre-grown in the presence of Zn2+.
The enzymes of R. erythropolis involved in the DBT desulfurization have already been purified and characterized (Matsubara et al., 2001; Ohshiro et al., 1999; Ohshiro et al., 1997; Watkins et al., 2003). Ohshiro et al. (1997) showed that the metal ions Cu2+, Zn2+ and Mn2+ significantly inhibited DBT monooxygenase activity of Rhodococcus erythropolis D- 1. Conversely, a positive effect on DBT sulfone monooxygenase activity was reported when
119 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
Al3+, Cd2+ and Zn2+ were added (Ohshiro et al., 1999). These results suggested that SH group(s) and metal(s) might be involved in these enzyme structures. On the contrary, the enzyme 2-(2’-hydroxyphenyl)benzenesulfinate desulfinase of R. erythropolis does not require a metal cofactor for catalysis (Nakayama et al., 2002; Watkins et al., 2003). In addition, Watkins et al. (2003) showed that the presence of zinc reduced the activity of DszB by 50%, due to the interference of substrate binding or catalysis. To date, there are no reports on the purification and characterization of Dsz enzymes from Gordonia sp.
This work shows the importance of zinc for the process of growth and DBT desulfurization by G. alkanivorans strain 1B. Our results suggest that zinc could be an essential metal cofactor for the catalysis of at least one of the enzymes involved in the metabolism of DBT desulfurization by G. alkanivorans strain 1B. Thus, the composition of the culture medium used to grow strain 1B must contain Zn2+ in a 10 mg l-1 concentration, instead of 0.5 mg l-1, to obtain a maximal desulfurization rate. In the future, it will be useful to purify and characterize the Dsz enzymes of strain 1B, in order to understand the role of zinc and other metal ions in their structure and catalysis.
This work has been supported by the contract POCTI/AMB/59108/04.
120 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
REFERENCES
Alves, L., Salgueiro, R., Rodrigues, C., Mesquita, E., Matos, J., Gírio, F.M., 2005. Desulfurization of dibenzothiophene, benzothiophene and other thiophene analogues by a newly isolated bacterium, Gordonia alkanivorans strain 1B. Appl. Biochem. Biotechnol. 120, 199-208.
Beard, S.J., Hughes, M.N., Poole, R.K., 1995. Inhibition of the cytochrome BD-terminated NADH oxidase system in Escherichia coli K-12 by divalent metal-cations. FEMS Microbiol. Lett. 131, 205-210.
Gray, K.A., Mrachko, G.T., Squires, C.H., 2003. Biodesulfurization of fossil fuels. Curr. Opin. Microbiol. 6, 229-235.
Gray, K.A., Pogrebinsky, O.S., Mrachko, G.T., Xi, L., Monticello, D.J., Squires, C.H., 1996. Molecular mechanisms of biocatalytic desulfurization of fossil fuels. Nat. Biotechnol. 14, 1705- 1709.
Kilbane, J.J. II., 2006. Microbial biocatalyst developments to upgrade fossil fuels. Curr. Opin. Microbiol. 17:305-314.
Konishi, J., Onaka, T., Ishii, Y., Susuki, M., 2000. Demonstration of the carbon-sulfur bond targeted desulfurization of benzothiophene by thermophilic Paenibacillus sp. Strain A11-2 capable of desulfurizing dibenzothiophene. FEMS Microbiol. Lett. 187, 151-154.
Lieberman, R.L., Rosenzweig, A.C., 2005. Crystal structure of a membrane-bound metalloenzyme that catalyses the biological oxidation of methane. Nature. 434, 177-182.
Lu, J., Nakajima-Kambe, T., Shigeno, T., Ohbo, A., Nomura, N., Nakahara, T., 1999. Biodegradation of dibenzothiophene and 4,6-dimethyldibenzothiophene by Sphingomonas paucimobilis strain TZS-7. J. Biosci. Bioeng. 88, 293-299.
Matsubara, T., Ohshiro, T., Nishina, Y., Izumi, Y., 2001. Purification, characterization, and overexpression of flavin reductase involved in dibenzothiophene desulfurization by Rhodococcus erythropolis D-1. Appl. Environ. Microbiol. 67, 1179-1184.
McCall, K.A., Huang, C.C., Fierke, C.A., 2000. Function and mechanism of zinc metalloenzymes. J. Nutr. 130, 1437S-1446S.
121 ZINC EFFECT ON DBT DESULFURIZATION BY G. alkanivorans
Nakayama, N., Matsubara, T., Ohshiro, T., Moroto, Y., Kawata, Y., Koizumi, K., Hirakawa, Y., Suzuki, M., Maruhashi, K., Izumi, Y., Kurane, R., 2002. A novel enzyme, 2’-hydroxybiphenyl- 2-sulfinate desulfinase (DszB), from a dibenzothiophene-desulfurizing bacterium Rhodococcus erythropolis KA2-5-1: gene overexpression and enzyme characterization. Biochim. Biophys. Acta. 1598, 122-130.
Ohshiro, T., Kojima, T., Torii, K., Kawasoe, H., Izumi, Y., 1999. Purification and characterization of dibenzothiophene (DBT) sulfone monooxygenase, an enzyme involved in DBT desulfurization, from Rhodococcus erythropolis D-1. J. Biosc. Bioeng. 88, 610-616.
Ohshiro, T., Suzuki, K., Izumi, Y., 1997. Dibenzothiophene (DBT) degrading enzyme responsible for the first step of DBT desulfurization by Rhodococcus erythropolis D-1: purification and characterization. J. Ferment. Bioeng. 83, 233-237.
Schilling, O., Vogel, A., Kostelecky, B., da Luz, H.N., Spemann, D., Spath, B., Marchfelder, A., Troger, W., Mayer-Klaucke, W., 2005. Zinc- and iron-dependent cytosolic metallo-beta- lactamase domain proteins exhibit similar zinc-binding affinities, independent of an atypical glutamate at the metal-binding site. Biochem. J. 385, 145-153.
Vallee, B.L., Auld, D.S., 1993. New perspectives on zinc biochemistry: cocatalytic sites in multi-zinc enzymes. Biochemistry. 32, 6493-6500.
Watkins, L.M., Rodriguez, R., Schneider, D., Broderick, R., Cruz, M., Chambers, R., Ruckman, E., Cody, M., Mrachko, G.T., 2003. Purification and characterization of the aromatic desulfinase, 2- (2’-hydroxyphenyl)benzenesulfinate desulfinase. Arch. Biochem. Biophys. 415, 14-23.
Yoshizawa, K., Yumura, T., 2003. A non-radical mechanism for methane hydroxylation at the diiron active site of soluble methane monooxygenase. Chem. Eur. J. 9, 2347-2358.
Yu, S.S.F., Chen, K.H.C., Tseng, M.Y.H., Wang, Y.S., Tseng, C.F., Chen, Y.J., Huang, D.S., Chan, S.I., 2003. Production of high-quality particulate methane monooxygenase in high yields from Methylococcus capsulatus (Bath) with a hollow-fiber membrane bioreactor. J. Bacteriol. 185, 5915-5924.
