AU715700B2 - Recombinant DNA encoding a desulfurization biocatalyst - Google Patents

Description

S F Ref: 290725D1

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

S.

S

S

S

*S.S

Name and Address of Applicant: Actual Inventor(s): Address for Service: Invention Title: Energy Biosystems Corporation 4200 Research Forest Drive The Woodlands Texas 77381 UNITED STATES OF AMERICA John Rambosek, Chris S. Piddington, Brian R.

Kovacevich, Kevin D. Young and Sylvia A. Denome Spruson Ferguson, Patent Attorneys Level 33 St Martins Tower, 31 Market Street Sydney, New South Wales, 2000, Australia Recombinant DNA Encoding a Desulfurization Biocatalyst The following statement is a best method of performing it full description known to me/us:of this invention, including the 5845 RECOMBINANT DNA ENCODING A DESULFURIZATION

BIOCATALYST

BACKGROUND

Sulfur contaminants in fossil fuels can create problems in refinery processes which can be costly to rectify. The sulfur contaminants that occur in fossil fuels fall into either of the following general classes: mineralized (inorganic, pyritic) sulfur and organic sulfur (sulfur that is covalently bound to carbonaceous molecules, referred to as organosulfur compounds). The presence of sulfur has been correlated with corrosion of pipeline, pumping and refining equipment, and with premature breakdown of combustion engines. Sulfur also poisons many catalysts which are used in the refining of fossil fuels. Moreover, the atmospheric emission of 15 sulfur combustion products, such as sulfur dioxide, leads to the form of acid deposition known as acid rain. Acid rain has lasting deleterious effects on aquatic and forest ecosystems, as well as on agricultural areas located S. downwind of combustion facilities. Monticello, D.J. and W.R. Finnerty, (1985) Ann. Rev. Microbiol. 39:371-389.

Regulations such as the Clean Air Act of 1964 require the removal of sulfur, either pre- or post-combustion, from virtually all coal- and petroleum-based fuels. Conformity with such legislation has become increasingly problematic 25 due to the rising need to utilize lower grade, highersulfur fossil fuels as clean-burning, low-sulfur petroleum reserves become depleted, as well as the progressive reductions in sulfur emissions required by regulatory authorities. Monticello, D.J. and J.J. Kilbane, "Practical Considerations in Biodesulfurization of Petroleum", IGT's d Intl. Symp. on Gas, Oil. Coal. and Env. Biotech., (Dec. 3-5, 1990) New Orleans,

LA.

One technique which is currently employed for the pre-combustion removal of organic sulfur from liquid fossil fuels, petroleum, is hydrodesulfurization (HDS). HDS is suitable for the desulfurization of fossil fuels wherein organosulfur compounds account for a significant, a major, proportion of all sulfur contaminants present. HDS is thus useful for treating crude oil or bitumen, petroleum distillate fractions or refining intermediates, liquid motor fuels, and the like. HDS is more particularly described in Shih, S.S. et al., "Deep Desulfurization of Distillate Components", Abstract No.

264B AIChE Chicago Annual Meeting, presented November 12, 1990, (complete text available upon request from the American Institute of Chemical Engineers); Gary, J.H. and G.E. Handwerk, (1975) Petroleum Refining: Technology and Economics, Marcel Dekker, Inc., New York, pp. 114-120, and Speight, (1981) The Desulfurization of Heavy Oils and Residue, Marcel Dekker, Inc., New York, pp. 119-127.

HDS is based on the reductive conversion of organic sulfur into hydrogen sulfide (H 2 S) in the presence of a metal catalyst. HDS is carried out under conditions of elevated 20 temperature and pressure. The hydrogen sulfide produced as a result of HDS is a corrosive gaseous substance, which is stripped from the fossil fuel by known techniques.

Elevated or persistent levels of hydrogen sulfide are known to poison (inactivate) the HDS catalyst, complicating the desulfurization of liquid fossil fuels that are high in sulfur.

Organic sulfur in both coal and petroleum fossil fuels is present in a myriad of compounds, some of which are termed labile in that they can readily be desulfur- 30 ized, others of which are termed refractory in that they do not easily yield to conventional desulfurization treatment, by HDS. Shih, S.S. et al. Frequently, then, even HDS-treated fossil fuels must be post-combustively desulfurized using an apparatus such as a flue scrubber.

Flue scrubbers are expensive to install and difficult to maintain, especially for small combustion facilities.

-3- Moreover, of the sulfur-generated problems noted above, the use of flue scrubbers in conjunction with HDS is directed to addressing environmental acid deposition, rather than other sulfur-associated problems, such as corrosion of machinery and poisoning of catalysts.

Recognizing these and other shortcomings of HDS, many investigators have pursued the development of microbial desulfurization (MDS). MDS is generally described as the harnessing of metabolic processes of suitable bacteria to the desulfurization of fossil fuels. Thus, MDS typically involves mild ambient or physiological) conditions, and does not involve the extremes of temperature and pressure required for HDS. It is also generally considered advantageous that biological desulfurizing agents can renew or replenish themselves under suitable conditions. Microbial desulfurization technology is reviewed in Monticello and Finnerty (1985), 39 ANN. REV. MICROBIOL.

371-389 and Bhadra et al. (1987), 5 BIOTECH. ADV. 1-27.

Hartdegan et al. (1984), 5 CHEM. ENG. PROGRESS 63-67 and 20 Kilbane (1989), 7 TRENDS BIOTECHNOL. (No. 4) 97-101 provide additional commentary on developments in the field.

Several investigators have reported mutagenizing naturally-occurring bacteria into mutant strains with the Sacquired capability of breaking down, catabolizing, dibenzothiophene (DBT). Hartdegan, F.J. et al., (May 1984) Chem. En Proress 63-67. DBT is representative of the class of organic sulfur molecules found in fossil fuels from which it is most difficult to remove sulfur by HDS. Most of the reported mutant microorganisms act upon 30 DBT nonspecifically, by cleaving carbon-carbon bonds, thereby releasing sulfur in the form of small organic breakdown products. One consequence of this microbial action is that the fuel value of a fossil fuel so treated is degraded. Isbister and Doyle, however, reported the derivation of a mutant strain of Pseudomonas which appeared to be capable of selectively liberating sulfur from -4- DBT, thereby preserving the fuel value of treated fossil fuels. U.S. Patent No. 4,562,156.

Kilbane recently reported the mutagenesis of a mixed bacterial culture, producing a bacterial consortium which appeared capable of selectively liberating sulfur from DBT by an oxidative pathway. Resour. Cons. Recycl. 3:69-79 (1990). A strain of Rhodococcus rhodocrous was subsequently isolated from the consortium. This strain, which has been deposited with the American Type Culture Collection under the terms of the Budapest Treaty as ATCC No. 53968 and also referred to as IGTS8, is a source of biocatalytic activity as described herein. Microorganisms of the ATCC No. 53968 strain liberate sulfur from forms of organic sulfur known to be present in fossil fuels, including DBT, by the selective, oxidative cleavage of carbon-sulfur bonds in organic sulfur molecules. Kilbane has described the isolation and characteristics of this strain in detail in U.S. Patent No. 5,104,801.

SUMMARY OF THE INVENTION 20 This invention relates in one aspect to a deoxyribonucleic acid (DNA) molecule containing one or more genes encoding one or more enzymes that, singly or in concert with each other, act as a biocatalyst that desulfurizes a fossil fuel that contains organic sulfur molecules. The 25 DNA molecule of the present invention can be purified and isolated from a natural source, or can be a fragment or portion of a recombinant DNA molecule that is, e.g., integrated into the genome of a non-human host organism.

The gene or genes of the present invention can be obtained from, a strain of Rhodococcus rhodochrous microorganisms having suitable biocatalytic activity. That is, the gene or genes of the present invention can be obtained from a non-human organism, a microorganism, that expresses one or more enzymes that, singly or in concert with each other, act as a desulfurizing biocatalyst.

Biocatalysis, as described more fully below, is the selective oxidative cleavage of carbon-sulfur bonds in organosulfur compounds. The present invention is particularly useful for the desulfurization of fossil fuels that contain organosulfur compounds,

DBT.

The invention further relates to recombinant

DNA

vectors, recombinant DNA plasmids and non-human organisms that contain foreign (recombinant, heterologous)

DNA

encoding a biocatalyst capable of desulfurizing a fossil fuel which contains organosulfur compounds. Such organisms are referred to herein as host organisms.

The invention described herein thus encompasses ribonucleic acid (RNA) transcripts of the gene or genes of the present invention, as well as polypeptide expression product(s) of the gene or genes of the present invention.

The present polypeptide expression products, after such post-translational processing and/or folding as is necessary, and in conjunction with any coenzymes, cofactors or coreactants as are necessary, form one or more protein S 20 biocatalysts (enzymes) that, singly or in concert with each other, catalyze (promote, direct or facilitate) the removal of sulfur from organosulfur compounds that are found in fossil fuels. This is accomplished by the selective, oxidative cleavage of carbon-sulfur bonds in said compounds. The biocatalyst of the present invention can be a non-human host organism, viable cultured) or non-viable heat-killed) containing the DNA of the present invention and expressing one or more enzymes encoded therein, or it can be a cell-free preparation 30 derived from said organism and containing said one or more biocatalytic enzymes.

In another aspect, the present invention relates to a method of desulfurizing a fossil fuel using the above mentioned non-human organism, said organism expressing a desulfurizing biocatalyst. Alternatively, the present invention relates to a method of desulfurizing a fossil 6 fuel using a biocatalyst preparation comprising one or more enzymes isolated from said organism. The process involves: 1) contacting said organism or biocatalyst preparation obtained therefrom with a fossil fuel that contains organic sulfur, such that a mixture is formed; and 2) incubating the mixture for a sufficient time for the biocatalyst expressed by or prepared from the organism to desulfurize the fossil fuel. The biocatalytically treated fossil fuel obtained following incubation has significantly reduced levels of organosulfur compounds, compared to a sample of the corresponding untreated fossil fuel.

Thus, according to one embodiment of the invention, there is provided a method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a Io transformed microorganism containing a recombinant DNA molecule of Rhodococcus origin wherein said transformed microorganism expresses a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the microorganism mixture under conditions sufficient to bring about the catalytic cleavage of organic carbon-sulfur bonds, whereby the organic sulfur content of the fossil fuel is significantly reduced.

In a preferred aspect of this embodiment of the invention, there is provided a method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a 20 biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: 2 (ORF-1), comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under 25 conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, whereby the organic sulfur content of the fossil fuel is significantly reduced.

a [I :\DayLib\LIBA12872.doc:TLT 6a In another preferred aspect of this embodiment of the invention, there is provided a method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: 3 (ORFcomprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, I0 whereby the organic sulfur content of the fossil fuel is significantly reduced.

In another preferred aspect of this embodiment of the invention, there is provided a method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur is molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: 5 (ORFcomprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, 20 whereby the organic sulfur content of the fossil fuel is significantly reduced.

In another preferred aspect of this embodiment of the invention, there is provided a method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur 25 molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID. Nos: 2 (ORF- 3 (ORF-3) and 5 (ORF-2), comprising the steps of: *9 a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, .o whereby the organic sulfur content of the fossil fuel is significantly reduced.

I :\DayLib\LIBA]2872.doc:TLT 6b According to another embodiment of the invention, there is provided a method for desulfurizing dibenzothiophene, comprising the step of contacting dibenzothiophene with a transformed microorganism containing a recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more reactions in the desulfurization of dibenzothiophene, wherein said recombinant nucleic acid molecule has the sequence of a nucleic acid molecule isolated from Rhodococcus, or a complement of said isolated nucleic acid molecule.

According to another embodiment of the invention, there is provided a method for desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a recombinant nucleic acid molecule which encodes an enzyme which catalyzes the conversion of dibenzothiophene to dibenzothiophene sulfone, wherein the enzyme comprises an active fragment of the enzyme having the amino acid sequence shown in ORF-3 (SEQ ID NO: 3).

According to a further embodiment of the invention, there is provided a method for i desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more steps in the conversion of dibenzothiophene sulfone to 2-hydroxybiphenyl, wherein the enzymes comprise an active fragment of the enzyme having the amino acid sequence shown in ORF- S: 20 1 (SEQ ID NO: 2).