122
Chapter 5
RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
DIBENZOTHIOPHENE DESULFURIZATION BY Gordonia alkanivorans STRAIN 1B USING RECYCLED PAPER SLUDGE HYDROLYZATE
L. Alves1, S. Marques1, J. Matos1, R. Tenreiro2, F.M. Gírio1
1INETI, Departamento de Biotecnologia, Estrada do Paço do Lumiar, 22, 1649-038 Lisboa, Portugal. 2Instituto de Ciência Aplicada e Tecnologia - Centro de Genética e Biologia Molecular, Edifício ICAT, Campus da FCUL, Campo Grande, 1749-016 Lisboa, Portugal.
ABSTRACT - Enzymatic hydrolyzates of recycled paper sludge (RPS) were tested as feedstock for biological desulfurization by Gordonia alkanivorans strain 1B. Only the hydrolyzate obtained after enzymatic mixture dialysis (dialyzed hydrolyzate) allowed dibenzothiophene (DBT) desulfurization, in spite of faster bacterial growth on non-dialyzed hydrolyzate did occur. For dialyzed RPS hydrolyzate, 250 µM DBT was consumed after 96 hours displaying a maximum specific productivity of 2-hydroxybiphenyl of 1.1 µmol g-1(dry cell weight) h-1. A comparison of the kinetic of biodesulfurization was assessed according to the type of hydrolyzate supplementation. Complete consumption of DBT was observed upon the addition of only phosphates and ammonia although the further addition of zinc did increase the 2-hydroxybiphenyl production by 14%. Strain 1B was able to desulfurize a model oil containing DBT, 4-methylDBT and 4,6-dimethylDBT, reducing by 63% the total sulfur content in 168 hours.
Keywords: Biodesulfurization, Dibenzothiophene, Gordonia alkanivorans, Biocatalyst, Industrial wastes, Recycled paper sludge.
126 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
INTRODUCTION
The rising consumption of fossil fuels along the world, due to the growing industrial activity, provokes an increase on waste products causing atmospheric pollution. These wastes products are particulates and various gases such as sulfur dioxide, nitrogen oxides and volatile organic compounds that are produced due to impurities in the fuels (Gupta et al., 2005). Thus, environmental authorities along the world have recognized this problem and are imposing increasingly stringent restrictions on the maximum sulfur concentration allowed in the fossil fuels. The process currently utilized in refineries to remove sulfur from these fuels is called hydrodesulfurization. However, heterocyclic sulfur compounds such as substituted dibenzothiophenes are very difficult to desulfurize by hydrodesulfurization.
Biological desulfurization of fossil fuels may offer an alternative process to remove the recalcitrant sulfur. One of the limiting factors to apply this process in an industrial scale is the cost associated to the production of biocatalysts, mainly due to the costs associated to the culture media formulation. At present, there is not any economically suitable method for large-scale preparation of biocatalysts (Ma et al., 2006). The utilization of alternative carbon sources derived from agro-industrial by-products or wastes may thereby represent an opportunity to cheaper culture media. These alternative substrates have widely been used as feedstock for several fermentation processes such as for the production of lactic acid (Bustos et al., 2005), polyhydroxybutyrate (Hu et al., 1999), ethanol (Karimi et al., 2006), pullulan (Israilides et al., 1998), xanthan gum (Yoo & Harcum, 1999), bacterial cellulose (Bae & Shoda, 2005), and xylanase (Nascimento et al., 2003). However, studies involving the DBT bacterial desulfurization have been carried out only using reagent-grade sugar- containing media.
The utilization of alternative carbon sources by desulfurizing bacteria might raise a problem associated to the presence of readily bioavailable sulfur compounds. Desulfurization of DBT is completely inhibited in the presence of several sulfur compounds in the culture medium.
127 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
The dsz promoter has been characterized from Rhodococcus erythropolis strain IGTS8 and it was found that the dsz genes expression is strongly repressed by sulfate or other sulfur compounds even in the presence of DBT (Li et al., 1996). Therefore, it is important to search for low-cost feedstocks containing low residual concentration or even a null content of sulfur.
Pulp and paper industry generates large amounts of waste throughout the year (Thomas, 2000). Concentrated sludge generated by the wastewater treatment facilities of recycled paper plants is currently a major disposal problem concerning the paper industry and it has to be urgently solved (Oral et al., 2005). Recycled paper sludge (RPS) (after neutralization) is approximately made up of 35% cellulose, 10% xylan and 20% lignin (on a dry-weight basis), being the remaining mainly inorganic ash. Due to this high polysaccharide content RPS appears as a promising feedstock for formulation of inexpensive culture media (Van Wyk & Mohulatsi, 2003) providing their polymeric carbohydrates are broken down into fermentable monosaccharides. This hydrolysis can be carried out by chemical or enzymatic methods. The latter is advantageous since it is more specific, it allows milder operation conditions leading to reduced production of biologically inhibitory compounds (such as sugar and lignin degradation products) and the biocatalysts are potentially reusable (Wen et al., 2004).
In this context, the aim of the present work was to evaluate the performance of the hydrolyzates obtained by enzymatic saccharification of RPS, as nutrient source for low-cost dibenzothiophene desulfurization by Gordonia alkanivorans strain 1B. To our knowledge, this is the first report on the utilization of alternative raw materials as substitutes for refined substrates (namely glucose) in biodesulfurization studies.
128 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
MATERIALS AND METHODS
Substrate
Recycled paper sludge
The present study used pressed RPS consisting of the solids resulting from wastewater treatment facility of the paper-recycling mill of Renova (Torres Novas, PT).
Enzymatic hydrolysis
After neutralization with hydrochloric acid (to reduce the carbonate content) RPS was suspended in 0.05 M sodium citrate buffer, pH 5.5, for an initial consistency of 7.5% (w/v), expressed in terms of total carbohydrate mass, and it was steam sterilized by autoclaving (at 121 ºC, 1 atm, for 15 min). This sludge suspension was incubated with the filter-sterilized enzyme solution containing a mixture of two commercial enzyme preparations (cellulolytic and xylanolytic, from Novozymes, Denmark): Celluclast® 1.5L, applied on a dosage of -1 ® -1 filter paper activity (FPase) of 10 U g carbohydrate; and Novozym 188, 0.4 ml g carbohydrate on sludge. The hydrolysis was carried out at 35 ºC in an orbital shaker (150 rpm) for 120 hours, under aseptic conditions. A control enzyme mixture was subjected to the same hydrolysis conditions but in the absence of sludge.
In an alternative approach in order to remove any sulfur compounds present in the commercial enzymatic mixture, this mixture was overnight dialyzed (cut-off = 12-14 kDa; Spectra/Por membranes, Spectrum Laboratories, Inc., Rancho Dominguez, CA, USA) against milli-Q water at 4 ºC.
The hydrolyzates obtained were filter sterilized and analyzed for sugar composition or used for supplementation of culture media.