According to yet another embodiment of the invention, there is provided a method for S. desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more steps in the 25 conversion of dibenzothiophene sulfone to 2-hydroxybiphenyl, wherein the enzymes comprise an active fragment of the enzyme having the amino acid sequence shown in ORF- 2 (SEQ ID NO: 0 an So--<

N::

[I:\DayLib\LBA]272. doc:TLTf 6c In yet another aspect, the invention relates to nucleic acid probes which hybridize to the recombinant DNA of the present invention.

In still other aspects, the present invention relates to the production of new non-human organisms containing the recombinant DNA of the present invention and preferably S expressing the biocatalyst encoded therein. Availability of the recombinant DNA of this invention greatly simplifies and facilitates the production and purification of biocatalysts for desulfurizing a fossil fuel. Costly and time consuming procedures involved in the purification of biocatalysts can be reduced, eliminating the need to generate the biocatalyst from one or more non-human organisms in which it is naturally present or has been 0 produced by mutagenesis. More specifically, non-human host organisms can be generated which express the gene or genes of the present invention at elevated levels. In addition, the invention described herein furthers the discovery of genes encoding desulfurization biocatalysts in additional non-human organisms. This objective can be accomplished using the nucleic acid probes of the present invention to screen DNA libraries prepared from one i or more additional non-human organisms in whom biocatalytic function is known or suspected to be present. Any genes present in such a a a a oo oo o o r. [I:\DayLib\LI BA]2872.doc:TLT -7organisms and encoding desulfurization biocatalysts or components thereof can be replicated at large scale using known techniques, such as polymerase chain reaction

(PCR).

PCR advantageously eliminates the need to grow the nonhuman organisms, in culture, in order to obtain large amounts of the DNA of interest.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1,is a flow diagram schematic illusting a stepwise procedure for the isolation of the recombinant DNA of the present invention.

Figure 2 is a schematic illustration of the Rhodococcus rhodochrous replication competent and chloramphenicol resistant vector pRF29, said vector having been derived from Rhodococcus fascians.

Figure 3 is a schematic illustration of the Rhodococcus rhodochrous replication competent and chloramphenicol resistant vector pRR-6.

Figure 4 is a schematic illustration of the restriction map for DNA plasmid pTOXI-1 encoding a biocatalyst 20 capable of carbon-sulfur bond cleavage.

Figure 5 is a schematic illustration of the restriction map for subclone pMELV-1, derived from plasmid pTOXI- 1.

Figure 6 is a schematic illustration of the restriction map for pMELV-1 and fragments thereof present as inserts in subclones pSMELV-1A, pSMELV-2A, pSMELV-3A and pSMELV-4A.

Figure 7 is a schematic illustration of the predicted locations within the sequence of pTOXI-1 of three nearly contiguous open reading frames (ORFs; specifically, ORF 1, ORF 2 and ORF 3) encoding polypeptide expression products responsible for the Dsz+ phenotype.

Figure 8 is a schematic illustration of the restriction map of pTOXI-l and fragments thereof present as inserts in subclones pENOK-1, pENOK-2, pENOK-3, pENOK-11, -8pENOK-13, pENOK-16, pENOK-18, pENOK-Nsi, pENOK-19

AND

Figure 9 is a schematic illustration of the restriction map of pRR-12.

Figure 10 is a schematic illustration of the restriction map of vector pKAMI. In the inset, the engineered cloning site present in pKAMI is shown in detail.

Figure 11 is a schematic illustration of the restriction map of pSBG-2, in which expression of a promoterless Dsz gene cluster from pTOXI-1 is driven by the chloramphenicol resistance promoter.

DETAILED DESCRIPTION OF THE PREFERRED

EMBODIMENT

In the petroleum extraction and refining arts, the term "organic sulfur" is generally understood as referring to organic molecules having a hydrocarbon framework to which one or more sulfur atoms (called heteroatoms) are covalently joined. These sulfur atoms can be joined directly to the hydrocarbon framework, by one or 'more carbon-sulfur bonds, or can be present in a substituent joined to the hydrocarbon framework of the molecule, a sulfonyl group (which contains a carbonoxygen-sulfur covalent linkage). The general class of organic molecules having one or more sulfur heteroatoms Sare sometimes referred to as "organosulfur compounds".

25 The hydrocarbon portion of these compounds can be aliphatic, aromatic, or partially aliphatic and partially aromatic.

Cyclic or condensed multicyclic organosulfur compounds in which one or more sulfur heteroatoms are linked to adjacent carbon atoms in the hydrocarbon framework by aromatic carbon-sulfur bonds are referred to as "sulfurbearing heterocycles". The sulfur that is present in many types of sulfur-bearing heterocycles is referred to as "thiophenic sulfur" in view of the five-membered aromatic ring in which the sulfur heteroatom is present. The -9simplest such sulfur-bearing heterocycle is thiophene, which has the composition

CHS.

Sulfur-bearing heterocycles are known to be stable to conventional desulfurization treatments, such as HDS. For this reason, they are said to be refractory or recalcitrant to HDS treatment. Sulfur-bearing heterocycles can have relatively simple or relatively complex chemical structures. In complex heterocycles, multiple condensed aromatic rings, one or more of which can be heterocyclic, are present. The difficulty of desulfurization increases with the structural complexity of the molecule. Shih et al_ That is, refractory behavior is most accentuated in complex sulfur-bearing heterocycles, such as dibenzothiophene (DBT, C 12

HS).

DBT is a sulfur-bearing heterocycle that has a condensed, multiple aromatic ring structure in which a fivemembered thiophenic ring is flanked by two six-membered benzylic rings. Much of the residual post-HDS organic sulfur in fossil fuel refining intermediates and com- 20 bustible products is thiophenic sulfur. The majority of this residual thiophenic sulfur is present in DBT and derivatives thereof having one or more alkyl or aryl radicals attached to one or more carbon atoms present in one or both flanking benzylic rings. Such DBT derivatives are said to be "decorated" with these radicals.