129 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
Bacterial strain and growth conditions
General conditions
G. alkanivorans strain 1B, originally isolated from oil-contaminated soil (Alves et al., 2005), was used in all cultivation assays. Unless otherwise stated, strain 1B was cultured in sulfur-free mineral (SFM) medium supplemented with a trace elements solution as described by Alves et al. (2007). DBT, 4-methyl DBT (4-mDBT), and/or 4,6-dimethyl DBT (4,6-dmDBT), dissolved in dimethylformamide, were added as source of sulfur. All G. alkanivorans liquid cultivations were carried out in duplicate shake-flasks, at 30 ºC and initial pH 7.5, with 150 rpm (orbital shaking). Harvested samples were analyzed for cell growth, organic compounds involved in desulfurization and sugar contents.
DBT desulfurization on RPS hydrolyzate
The RPS hydrolyzates obtained either with dialyzed enzymes (dialyzed hydrolyzate) or non-dialyzed enzymes (non-dialyzed hydrolyzate), were used as carbon source on a concentration of 10 g l-1 glucose in medium containing 0.25 mM DBT. Growth controls were carried out using reagent-grade glucose, xylose or cellobiose on a concentration of 10 g l-1.
Effect of supplementation of RPS hydrolyzate
In order to investigate the possibility of using the RPS hydrolyzate obtained with dialyzed enzymes as the component of culture medium (on the concentration previously described), cultivations were carried out on the following formulations: 1) RPS hydrolyzate; 2) RPS hydrolyzate + phosphates (KH2PO4 and Na2HPO4·2H2O); 3) RPS hydrolyzate + ammonia
(NH4Cl); 4) RPS hydrolyzate + phosphates + ammonia; 5) RPS hydrolyzate + phosphates + magnesium (MgCl2·6H2O); 6) RPS hydrolyzate + phosphates + ammonia + magnesium; 7)
RPS hydrolyzate + phosphates + ammonia + magnesium + Zinc (ZnCl2); 8) RPS
130 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION hydrolyzate + phosphates + ammonia + magnesium + trace elements solution. Phosphates, ammonia and magnesium were added on the concentration present on SFM medium and Zn2+ was used on a final concentration of 10 mg l-1. DBT was used as sulfur source, on a concentration of 0.25 mM.
Desulfurization of model oil
The model oil used consisted of n-heptan with DBT, 4-mDBT and 4,6-dmDBT dissolved on a concentration of 2 mM each. In this assay it was used a water/oil ratio of 10:1. The culture medium used was the formulation 7 described on the previous section.
Analytical methods
Cell growth was directly monitored by measuring the absorbance of the culture broth samples at 600 nm. Dry cell weight (DCW) was determined using 0.22 µm cellulose acetate membranes, after drying for 18 h at 100 ºC to constant weight. Filter paper activity (FPase), describing the cellulolytic activity, was assayed using Whatman number 1 filter paper as substrate. Enzyme activity was expressed in international units (U) as the amount of enzyme required to release 1 µmol min-1 of glucose reducing equivalent under the assay conditions. Reducing sugars were estimated by the dinitrosalycilic acid method (Miller, 1959).
Sugars were measured by high-performance liquid chromatography using a Waters LC1 module 1 plus (Millford, LA) instrumentation equipped with a differential refractive index detector. An Aminex HPX-87H column (Bio-Rad Laboratories, Hercules, CA, USA) was -1 used, operated at 50 ºC with 0.005 M H2SO4 as mobile phase at a flow rate of 0.4 ml min . The determination of the organic compounds involved in desulfurization was performed as described by Alves et al. (2007). In all GC measurements, anthracene was used as internal standard to minimize the variation.
131 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
RESULTS AND DISCUSSION
DBT desulfurization on RPS hydrolyzate
Table 1 shows the sugar composition of both RPS hydrolyzates obtained. The sludge hydrolysis yield was calculated based on the total sugar concentration in the hydrolyzate (corrected for concentration in the control assay) relatively to the content of polysaccharides (potential glucose and xylose) of the feedstock. Since hydrolysis was carried out under optimal conditions defined previously (according to a previous study), a very high yield (93%) was obtained. A lower hydrolysis yield (72%) was obtained for dialyzed RPS hydrolyzate.
Table 1. Major carbohydrate composition of both hydrolyzates obtained from recycled paper sludge (RPS). Control assays (with dialyzed or non-dialyzed enzyme mixtures) were performed in the absence of RPS.
Concentration (g l-1) Hydrolysis Total Yield (%) Glucose Xylose Cellobiose sugars RPS hydrolyzate 56.2 12.9 6.8 75.9 93
Control 5.1 0.9 0.4 6.4 -
Dialyzed RPS 36.3 12.7 4.9 53.9 72 hydrolyzate
Control 0.0 0.0 0.0 0.0 -
This difference reflects the lower extension of hydrolysis of the cellulosic fraction resulting from a reduction on the efficiency of the cellulase formulation due to the dialysis process. The fact that both enzyme formulations (non-dialyzed and dialyzed) were applied at the same dosage in terms of cellulase activity implies that the initial hydrolysis rate should be the same. However, it is not possible to assure that enzymatic activities have equivalent
132 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION stabilities. Hence, this reduction on efficiency observed for dialyzed enzymes might be due to a reduction on enzymes stability, resulting from a loss of some small-molecule stabilizer components during dialysis.
Despite the difference on sugar concentrations, both RPS hydrolyzates used in this work were composed of glucose, xylose and cellobiose (Table 1). Since strain 1B has no ability to use xylose and cellobiose (data not shown), the assays were performed using diluted RPS on a initial concentration of 10 g l-1 of glucose. Figure 1 shows the time course of G. alkanivorans strain 1B cultivation using RPS hydrolyzates as carbon source. The bacterium attained maximum growth at 100 – 120 h of cultivation, and the turbidity at 600 nm at this time was around 10 for both cases (Fig. 1A and 1B). The maximum specific growth rates,
µmax, for growth with non-dialyzed hydrolyzate (Fig. 1A) and dialyzed hydrolyzate (Fig. 1B) were 0.051 h-1 and 0.035 h-1, respectively. It is concluded that strain 1B grows better using hydrolyzates in comparison with the corresponding culture medium containing commercial -1 grade glucose as the only carbon source (µmax = 0.019 h ) (Alves et al., 2005). The RPS hydrolyzates not only contain the products of the RPS conversion but also all the components provided by the enzymatic formulation. Commercial enzyme preparations are obtained by growth of microbial organisms, followed by concentration of the culture cell- free broth, for obtaining an extract with a high enzymatic activity.