DBT

itself is accepted in the relevant arts as a model compound illustrative of the behavior of the class of compounds encompassing DBT and alkyl- and/or aryl-decorated derivatives thereof in reactions involving thiophenic 30 sulfur. Monticello and Finnerty (1985), Microbial desulfurization of fossil fuels, 39 ANNUAL REVIEWS IN MICRO- BIOLOGY 371-389, at 372-373. DBT and radical-decorated derivatives thereof can account for a significant percentage of the total sulfur content of particular crude oils, coals and bitumen. For example, these sulfur-bearing heterocycles have been reported to account for as much as 70 wt% of the total sulfur content of West Texas crude oil, and up to 40 wt% of the total sulfur content of some Middle East crude oils. Thus, DBT is considered to be particularly relevant as a model compound for the forms of thiophenic sulfur found in fossil fuels, such as crude oils, coals or bitumen of particular geographic origin, and various refining intermediates and fuel products manufactured therefrom. Id. Another characteristic of DBT and radical-decorated derivatives thereof is that, following a release of fossil fuel into the environment, these sulfur-bearing heterocycles persist for long periods of time without significant biodegradation. Gundlach et al. (1983), 221 SCIENCE 122-129. Thus, most prevalent naturally occuring microorganisms do not effectively metabolize and break down sulfur-bearing heterocycles.

A fossil fuel that is suitable for desulfurization treatment according to the present invention is one that e: contains organic sulfur. Such a fossil fuel is referred to as a "substrate fossil fuel". Substrate fossil fuels 20 that are rich in thiophenic sulfur (wherein a significant fraction of the total organic sulfur is thiophenic sulfur, present in sulfur-bearing heterocycles, DBT) are particularly suitable for desulfurization according to the method described herein. Examples of such substrate fossil fuels include Cerro Negro or Orinoco heavy crude oils; Athabascan tar and other types of bitumen; petroleum refining fractions such as light cycle oil, heavy atmospheric gas oil, and No. 1 diesel oil; and coal-derived liquids manufactured from sources such as Pocahontas #3, 30 Lewis-Stock, Australian Glencoe or Wyodak coal.

Biocatalytic desulfurization (biocatalysis or BDS) is the excision (liberation or removal) of sulfur from organosulfur compounds, including refractory organosulfur compounds such as sulfur-bearing heterocycles, as a result of the selective, oxidative cleavage of carbon-sulfur bonds in said compounds by a biocatalyst. BDS treatment -11yields the desulfurized combustible hydrocarbon framework of the former refractory organosulfur compound, along with inorganic sulfur substances which can be readily separated from each other by known techniques such as frational distillation or water extraction. For example, DBT is converted into hydroxybiphenyl or dihydroxybiphenyl, or a mixture thereof, when subjected to BDS treatment.

BDS

is carried out by a biocatalyst comprising one or more non-human organisms microorganisms) that functionally express one or more enzymes that direct, singly or in concert with each other, the removal of sulfur from organosulfur compounds, including sulfur-bearing heterocycles, by the selective cleavage of carbon-sulfur bonds in said compounds; one or more enzymes obtained from such microorganisms; or a mixture of such microorganisms and enzymes. Organisms that exhibit biocatalytic activity are referred to herein as being CS+ or Dsz+. Organisms that lack biocatalytic activity are referred to herein as being CS- or Dsz-.

20 As summarized above, the invention described herein relates in one aspect to a DNA molecule or fragment thereof containing a gene or genes which encode a biocatalyst capable of desulfurizing a fossil fuel that contains S. organosulfur compounds. The present DNA molecule or fragment thereof can be purified and isolated DNA obtained from, a natural source, or can be recombinant (heterologous or foreign) DNA that is, present in a nonhuman host organism. The following discussion, which is not to be construed as limiting on the invention in any 30 way but is presented for purposes of illustration, recounts the isolation of DNA encoding a desulfurizing biocatalyst from a strain of Rhodococcus rhodochrous,

ATCC

No. 53968, that is known to express suitable biocatalytic activity. This preferred strain of Rhodococcus rhodocrous is disclosed in U.S. Patent No. 5,104,801 (issued 1992), the teachings of which are incorporated herein by refer- -12ence, and has been referred to in the literature as IGTS8.

IGTS8 was developed by investigators at the Institute of Gas Technology in Chicago IL. Other organisms that are known to express suitable biocatalytic activity and thus are viewed as suitable sources of the DNA of the present invention include the strain of Bacillus sulfasportare described in U.S. Patent 5,002,888 and deposited with the American Type Culture Collection as ATCC No. 53969 and the Corynebacterium strain described in Omori et al. (1992), Desulfurization of dibenzothiophene by Corynebacterium sp.

strain SYL, 58 APPL. ENV. MICROBIOL. (No. 3) 911-915. The isolation of the DNA of the present invention from the ATCC No. 53968 microorganism is schematically depicted in Figure 1, and will now be described.

Mutant strains of R. rhodochrous that are incapable of cleaving carbon-sulfur bonds (CS- or Dsz-), are produced by exposing a strain of R. rhodochrous,

ATCC

No. 53968, that exhibits biocatalytic activity (that is CS+ or Dsz+), to a mutagen under appropriate conditions 20 that are known to or readily ascertainable by those skilled in the art. Suitable mutagens include radiation, ultraviolet radiation, and chemical mutagens, e.g., N-methyl-N'-nitrosoguanidine (NTG), hydroxylamine, ethylmethanesulphonate (EMS) and nitrous acid. Mutants thus formed are allowed to grow in an appropriate medium and screened for carbon-sulfur bond cleavage activity. Mutants identified as lacking carbon-sulfur bond cleavage activity are termed CS-. Any method of screening which allows for an accurate detection of carbon-sulfur bond S 30 cleavage activity is suitable in the method of the present invention. Suitable methods of screening for this activity include exposing the different mutants to carbonsulfur bond containing molecules DBT) and measuring carbon-sulfur bond cleavage. In a preferred embodiment, the mutants are exposed to DBT, such that the breakdown product, hydroxybiphenyl (HBP), which fluoresces under -13short wave ultraviolet light, is produced. HBP can also be detected colorimetrically through its blue reaction product with Gibbs' reagent. Other methods include gas and liquid chromatography, infrared and nuclear magnetic resonance spectrometry. See Kodama, et al., Applied and Environmental Microbiology, pages 911-915 (1992) and Kilbane and Bielaga, Final Report D.O.E. Contract No. DE- AC22-88PC8891 (1991). Once CS- mutants are identified and isolated, clones thereof are propagated using standard techniques and subjected to further analysis.

Concurrent with the mutagenesis of the above-described culture of the CS+ organism, R. rhodochrous, a second culture of the same CS+ organism is maintained in culture. CS+ organism DNA is extracted from this culture of R. rhodocrous. Various methods of DNA extraction are suitable for isolating the DNA of this organism.