Hence, not only protein concentration is achieved, but most of all residual components present on culture media are also concentrated and these nutrients could be used by strain 1B, acting as growth promoters. This explains the faster growth achieved for the culture media containing RPS hydrolyzates. The slower growth of strain 1B with the dialyzed hydrolyzate is also justified by the removal of some of these nutrients present on the enzymatic formulation due to dialysis, attenuating this nutritional effect. Although this effect was already predicted, dialysis was performed aiming to remove possible sulfur compounds present on the enzymatic formulation that could be more easily assimilated than DBT by bacterial cells. The absence of DBT desulfurization in the non-dialyzed hydrolyzate (Fig. 1A) might be explained by the presence of other S-sources of faster assimilation than
133 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
DBT. It has been reported (Takada et al., 2005) that the presence of other sulfur sources in the culture medium, even in small concentrations, inhibits the desulfurization of DBT. The dialysis of the enzyme mixture is therefore essential for the occurrence of DBT desulfurization, unless a recombinant strain (lacking dsz substrate repression) is used.
125 12 A ) 100 10 -1 8 M) 75 µ 6
50 ; Glucose (g l
2-HBP ( 2-HBP 4
600 nm 600 25
2 OD
0 0 0 25 50 75 100 125 150
125 12 B
10 ) 100 -1
8
M) 75 µ 6
50 ; Glucose (g l ( 2-HBP 4 600 nm
25 OD 2
0 0 0 25 50 75 100 125 150
Time (h)
Fig. 1. Time course of G. alkanivorans strain 1B cultivation using: Panel A, non- dialyzed hydrolyzate; Panel B, dialyzed hydrolyzate. ●, strain 1B growth; ■, 2-HBP concentration; S, glucose concentration.
134 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
As shown in figure 1B, the detection of 2-HBP only occurs after 50 h of cultivation with the dialyzed hydrolyzate, probably due to the presence in the culture medium of a residual concentration of sulfur contaminants that were not completely removed by dialysis. In this cultivation, 250 µM of DBT was consumed after 96 hours of culture (data not shown), but less than 125 µM of 2-HBP was determined for this time, which means that only 50% of the 2-HBP produced was detected. This result is in agreement with a previous study involving the same strain (Alves et al., 2005). The maximum specific productivity of 2-HBP was 1.1 µmol g-1(DCW) h-1.
Effect of supplementation of RPS hydrolyzate
In the previous assays, it was shown that dialyzed RPS hydrolyzate could be used as carbon source in a process of DBT desulfurization by G. alkanivorans strain 1B. Subsequently, and having in mind the already stated rich composition of the enzymatic formulation used, the possibility of using this hydrolyzate as a complete culture medium was investigated. In these assays, dialyzed RPS hydrolyzate was used on an initial glucose concentration of 10 g l-1, being supplemented using different nutrient formulations as stated previously in materials and methods section.
Figure 2 shows the time course of growth and production of 2-HBP by strain 1B obtained for each medium formulation. RPS hydrolyzate without additional nutrients (formulation 1), with phosphates (formulation 2) or with phosphates + magnesium (formulation 5) did not allow a significant growth neither 2-HBP production. Although formulation 3 (RPS hydrolyzate with ammonia) has allowed a high initial bacterial growth, this ended after the first 72 hours of culture, probably due to the absence of some nutrients (Fig. 2A). The same behavior occurred with 2-HBP production after 96 hours of cultivation. A complete consumption of the glucose and DBT present in culture medium (data not shown) was observed when formulations 4, 6, 7 and 8 were used. This indicates that these formulations can be used as complete culture medium for DBT desulfurization by G. alkanivorans strain 1B.
135 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
3.0
2.5 A
2.0
600nm 1.5
Ln OD Ln 1.0
0.5
0.0 0 20 40 60 80 100 120 140
200
160 B M)
µ 120
80 2- HBP ( ( HBP 2-
40
0 0 20 40 60 80 100 120 140 Time (h)
Fig. 2. Time course of growth (Panel A) and 2-HBP production (Panel B) for all the formulations tested in the RPS hydrolyzate supplementation assays: , formulation 1; S, formulation 2; ○, formulation 3; ■, formulation 4; □, formulation 5; +, formulation 6; ●, formulation 7; ×, formulation 8.
-1 Growth occurred throughout 120 hours and a µmax value of 0.031 h was achieved for formulation 7 and 0.027 h-1 for formulations 4, 6 and 8. The desulfurization patterns, in terms of 2-HBP productivity, were also very similar for formulations 4, 6, 7 and 8. However, formulation 7 seems to be more favorable to a faster desulfurization since the 2- HBP maximum productivity obtained was 14% higher (5.7 µM h-1) comparatively to the
136 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION value of about 5 µM h-1 obtained for formulations 4, 6 and 8 (Fig. 2B). Formulation 7 consists of the composition of SFM medium supplemented with zinc. In a previous work it was found that this metal ion has an important role for the DBT desulfurization ability of G. alkanivorans strain 1B (data not shown).
Desulfurization of model oil
Desulfurization of DBT, 4-mDBT and 4,6-dmDBT in a model oil system (containing 2mM of each sulfur source dissolved in n-heptan) was carried out with G. alkanivorans strain 1B using as culture medium dialyzed RPS hydrolyzate supplemented with phosphates, ammonia, magnesium and zinc, corresponding to formulation 7. The volume ratio of oil-to- water was 0.1. Strain 1B cannot use n-heptan as carbon source (data not shown). Figure 3 shows the bacterial growth and the sulfur and glucose consumption during 192 hours of cultivation.
10 3.0 ) -1 2.5 8
2.0 ) 6 0 1.5 4 1.0 (OD/OD Ln
2 0.5 Total sulfur (mM); Glucose (g l
0 0.0 0 24 48 72 96 120 144 168 Time (h)
Fig. 3. Time course of G. alkanivorans strain 1B cultivation with model oil: ●, strain 1B growth; ■, total sulfur concentration; S, glucose concentration.
137 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
The results show that strain 1B consumed almost all the glucose present in the culture -1 medium after 96 hours of culture, with a µmax of 0.062 h and total sulfur was decreased 2.7- fold, to 2.23 mM after 168 hours of cultivation. The specific desulfurization rates after 24, 48 and 72 hours were 22.2, 11.1 and 4.8 µmol g-1(DCW) h-1, respectively.
Two alkylated dibenzothiophenes, which are recalcitrant to hydrodesulfurization technology, were used in this work. In fact, dibenzothiophenes substituted in positions 4 and 6 are the most recalcitrant to hydrodesulfurization (Okada et al., 2002). In Figure 4, the time course of each individual sulfur source degradation is shown.
2.0
1.6
1.2
0.8 Sulfur source [mM] source Sulfur 0.4
0.0 0 24 48 72 96 120 144 168
Time (h)
Fig. 4. Time course of sulfur sources degradation during G. alkanivorans strain 1B cultivation with model oil: ■, 4-methyl DBT concentration; S, 4,6-dimethyl DBT concentration; , DBT concentration.
Strain 1B simultaneously utilized all sulfur sources present in model oil desulfurizing, although DBT is preferentially utilized relatively to the alkylated dibenzothiophenes. On the contrary, Prince & Grossman (2003) stated that these alkylated compounds were removed in preference relatively to their unalkylated parent.