Suitable methods include phenol and chloroform extraction.

See Maniatis et al., Molecular Clonin. A Laboratory Manual, 2d, Cold Spring Harbor Laboratory Press, page 20 16.54 (1989), herein referred to as Maniatis et al..

Once the DNA is extracted from R. rhodochrous 1, the DNA is cut into fragments of various kilobase lengths, which collectively make up DNA library 5. Various methods of fragmenting the DNA of R. rhodochrous to purify DNA therefrom, including the DNA of the present invention, can be used, enzymatic and mechanical methods. Any four-base recognition restriction endonuclease such as TaqI or Sau 3A is suitable for fragmenting the DNA.

Suitable methods of fragmenting DNA can be found in Mani- 30 atis et al..

The various DNA fragments are inserted into several CS- mutant clones of R. rhodochrous with the purpose of isolating the fragment of DNA that encodes the biocatalyst of the present invention. The transformation of a previously CS- mutant cell to a CS+ transformed cell is evidence that the inserted DNA fragment encodes a bio- -14catalyst. Any method of inserting DNA into R. rhodochrous which allows for the uptake and expression of said fragment is suitable. In a preferred embodiment, electroporation is used to introduce the DNA fragment into R.

rhodochrous. See Maniatis et al..

Once transformed, CS+ mutant R. rhodochrous 7 has been produced and identified, DNA fragment 9 encoding the CS+ biocatalyst can be identified and isolated. The encoded biocatalyst can then be produced using the isolated DNA in various methods that are well known and readily available to those skilled in the art. In addition, the isolated DNA can be sequenced and replicated by known techniques, the techniques described in Maniatis et al..

As noted previously, the above-described method for isolating the DNA of the present invention can be applied to CS+ organisms other than R. rhodocrous microorganisms, of the strain ATCC No. 53968. Thus, Bacillus sulfasportare ATCC No. 53969 or Corynebacterium sp. SY1 can be 20 used as the source organism for the DNA of the present invention. Furthermore, once isolated, the DNA of the present invention can be transfected into a non-human host organism other than a CS- mutant of the source organism.

Thus, the DNA of the present invention can be transfected into, a suitable strain of Escherichia coli bacteria. Other types of non-human host organism can also be used, including unicellular organisms yeast) and cells established in culture from multicellular organisms.

Other methods of isolating the DNA of the present 30 invention, include variations on the rationale described above and depicted in Figure 1. For example, it would be possible to randomly insert a CS- DNA plasmid into clones of a CS+ strain of R. rhodochrous. DNA encoding a CS+ biocatalyst could then be identified by screening for clones that have been transformed from CS+ to CS-.

The recombinant DNA molecule or fragment thereof of the present invention is intended to encompass any DNA resulting from the insertion into its chain, by chemical or biological means, of one or more genes encoding a biocatalyst capable of selectively cleaving carbon-sulfur bonds, said gene not originally present in that chain.

Recombinant DNA includes any DNA created by procedures using restriction nucleases, nucleic acid hybridization, DNA cloning, DNA sequencing or any combination of the preceding. Methods of construction can be found in Maniatis et al., and in other methods known by those skilled in the art.

Procedures for the construction of the DNA plasmids or vectors of the present invention include those described in Maniatis et al. and other methods known by those skilled in the art. Suitable plasmid vectors include pRF- 29 and pRR-6 depicted in Figures 2 and 3, respectively.

S" The terms "DNA plasmid" and "vector" are intended to encompass any replication competent plasmid or vector 20 capable of having foreign or exogenous DNA inserted into it by chemical or biological means and subsequently, when transfected into an appropriate non-human host organism, of expressing the product of the foreign or exogenous

DNA

insert of expressing the biocatalyst of the present invention). In addition, the plasmid or vector must be receptive to the insertion of a DNA molecule or fragment thereof containing the gene or genes of the present invention, said gene or genes encoding a biocatalyst that selectively cleaves carbon-sulfur bonds in organosulfur S. 30 compounds. Procedures for the construction of DNA plasmid vectors include those described in Maniatis et al. and others known by those skilled in the art.

The plasmids of the present invention include any DNA fragment containing a gene or genes encoding a biocatalyst that selectively cleaves carbon-sulfur bonds in organosulfur compounds. The term "plasmid" is intended to -16encompass any DNA fragment. The DNA fragment should be transmittable to a host microorganism by transformation or conjugation. Procedures for the construction or extraction of DNA plasmids include those described in Maniatis et al. and others known by those skilled in the art.

The transformed non-human host organisms of the present invention can be created by various methods by those -skilled in the art. For example, transfection electroporation as explained by Maniatis et al. can be used. By the term "non-human host organism" is intended any non-human organism capable of the uptake and expression of foreign, exogenous or recombinant DNA,

DNA

not originally a part of the organism's nuclear material.

The method of desulfurizing a fossil fuel of the present invention involves two aspects. First, a host organism or biocatalytic preparation obtained therefrom is contacted with a fossil fuel to be desulfurized. This can be done in any appropriate container, optionally fitted .with an agitation or mixing device. The mixture is com- 20 bined thoroughly and allowed to incubate for a sufficient time to allow for cleavage of a significant number of carbon-sulfur bonds in organosulfur compounds, thereby producing a desulfurized fossil fuel. In one embodiment, an aqueous emulsion is produced with an aqueous culture of 25 the organism and the fossil fuel, allowing the organism to propagate in the emulsion while the expressed biocatalyst Scleaves carbon-sulfur bonds.

Variables such as temperature, mixing rate and rate of desulfurization will vary according to the organism 30 used. The parameters can be determined through no more than routine experimentation.

Several suitable techniques for monitoring the rate and extent of desulfurization are well-known and readily available to those skilled in the art. Baseline and timecourse samples can be collected from the incubation mixture, and prepared for a determination of the residual -17organic sulfur in the fossil fuel. The disappearance of sulfur from organosulfur compounds, such as DBT, in the sample being subjected to biocatalytic treatment can be monitored using, X-ray fluorescence (XRF) or atomic emission spectrometry (flame spectrometry). Preferably, the molecular components of the sample are first separated, by gas chromatography.

The nucleic acid probes of the present invention include any nuclear material capable of hybridizing to at least a portion of the DNA of the present invention. The term "nucleic acid probe" includes any nuclear material capable of hybridizing to DNA.