138 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
The specific desulfurization rates in the model oil for DBT, 4-mDBT and 4,6-dmDBT were 0.78, 0.68 and 0.41 µmol g-1(DCW) h-1, respectively. 4-mDBT has also been removed more extensively than 4,6-dmDBT, in the work performed by Prince & Grossman (2003). Okada et al. (2002) have also verified that desulfurization activities against alkyl DBTs decreased with increasing molecular weight of DBT derivatives.
Figure 5 shows the chromatograms obtained by GC analysis for an initial (Fig. 5A) and a final (Fig. 5B) sample of model oil desulfurization with G. alkanivorans strain 1B. The peak with a retention time of 19.1 min corresponds to the internal standard used (anthracene). The peaks at 18.1, 20.7 and 23.2 min correspond to DBT, 4-mDBT and 4,6-dmDBT, respectively. The peak with retention time of 11.8 min corresponds to 2-HBP resulting from DBT desulfurization. The desulfurization of 4,6-dmDBT produces only one compound, detected at retention time of 16.7 min. The desulfurization of 4,6-dmDBT by Nocardia globerula R-9 also produces only one product that was identified by HPLC/GC–MS as monohydroxy dimethyl biphenyl (Luo et al., 2003). A different result was described for 4,6-dmDBT desulfurization by Sphingomonas paucimobilis strain TZS-7, which produces 3 compounds, suggesting that this compound is degraded through a ring-destructive pathway (Lu et al., 1999). In G. alkanivorans strain 1B, the degradation of 4,6-dmDBT was assumed to occur through the 4S pathway, which involves the carbon-sulfur bond cleavage reaction and generates hydroxybiphenyl compounds derived from the substrate.
The desulfurization of 4-mDBT by strain 1B occurs by a different mechanism than the one for 4,6-dmDBT, since 2 compounds were produced, in a proportion of 1:5, detected at retention times of 13.9 and 14.5 min. This could suggest the cleavage of some carbon-sulfur bonds of 4-mDBT by strain 1B.
139 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
A
B
Fig. 5. GC chromatograms obtained for the initial (Panel A) and final (Panel B) samples taken from G. alkanivorans strain 1B cultivation with model oil.
CONCLUSIONS
Altogether, these results clearly show that RPS hydrolyzates can be employed as nutrient for DBT desulfurization by G. alkanivorans strain 1B (even with a low enrichment of the
140 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION medium), providing process efficiencies similar to those achieved when using conventional medium containing commercial grade sugars. This procedure has a double profit because a harmful environmental waste is removed and RPS can be used as an alternative carbon source for a biotechnological process, as already reported for other industrial applications such as lactic acid (Lin et al., 2005) and ethanol production (Lark et al., 1997).
Realistically, for a potential industrial application, an inexpensive culture medium would have to be employed in a large-scale process. Therefore, the cost of commercial enzyme formulations might economically hamper the present process, unless the enzymes added in the hydrolysis step are recovered and reused. This might be achieved by developing a process based on an enzymatic membrane reactor, in which a semi-permeable ultrafiltration membrane is used to retain the enzymes in the reactor while preserving their activity.
Acknowledgments
This work has been partially supported by the contract POCTI/AMB/59108/04.
141 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
REFERENCES
Alves, L., Melo, M., Mendonça, D., Simões, F., Matos, J., Tenreiro, R., Gírio, F.M., 2007. Sequencing, cloning and expression of the dsz genes required for dibenzothiophene sulfone desulfurization from Gordonia alkanivorans strain 1B. Enz. Microb. Technol. 40, 1598-1603.
Alves, L., Salgueiro, R., Rodrigues, C., Mesquita, E., Matos, J., Gírio, F.M., 2005. Desulfurization of dibenzothiophene, benzothiophene and other thiophene analogues by a newly isolated bacterium, Gordonia alkanivorans strain 1B. Appl. Biochem. Biotechnol. 120, 199-208.
Bae, S.O., Shoda, M., 2005. Production of bacterial cellulose by Acetobacter xylinum BPR2001 using molasses medium in a jar fermentor. Appl. Microbiol. Biotechnol. 67, 45-51.
Bustos, G., Moldes, A.B., Cruz, J.M., Domínguez, J.M., 2005. Production of lactic acid from vine- trimming wastes and viticulture lees using a simultaneous saccharification fermentation method. J. Sci. Food Agric. 85, 466-472.
Gupta, N., Roychoudhury, P.K., Deb, J.K., 2005. Biotechnology of desulfurization of diesel: prospects and challenges. Appl. Microbiol. Biotechnol. 66, 356-366.
Hu, P.H., Chua, H., Huang, A.L., Ho, K.P., 1999. Conversion of industrial food wastes by Alcaligenes latus into polyhydroxyalkanoates. Appl. Biochem. Biotechnol. 77, 445-454.
Israilides, C.J., Smith, A., Harthill, J.E., Barnett, C., Bambalov, G., Scanlon, B., 1998. Pullulan content of the ethanol precipitate from fermented agro-industrial wastes. Appl. Microbiol. Biotechnol. 49, 613-617.
Karimi, K., Emtiazi, G., Taherzadeh, M.J., 2006. Ethanol production from dilute-acid pretreated rice straw by simultaneous saccharification and fermentation with Mucor indicus, Rhizopus oryzae, and Saccharomyces cerevisiae. Enz. Microb. Technol. 40, 138-144.
Lark, N., Xia, Y.K., Qin, C.G., Gong, C.S., Tsao, G.T., 1997. Production of ethanol from recycled paper sludge using cellulase and yeast, Kluveromyces marxianus. Biomass Bioenergy. 12, 135- 143
Li, M.Z., Squires, C.H., Monticello, D.J., Childs, J.D., 1996. Genetic analysis of the dsz promoter and associated regulatory regions of Rhodococcus erythropolis IGTS8. J. Bacteriol. 178, 6409-6418.
142 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
Lin, J.Q., Lee, S.M., Koo, Y.M., 2005. Modeling and simulation of simultaneous saccharification and fermentation of paper mill sludge to lactic acid. J. Microbiol. Biotechnol. 15, 40-47.
Lu, J., Nakajima-Kambe, T., Shigeno, T., Ohbo, A., Nomura, N., Nakahara, T., 1999. Biodegradation of dibenzothiophene and 4,6-dimethyldibenzothiophene by Sphingomonas paucimobilis strain TZS-7. J. Biosc. Bioeng. 88, 293-299.
Luo, M.F., Gou, Z.X., Xing, J.M., Liu, H.H., Chen, J.Y., 2003. Microbial desulfurization of model and straight-run diesel oils. J. Chem. Technol. Biotechnol. 78, 873–876
Ma, C.-Q., Feng J.-H., Zeng, Y.-Y., Cai, X.-F., Sun, B.-P., Zhang, Z.-B., Blankespoor, H.D., Xu, P., 2006. Methods for the preparation of a biodesulfurization biocatalyst using Rhodococcus sp. Chemosphere. 65, 165–169.
Miller, G.L., 1959. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 31, 426-428.