The invention will now be further illustrated by the following specific Examples, which are not to be viewed as limiting in any way.

EXAMPLE 1. ISOLATION OF DNA ENCODING A DESULFURIZATION ACTIVE BIOCATALYST.

As used herein, the term "Dsz+" refers to the ability of an organism to utilize thiophenic compounds such as 20 dibenzothiophene (DBT) as the sole source of sulfur by the selective cleavage of carbon-sulfur bonds therein. Rhodococcus rhodochrous strain IGTS8 demonstrates the Dsz phenotype. The term "Dsz-" referrs to an organism's inability to utilize said thiophenic compounds as a sole source of sulfur by the selective cleavage of carbonsulfur bonds therein.

*go.

Materials Bacterial strains and plasmids Rhodococcus rhodochrous strain IGTS8 (ATCC No.

53968), obtained from the Institute of Gas Technology (Chicago, IL), was used as a parent strain for production of mutant strains which have lost the desulfurization phenotype Strain IGTS8 was also used for isolation of DNA fragments capable of complementing said -18mutants to produce Dsz+ mutants therefrom. Rhodococcus vector pRF-29 was obtained from the Institute of Gas Technology. The construction of pRF-29 is described in Desomer, et al. (1990), Transformation of Rhodococcus fascians by High-Voltae Electroporation and Development of R. fascians Cloning Vectors, APPLIED AND ENVIRONMENTAL MICROBIOLOGY 2818-2825. The structure of pRF-29 is schematically depicted in Figure 2.

Escherichia coli strain JM109 was used as a host in transformation with plasmid constructs derived from the plasmids pUC18 and pUC19 (Bethesda Research Laboratories, Bethesda, MD).

Enzymes and Reagents Restriction endonucleases were purchased from Bethesda Research Laboratories (BRL) and New England Biolabs (Beverly, MA). T4 ligase and the Klenow fragment of E. coli DNA polymerase I were purchased from BRL. HKTM Phosphatase was purchased from Epicentre Technologies (Madison, WI). All enzymes were used in accordance with 20 manufacturers recommendations. Enzyme assay substrates Dibenzothiophene (DBT), Dibenzothiophene 5-oxide (DBT sulfoxide) and Dibenzothiphene sulfone (DBT sulfone) were purchased from Aldrich (Milwaukee, WI). Gibb's Reagent, *0 2, 6 -dicholoroquinone-4-chloroimide, was purchased from Sigma (St. Louis, MO). Chemical mutagen N-methyl-N'- S nitro-N-nitrosoguanidine (NTG) was also purchased from Sigma.

Growth Media and Conditions E. coli JM109 was grown in L-broth (Difco, Detroit, MI). Transformants were selected on L-plates supplemented with 1.5% agar and containing 125gg/ml ampicillin.

E.

coli strains were grown at 37'C. Rhodococcus strains were maintained on Rhodococcus Media (RM) composed per liter of: 8.0g Nutrient Broth (Difco), 0.5g yeast extract, 10.0g -19glucose. Transformants of Rhodococcus strains were selected on RM plates supplemented with 1.5% agar and containing 25pg/ml chloramphenicol. For expression of the Dsz+ phenotype, Rhodococcus strains were grown in Basal Salts Media (BSM) composed per liter of: 2.44g KH 2

PO

4 5.57g Na 2

HPO

4 2.0g NH 4 C1,-0.2 g MgC1 2 .6H 2 0, 0.001g CaC12.2HO2, 0.001g FeC16H 2 O, 0.004g MnC1 2 .4H 2 0, 6.4ml glycerol. Optionally, BSM can be supplemented with glucose. Rhodococcus strains were grown at Methods Sulfur Bioavailability Assay The sulfur bioavailability assay, described in U.S.

Patent 5,104,801, examines an organism's ability to liberate organically bound sulfur from substrates DBT, DBT sulfoxide, DBT sulfone) for use as the sole source of sulfur for growth. In the assay, BSM, which contains no sulfur, is supplemented with one or more sulfur containing substrates, DBT. The organism's ability to liberate sulfur therefrom is measured by its ability to grow with proper incubation, as monitored by optical density at 600 nm.

Gibbs Assay for 2-Hydroxybiphenvl The oxidative product of DBT, DBT sulfoxide and DBT 25 sulfone incubated with strain IGTS8 is 2-hydroxybiphenyl (2-HBP). The Gibbs assay colorimetrically quantitates the amount of 2-HBP produced from DBT and its above-mentioned oxidative derivatives. The assay measures 2-HBP produced in culture supernatants after incubation with DBT. The 30 media must be adjusted to pH 8.0 before the Gibb's reagent is added. Gibb's Reagent, 2,6- dicholoroquinone-4-chloroimide (10mg/ml in ethanol), is added to culture supernatants at 1:100 Color development is measured as absorbance at 610nm after a 30 minute incubation at room temperature.

HPLC Assay for 2-Hdroxvbiphenyl 2-HBP production cultures incubated with DBT can also be detected by HPLC using instrumentation available from Waters, Millipore Corporation, Milford, MA. Reagent alcohol is added to culture broth at 1:1 in order to solubilize all remaining DBT and 2-HBP. Samples are agitated for 5 min at 220 rpm. Extracted broth samples are removed and centrifuged to remove cellular mass.

Clarified supernatants are then analyzed by HPLC with the following conditions: Column: Waters 4p Phenyl Novapak Detection Parameters: DBT 233nm, 1.0 AUFS 2-HBP 248nm, 0.2 AUFS Quantitative Detection Limits: DBT 10 250 pM 2-HBP 6 60 'pM Mobile Phase: Isocratic 70% Acetonitrile Retention times: *DBT 4.5 minutes 2-HBP 2.9 minutes IGTS8 Mutagenesis In order to generate mutant strains of R. rhodochrous which did not metabolize DBT (Dsz- mutants), biocatalyst source strain IGTS8 (Dsz+) was subjected to mutagenesis by short-wave UV light and to chemical mutagenesis with Nethyl-N'-nitro-N-nitrosoguanidine (NTG). With UV ex- 30 posure mutagenesis, a kill rate of greater than 99% was targeted. Continuously stirred R. rhodochrous cells at an optical density (A60) of 0.3 were subjected to UV exposure from a Mineralight Lamp Model UVG-254 (Ultra-violet Pro- -21ducts, Inc., San Gabriel, CA) at a distance of 10 cm for to 65 seconds to obtain this kill rate (97.9-99.9%).