Nascimento, R.P., Coelho, R.R.R., Marques, S., Alves, L., Girio, F.M., bom, E.P.S., Amaral-Collaco, M.T., 2003. Production and partial characterisation of xylanase from Streptomyces sp. strain AMT-3 isolated from Brazilian cerrado soil. Enzyme Microb. Technol. 31, 549-555.
Okada, H., Nomura, N., Nakahara, T., Maruhashi, K., 2002. Analyses of substrate specificity of the desulfurizing bacterium Mycobacterium sp. G3. J. Biosc. Bioeng. 93, 228-233.
Oral, J., Sikula, J., Puchyr, R., Hajny, Z., Stehlik, P., Bebar, L., 2005. Processing of waste from pulp and paper plant. J. Clean. Prod. 13, 509-515.
Prince, R.C., Grossman, M.J., 2003. Substrate preferences in biodesulfurization of diesel range fuels by Rhodococcus sp. strain ECRD-1. Appl. Environ. Microbiol. 69, 5833-5838.
Takada, M., Nomura, N., Okada, H., Nakajima-Kambe, T., Nakahara, T., Uchiyama, H., 2005. De- repression and comparison of oil-water separation activity of the dibenzothiophene desulfurizing bacterium, Mycobacterium sp. G3. Biotechnol. Lett. 27, 871-874.
Thomas, S., 2000. Production of lactic acid from pulp mill solid waste and xylose using Lactobacillus delbrueckii (NRRL B445). Appl.Biochem.Biotechnol. 84-86, 455-468.
Van Wyk, J.P.H., Mohulatsi, M. 2003. Biodegradation of wastepaper by cellulase from Trichoderma viride. Bioresour. Technol. 86, 21-23.
143 RPS HYDROLYZATE AS NUTRIENTS SOURCE FOR DBT DESULFURIZATION
Wen, Z.Y., Liao, W., Chen, S.L., 2004. Hydrolysis of animal manure lignocellulosics for reducing sugar production. Bioresour. Technol. 91, 31-39.
Yoo, S.D., Harcum, S.W., 1999. Xanthan gum production from waste sugar beet pulp. Bioressour. Technol. 70, 105-109.
144
Chapter 6
GLOBAL ANALYSIS AND CONCLUSIONS
GLOBAL ANALYSIS AND CONCLUSIONS
Sulfur is usually one of the most abundant elements in crude oil, which is organically bound, mainly in the form of condensed thiophenes. The problem with fossil fuels is that the combustion products, as nitrogen oxides and sulfur oxides, are hard on the planet. Due to the increasing emissions of such compounds into the atmosphere, governments throughout the world have recognized the problem and act to reduce it through legislation. The easiest way to limit the amount of sulfur dioxide emitted into the air is to restrict the amount of sulfur in fuel. During the last two decades, legislation in several countries has limited the sulfur concentration in transportation fuels down to very low levels.
Fuel desulfurization seems like a reasonable approach to reduce air pollution, except for one inevitable problem: it costs money, and as the extent of desulfurization increases, the costs escalate rapidly. Therefore, as an alternative promising method, microbial desulfurization has attracted attention for its application to the desulfurization of fossil fuels. It has been demonstrated that microorganisms metabolize DBT aerobically or anaerobically (Ohshiro & Izumi, 1999).
Although occasional reports on anaerobic bacteria that can desulfurize DBT have occurred, the rate and extent of petroleum desulfurization by currently available anaerobic cultures and the lack of knowledge on the biochemistry and genetics of such microorganisms make the development of a commercial process unlikely (Bahrami et al., 2001). Much more is known about the pathway for the metabolism of DBT by aerobic microorganisms, especially due to the knowledge achieved in the studies involving bacteria belonging to the genus Rhodococcus. Indeed, the first advances in biodesulfurization area were achieved from the isolation of Rhodococcus erythropolis strain IGTS8, and during many years the main results were obtained with this strain. Very little was known about desulfurization ability of bacterial strains belonging to other genera.
149 GLOBAL ANALYSIS AND CONCLUSIONS
In this context, the focus of this dissertation was the isolation of a novel bacterial strain belonging to a genus other than the well-known Rhodococcus, which can desulfurize DBT, and the study of its physiology and genetics. Moreover, the possibility of using alternative carbon sources to grow this isolated bacterium, aiming to reduce the costs of the biocatalysts production, was also investigated.
The first step to obtain novel bacteria able to utilize DBT as the only sulfur source was to select the sites for the sampling collection. It was selected an area with high contamination of hydrocarbons such as the grounds at EXPO-98 (Lisbon, Portugal), where the former Petrogal refinery was located. The isolation procedure allowed the attainment of some bacterial strains with ability to grow in a culture medium with DBT. From these strains, the bacterium showing the best growth in a culture medium in which DBT was the only sulfur source was selected.
Then, this strain was identified by characterization of its morphological and physiological characteristics having in mind its identification. These data, in addition to the chemical composition of the bacterial cell wall and the knowledge of the first 500 bases amongst the 16S rDNA, allowed to identify the strain 1B as Gordonia alkanivorans.
The actinomycete genus Gordonia has attracted much interest in recent years for various reasons. Most species were isolated due to their abilities to degrade xenobiotics, environmental pollutants, or otherwise slowly biodegreadable natural polymers, as well as to transform or synthesize potential useful compounds. The variety of chemical compounds transformed, biodegraded, and synthesized by gordoniae makes these bacteria potentially useful for environmental and industrial biotechnology (Arenskötter et al., 2004).
The species G. alkanivorans was described for the first time by Kummer et al. (1999). However, the first report involving biodesulfurization studies on this species was the work described in chapter 2 of this dissertation (Alves et al., 2005). G. alkanivorans strain 1B has shown a good ability to desulfurize several sulfur sources, including DBT, DBT sulfone and benzothiophene. In fact, its ability to use DBT and DBT sulfone is an important
150 GLOBAL ANALYSIS AND CONCLUSIONS characteristic, since the majority of the described bacteria can only metabolize one of these compounds. On the other hand, strain 1B can also use DBT derivatives, 4m-DBT and 4,6dm-DBT as sulfur source, as described in chapter 5.
The repression effect on desulfurization by sulfates, the fact that 2-HBP is the only metabolite detected, in addition to the fact of DBT, DBT sulfone or 2-HBP can not be used as sole carbon sources, suggested that strain 1B proceed DBT desulfurization using the 4S metabolic pathway. It is generally accepted that 4S pathway is the only of the known desulfurization pathways that can be applied in a future industrial process since there is no decrease in the carbon content of the desulfurized fuel.
The metabolic pathway used by strain 1B to desulfurize DBT was confirmed through the identification, sequencing and characterization of desulfurization genes, as described in chapter 3. To perform this work, the primers were selected based on known sequence of dsz genes of R. erythropolis strain IGTS8. These primers allowed the amplification of some sequences of dsz genes of strain 1B. These sequences were used to design other PCR primers that are G. alkanivorans specific and a sequence of 4581 bp was obtained that included the full sequence of dszABC genes.