For NTG mutagenesis, cell suspensions were treated with 500 pg/ml NTG for a duration determined to achieve a kill rate of 30%-50%. Combination mutagenesis utilizing both NTG and UV was also done. For these an overall kill rate of greater than 99.9% was used. Colonies surviving mutagenesis were picked onto RM plates and screened for the Dsz- phenotype as described below.

Screening Example A: Initially, a DBT-spray plate screen was used to select Dsz- mutants. Mutant colonies were replica plated onto Basal Salts Media (BSM) electrophoretic-grade agarose plates which contained no added sulfur. Colonies were allowed to grow at 30'C for 24hr.

The plates were then sprayed with an even coating of DBT dissolved in ether and incubated at 30'C for 90 minutes. The plates were then wiped clean and observed under short-wave UV light. The observed end product of DBT metabolism, 2 -hydroxybiphenyl (2-HBP) fluoresces under short-wave UV light. Colonies that produce 2-HBP are thus identified by fluorescent spots on the agarose. Colonies that do not produce 2-HBP (that are Dsz-) do not produce fluorescent spots.

Screening Example B: A simpler variation of screening 25 involved replica plating surviving mutagenized colonies to BSM agarose plates supplemented with 1.

2 ml/liter of a saturated ethanol solution of DBT. After 24 hours, production of 2-HBP can be visualized under UV illumination as above.

30 Mutants which did not appear to produce 2-HBP by the above-described screening methods were examined with the sulfur bioavailability assay, with DBT as the sole source of sulfur. Growth of potential mutants was examined in I -22- 1.25ml liquid fermentations in BSM plus DBT media dispensed in 24-well plates (Falcon). After a one day incubation at 30°C, 2-HBP production was monitored by the Gibbs colorimetric assay. Strains which continue to demonstrate the Dsz- phenotype were incubated in larger volumes of BSM plus DBT and analyzed for 2-HBP or intermediates by the HPLC method. Because BSM is a defined minimal medium, a duplicate control culture which contained supplemental inorganic sulfur was grown in order to distinguish true Dsz- mutants from auxotrophic mutants.

Mutants which failed to grow in both the control and experimental media were assumed to be auxotrophic mutants.

Of 1970 individually analyzed potential mutants, two were identified as Dsz-. One mutant, GPE-362, was generated by NTG mutagenesis. The other, CPE-648, was generated by combination NTG/UV mutagenesis. Both GPE-362 and CPE-648 grow slowly in sulfur bioavailability assays, presumably from trace amounts of sulfur on the glassware or in the media components. However, no detectable amounts of 2-HBP were produced by either mutant after an extended incubation of 6 to 10 days with DBT, as assessed with either the Gibbs assay or the HPLC assay. Thus, independently produced R. rhodocrous IGTS8 mutants GPE-362 and CPE-648 were Dsz- organisms.

Vector Construction Vector constructs were derived from R. rhodochrous and confer chloramphenicol resistance. All constructs were developed in E. coli strain JM109. Transformation of JM109 was carried out with the Gene Pulsar (Bio-Rad Laboratories, Richmond, CA) according to manufacturer's recommendations. Plasmid isolation from JM109 was performed by standard methods (Birnboim and Doly (1979), A rapid alkaline extraction procedure for screening recombinant plas- Smid DNA, 7 NUCLEIC ACIDS RES. 1513-1523; Maniatis et al.

35 (1982), MOLECULAR CLONING: A LABORATORY MANUAL (Cold -23- Spring Harbor Laboratory Press). Transformants containing correct vector constructs were identified by restriction analysis.

Vector Construct A: pRR-6 (Figure 3) contains the Rhodococcus origin of replication and Chloramphenicol resistance marker (CmR). The ori and CmR have been removed from pRF-29 as a 6.9kb XhoI/Xba (partial) fragment. The ends were made blunt with Klenow and ligated to SaII/Xbal cut pKF39. pKF39 is pUCl8 with the Small site replaced with a BgIII site. A unique NarI site is available for cloning in pRR-6. NarI ends are compatible with 4-base recognition endonuclease TaqI.

Transformation of Rhodococcus rhodochrous Transformation of IGTS8 and Dsz- mutants thereof can be achieved by electroporation. The following conditions were used in all transformations of Rhodococcus rhodochrous. Cells were grown in RM to mid-log phase and harvested by centrifugation (5000xg), then washed three times in cold, deionized, distilled water and concentrated 50-fold in 10% glycerol. The resulting cell concentrate could be used for electroporation directly or stored at Electroporations were carried out with the Gene Pulser (Bio-Rad) apparatus. 100 pl cells were mixed with 25 transformation DNA in a 2-mm gapped electrocuvette (Bio- Rad) and subjected to a 2.5 kV pulse via the pulse controller (25 pF capacitor, 200 0 external resistance).

Pulsed cells were mixed with 400 P RM and incubated for 4 hours at 30*C with regular agitation. Cells were then plated to RM supplemented with proper antibiotic.

When IGTS8 was transformed with pRF-29, chloramphenicol resistant colonies were cleanly selected at a frequency of 105 10 6 /pg DNA on plates containing pg/ml chloramphenicol.

-24- Small Scale Plasmid Preparation from R. rhodochrous A single colony of Rhodococcus rhodochrous was used to inoculate 2 to 7ml of RM plus 25 pg/ml chloramphenicol.

The culture was incubated for two days at 30'C with shaking. Cells were pelleted by centrifugation and resuspended in 300 p1 sucrose buffer (20% sucrose, 0.05 M Tris- Cl pH 8.0, 0.01 M EDTA 0.05 M NaC1, 10 mg/ml lysozyme) and incubated at 37'C for 1 hour. 300 p1 Potassium acetateacetate solution, pH 4.8 (60 ml 5 M KOAc, 11.5 ml Glacial acetic acid, 28.5 ml dH 2 was added and the mixture was gently mixed by inversion. The mixture was placed on ice for 5 minutes and then cellular debris was pelleted by centrifugation. 500 p1 supernatant was removed to a fresh tube to which RNAse was added to 0.05 pg/pl and incubated for 20 minutes at 37°C. The sample was then phenol:chloroform extracted and the aqueous layer was precipitated at -80°C with an equal volume of isopropanol.