Evidence for the role played by the desulfurization process in nature is provided by the finding that in most soils sulfur is predominantly present as sulfonates and sulfate esters, rather than as sulfate, and the use of organosulfur compounds may be crucial for the survival of many bacterial cultures (Mirleau et al., 2005). The desulfurization process is widely spread among the microbial organisms and diverse species have often been found to possess identical, or nearly identical, dsz gene sequences with generally more than 99% identity to the dsz genes of R. erythropolis IGTS8 (Kilbane et al., 2006).
The dsz genes of G. alkanivorans strain 1B show about 89% identity with the 4S metabolic pathway dsz genes of R. erythropolis IGTS8, but the overall gene order and structure is preserved including an overlap of the dszA and dszB genes, which is 4 bp in R. erythropolis IGTS8 but 64 bp in G. alkanivorans strain 1B (Alves et al., 2007). These data confirm the
151 GLOBAL ANALYSIS AND CONCLUSIONS results that suggest that strain 1B uses 4S metabolic pathway to desulfurize DBT (see chapter 2). Indeed, a recombinant bacterium including dszAB genes was able to produce 2- HBP, the final product of 4S pathway, but only when DBT sulfone was used.
G. alkanivorans strain 1B shows a good potential to be used in a biodesulfurization process of fossil fuels. However, its low desulfurization activity is a problem for a possible application in a future industrial process. Thus, the next step in this work consisted of the optimization of some factors that influence the growth and desulfurization ability of strain 1B.
The composition of the culture medium is an important parameter for growth of desulfurizing bacteria, since some metal ions can be necessary as enzymatic cofactors. Therefore, the suppression of some metal ions present in the mineral medium usually used to grow this bacterial strain was tested. It was found that the absence of zinc significantly decreased the growth and DBT desulfurization ability of strain 1B. Without zinc, strain 1B showed 20% less desulfurization activity than in the conditions described in chapter 2 (0.5 mg l-1 Zn). On the other hand, when Zn concentration was increased to 10 mg l-1, the 2-HBP productivity was also increased by 126% in comparison with the assay without zinc (chapter 4).
In addition, this metal ion seems to influence bacterial growth only when DBT is present in the culture medium. Using sulfates as sulfur source, strain 1B shows the same growth behavior in the presence or absence of zinc. The assays with resting cells pre-grown in presence and absence of zinc confirm the previous indication. In fact, resting cells pre- grown with zinc showed a 2-HBP specific productivity of 2.29 µmol g-1 (DCW) h-1, 7.6-fold higher than the specific productivity achieved by resting cells pre-grown without zinc. These results suggest that zinc ion might play a key role in the activity of one or more desulfurization enzymes of G. alkanivorans strain 1B. Thus, zinc ion concentration must be increased from 0.5 mg l-1 to 10 mg l-1 in the culture medium to grow strain 1B.
152 GLOBAL ANALYSIS AND CONCLUSIONS
The establishment of biodesulfurization in a commercial scale is not a reality yet, due to several factors that limit its application. Some key factors affect the economic viability of such processes, as biocatalyst activity and cost, differential in product selling price, sale or disposal of co-products or wastes from the treatment process, and the capital and operating costs of unit operations in the treatment scheme (Kaufman et al., 1997). In fact, the cost associated with the composition of the culture medium utilized to obtain the biocatalysts is one of the main factors limiting BDS application to an industrial process. Glucose is the currently carbon source used in the majority of the biodesulfurization studies. Therefore it is of great interest the use of some alternative carbon source that allows the bacterial growth and DBT desulfurization. These carbon sources can advantageously be obtained from agro- industrial materials, such as recycled paper sludge. The main problem associated to these alternative carbon sources might be the presence of sulfur sources that are more easily metabolized by bacteria.
The enzymatic hydrolysis of polysaccharides on RPS material provided a hydrolyzate containing mainly glucose and xylose, in addition to the nutrients from the enzyme formulation used. As presented in chapter 5, the hydrolyzate obtained without dialysis of enzymatic mixture allowed a strong bacterial growth but without DBT desulfurization. This fact was due to the presence of other sulfur sources in the enzyme mixture. When hydrolysis was performed with dialyzed enzyme, the hydrolyzate allowed a good desulfurization by strain 1B.
During the growth of G. alkanivorans strain 1B using dialyzed hydrolyzate, the stationary phase occurred after 4 days of cultivation (chapter 5), which represents a significant improvement as compared to the growth obtained in a SFM medium with commercial grade glucose (chapter 2). Similarly, after 4 days of cultivation with RPS hydrolyzate and with commercial glucose, 106 µM 2-HBP and 38 µM 2-HBP were detected, respectively. However, the maximum specific productivity of 2-HBP was very similar in both cases. These results demonstrate that RPS hydrolyzate can be employed as nutrient for DBT desulfurization by G. alkanivorans strain 1B.
153 GLOBAL ANALYSIS AND CONCLUSIONS
The possibility of using RPS hydrolyzate to grow the strain 1B with the minimum supplementation was investigated. It was demonstrated that the supplementation of the hydrolyzate only with phosphate and ammonia was enough to support strain 1B growth. However, the RPS hydrolyzate supplementation described as formulation 7 (see chapter 5) was more favorable to a faster desulfurization since the 2-HBP maximum productivity was increased by 14% comparatively to the productivity obtained for formulations 4, 6 and 8.
One reason for this behavior was the presence of a higher concentration of zinc ion that allows a better bacterial growth and DBT desulfurization by strain 1B (chapter 4). From a practical point of view, the increase of desulfurization activity by G. alkanivorans strain 1B, in addition to the use of a less expensive culture medium based on RPS hydrolyzate, is an important improvement for a future application. To our knowledge, this is the first report describing the use of an alternative carbon source in BDS.
In this thesis, the desulfurization of any fossil fuel was not studied. However, the ability of G. alkanivorans strain 1B to desulfurize a model oil containing some sulfur compounds recalcitrant to hydrodesulfurization process was addressed. In chapter 5, it was demonstrated that strain 1B could desulfurize this model oil, containing DBT, 4m-DBT and 4,6dm-DBT dissolved in n-heptan, with a volume ratio of oil-to-water of 0.1. In these -1 conditions, strain 1B showed a faster growth, µmax of 0.062 h , and a higher desulfurization activity, 3.8 mM consumption of total sulfur after 7 days of cultivation. This enhanced behavior can be explained by the fact that sulfur compounds are more accessible to the cells and, by the opposite, the desulfurization products, as 2-HBP, also are dissolved in the organic fraction, being in this case less toxic to the cells.
Other reports studying two phase solvent:water systems have described the increase of DBT desulfurization rates in the presence of 40-50% n-tetradecane (Ohshiro et al., 1995) or kerosene (Ohshiro et al., 1996) or 96% hexadecane (Kaufman et al., 1998). The increased rates in higher hydrocarbon fractions might suggest transfer of DBT through the interface
154 GLOBAL ANALYSIS AND CONCLUSIONS between the aqueous and hydrocarbon phases or adsorption of cells at the interface (Maghsoudi et al., 2001).
The ultimate purpose of this work was to isolatate a new bacterium that presents good desulfurization ability using the 4S metabolic pathway. It was shown that G. alkanivorans strain 1B presents desulfurization activities very similar to some reported R. erythropolis strains. Desulfurization rates for non-engineered Rhodococcus spp. are 1 to 5 mg of sulfur g-1(DCW) h-1 (Kaufman et al., 1998), and the desulfurization rate of strain 1B in model oil was 0.7 mg of sulfur g-1 (DCW) h-1. This is a higher rate than the desulfurization rate reported for the wild type Gordonia sp. strain CYKS1 growing in a commercial grade glucose, which was around 0.12 mg of sulfur g-1 (DCW) h-1 (Chang et al., 2000).
However, some recombinant strains carrying multiple copies of the dsz genes of R. erythropolis KA 2-5-1 increased the desulfurization activity up to 46 mg of sulfur g-1(DCW) h-1 (Hirasawa et al., 2001). Recent research has resulted in a 200-fold increase in expression of key desulfurization genes in the best strains (Van Hamme et al., 2003).
Because biodesulfurization rates are still relatively low for the requirements of an industrial process, little research has focused on process development. Due to these application restrictions, it is widely accepted that biodesulfurization might be a complementary method to the currently available hydrodesulfurization. This complement is justified since DBTs are recalcitrant in HDS but are the preferred substrates for BDS. The critical success factors for the industrial implementation will be the cost-effectiveness and the ability of the bioprocess to be integrated into existing petrochemical operations.
TRENDS FOR FUTURE WORK
The key research needs for bioprocesses to upgrade fossil fuels are the development of desulfurization biocatalysts obtained at low cost and with higher specific activity and
155 GLOBAL ANALYSIS AND CONCLUSIONS broader substrate range. The search for other alternative carbon sources derived from agro- industrial by-products or wastes, which can be easily obtained with cheaper methods (instead of hydrolysis), could reduce even more the cost involved in biocatalyst production.
It will be interesting to find out if strain 1B has also the ability to degrade carbazole. Bacterial strains that can metabolize both DBT and carbazole could be used for the simultaneous removal of both sulfur and nitrogen from petroleum. The physiological study of G. alkanivorans strain 1B in crude oil or in gasoline/diesel will be essential for the knowledge of its desulfurization ability in a very complex mixture of hydrocarbons, with the possibility of those being toxic to the cells.
Another future approach will be to investigate the exact role of some metals, especially zinc, in the activity of strain 1B enzymes. The purification of these enzymes will allow their complete characterization, verifying whether they are metallo-proteins or not. This future characterization of the enzyme structure might allow an improvement of their activity.
Finally, the attainment of recombinant bacteria carrying the dszABC genes of strain 1B that can desulfurize DBT is an obvious step to increase desulfurization activity. On the other hand, the lack of dsz substrate repression in the recombinant strains will permit the use of any alternative carbon source, even in the presence of high concentrations of sulfur sources. Despite the desulfurization improvement achieved throughout this work for strain 1B, much more developments must be carried out to achieve the promising results for other improved bacterial strains.
156 GLOBAL ANALYSIS AND CONCLUSIONS
REFERENCES
Alves, L., Melo, M., Mendonça, D., Simões, F., Matos, J., Tenreiro, R., Gírio, F.M., 2007. Sequencing, cloning and expression of the dsz genes required for dibenzothiophene sulfone desulfurization from Gordonia alkanivorans strain 1B. Enz. Microb. Technol. 40, 1598-1603.
Alves, L., Salgueiro, R., Rodrigues, C., Mesquita, E., Matos, J., Gírio, F.M., 2005. Desulfurization of dibenzothiophene, benzothiophene and other thiophene analogues by a newly isolated bacterium, Gordonia alkanivorans strain 1B. Appl. Biochem. Biotechnol. 120, 199-208.
Arenskötter, M., Bröker, D., Steinbüchel, A., 2004. Biology of the metabolically diverse genus Gordonia. Appl. Environ. Microbiol. 70, 3195-3204.
Bahrami, A., Shojaosadati, S.A., Mohebali, G., 2001. Biodegradation of dibenzothiophene by thermophilic bacteria. Biotechnol. lett. 23, 899-901.
Chang, J.H., Chang, Y.K., Ryu, H.W., Chang, H.N., 2000. Desulfurization of light gas oil in immobilized-cell systems of Gordona sp. CYKS1 and Nocardia sp. CYKS2. FEMS Microbiol. Lett. 182, 309-312.
Hirasawa, K., Ishii, Y., Kobayashi, M., Koizumi, K., Moruhashi, K., 2001. Improvement of desulfurization activity in Rhodococcus erythropolis KA2-5-1 by genetic engineering. Biosci. Biotechnol. Biochem. 65, 239-246.
Kaufman, E.N., Harkins, J.B., Barole, A.P., 1998. Comparison of batch-stirred and electrospray reactors for biodesulfurization of dibenzothiophene in crude oil and hydrocarbon feedstocks. Appl. Biochem. Biotechnol. 73, 127-143.
Kaufman, E.N., Harkins, J.B., Rodriguez, M., Tsouris, C., Selvaraj, P.T., Murphy, S.E., 1997. Development of an electro-spray bioreactor for crude oil processing. Fuel Process Technol. 52, 127-144.
Kilbane, J.J. II., 2006. Microbial biocatalyst developments to upgrade fossil fuels. Curr. Opin. Microbiol. 17, 305-314.
Kummer, C., Schumann, P., Stackebrandt, E., 1999. Gordonia alkanivorans sp. nov., isolated from tar-contaminated soil. Int. J. Syst. Bacteriol. 49, 1513-1522.
157 GLOBAL ANALYSIS AND CONCLUSIONS
Maghsoudi, S., Vossoughi, M., Kheirolomoom, A., Tanaka, E., Katoh, S., 2001. Biodesulfurization of hydrocarbons and diesel fuels by Rhodococcus sp. strain P32C1. Biochem. Eng. J. 8, 151-156.
Mirleau, P., Wogelius, R., Smith, A., Kertesz, M.A., 2005. Importance of organosulfur utilization for survival of Pseudomonas putida in soil and rhizosphere. Appl Environ Microbiol. 71, 6571-6577.
Ohshiro, T., Hirata, T., Hashimoto, I., Izumi, Y., 1996. Characterization of dibenzothiophene desulfurization reaction by whole cells of Rhodococcus erythropolis H-2 in the presence of hydrocarbon. J. Ferm. Bioeng. 82, 610-612.
Ohshiro, T., Hirata, T., Izumi, Y., 1995. Microbial desulfurization of dibenzothiophene in the presence of hydrocarbon. Appl. Microbiol. Biotechnol. 44, 249-252.
Ohshiro, T., Izumi, Y., 1999. Microbial desulfurization of organic sulfur compounds in petroleum. Biosci. Biotechnol. Biochem. 63, 1-9.
Van Hamme, J.D., Singh, A., Ward, O.P., 2003. Recent advances in petroleum microbiology. Microbiol. Mol. Biol. Rev. 67, 503-549.
158