DNA

was pelleted by centrifugation and resuspended in 0.3 M NaOAc pH 8.0. DNA was precipitated again at -80°C with an equal volume of isopropanol. DNA was pelleted by centrifugation and resuspended in 0.3 M NaOAc pH 8.0. DNA was precipitated again at -80°C with two volumes of 95% EtOH.

Pelleted DNA was washed with 70% EtOH and resuspended in 50 pl TE (Tris EDTA).

25 Isolation of Genomic DNA from R. rhodochrous Strain IGTS8 IGTS8 genomic DNA was isolated as described. 20 ml RM was inoculated with a single colony of IGTS8 and incubated at 30°C for 48 hours with shaking at 220 rpm. Cells were harvested by centrifugation (5000xg). Cells were resuspended in 10ml TE (10 mM Tris Base, 1 mM EDTA) with 100 mg lysozyme and incubated for 30 minutes at Cells were lysed by adding 1 ml of 20% sodium dodecyl sulfate (SDS). 10 ml of TE-saturated phenol and 1.5 ml M NaCI were added immediately and the mixture was gently 35 agitated for 20 minutes at room temperature. Phenol was removed by centrifugation, and the aqueous layer was extracted twice with an equal volume of chloroform. An equal volume of isopropanol was added to the aqueous layer to precipitate the DNA. DNA was spooled onto a pasteur pipette and redissolved in TE. DNA was then RNased with pg/ml MI or 1 hour at 37*C. The sample was made to a final concenia lon of 100 mM NaC1 and 0.4% SDS and proteased with 100 pg/ml protease K. The sample was then extracted with phenol and chloroform and precipitated with isopropanol as before. The purified genomic DNA, which included the DNA of the present invention, was resuspended in TE.

Construction of Plasmid Library of IGTS8 Genomic DNA from the Dsz+ source organism (IGTS8) was cut with TaqI in order to produce fragments 0.5 23 kb in length. Cut DNA was electrophoresed through 0.8% low melting temperature agarose and DNA fragments greater than kb in length were isolated and purified by standard methods (Maniatis, T. et al. (1982), MOLECULAR CLONING:

A

LABORATORY MANUAL (Cold Spring Harbor Laboratory Press)).

Vector pRR-6 was cut with NarI to completion. The vector ends were dephosphorylated with HKTM phosphatase to prevent religation of the vector. The size-fractioning genomic DNA was ligated to cut and dephosphorylated pRR-6.

25 Molecular Complementation of Dsz- Mutant Strain CPE-648 Plasmid library ligations (above) were used to transform Dsz- mutant strain CPE-648 by electroporation as described. Negative control transformations of CPE-648, which did not contain DNA (mock transformations), were 30 also performed. After the four hour incubation in RM, the cells were spun out of suspension by centrifugation and the supernatant was removed. The cells were resuspended in BSM with no sulfur. These cells were used to inoculate 250 ml of BSM supplemented with 300 pl of a saturated -26ethanol solution of DBT. By this procedure, clones which are capable of complementing the Dsz- mutation will be selected by the sulfur bioavailability assay. Strains containing the complementing sequences the DNA of the present invention) will successfully remove the sulfur from DBT and grow preferentially.

After 6 days incubation at 30*C, the cultures were assayed for 2-HBP by HPLC. Accumulation of 2-HBP was detected in experimental cultures while no accumulation of 2-HBP was detected in control cultures. The culture producing 2-HBP was spread onto RM plates supplemented with chloramphenicol to obtain single colonies that were harboring plasmids. These plates were replica-plated to BSM agarose plates supplemented with 1.2 ml/liter of a saturated ethanol solution of DBT. After 24 hours incubation at 30°C, 2-HBP could be detected around some individual colonies under short wave UV illumination.

These colonies presumably harbored plasmids which complemented the Dsz- mutant by restoring the former Dsz+ phenotype.

Characterization of Clones Complementing Dsz- Mutant CPE-648 Two independent plasmid libraries successfully complemented mutant CPE-648 to Dsz* as described above.

25 Plasmid DNA was isolated from single colonies which demon- S. strated 2-HBP production on BSM plus DBT plates (above) from cultures transformed with each of the two libraries.

*This plasmid DNA was used to transform E. coli strain JM109. Plasmid DNA was isolated and cut with restriction endonucleases in order to build a restriction map of the clones. Each of the two libraries yielded a single complementing clone. By restriction pattern similarities, the two clones appear to have overlapping sequences.

These clones have been designated pTOXI-1 (Figure 4) and 35 pTOXI-2, respectively. pTOXI-1 contains an insert of

S

-27approximately 6.6kb. pTOXI-2 contains an insert of approximately 16.8kb.

Complementation of Dsz" Mutant GPE-362 Dsz- mutant GPE-362 was transformed with plasmids pTOXI-1 and pTOXI-2. As a control, GPE-362 was also transformed with vector pRR-6. Transformants containing plasmid DNA were selected on RM plus chloramphenicol plates. cmR colonies were transferred to BSM agarose plates supplemented with DBT. After 24 hr. incubation at 30°C, 2-HBP production could be seen around colonies containing either pTOXI-1 or pTOXI-2 by short wave UV illumination. No 2-HBP could be detected around colonies containing only vector pRR-6.

Overexpression of the Dsz Trait Uon Reintroducton of Cloned

DNA

Plasmids pTOXI-1 and pTOXI-2 were transformed into Dsz- mutant strain CPE-648. Transformants containing plasmid DNA were selected on RM plus chloramphenicol plates. The specific activity of individual clones was examined by the following protocol.

Single colonies of CPE-648 containing either pTOXI-l r pTOXI-2 were used to inoculate 25 ml RM plus 25 pg/ml chloramphenicol in a 250 ml flask. As a ositive control parent strain IGTS8 was also grown in 25 ml RM. After 48 S 25 hours of growth at 30'C, 225 rpm shaking, 2.5 ml of the cultures were crossed into 25 ml BSM supplemented with 0.7 mM DMSO. Cultures were incubated for an additional hours at 30*C. The optical density of each culture was measured at 600 nm against an appropriate blank. DBT was added to a final concentration of 150 pM and the cultures were incubated for 3 hours at 300C. An equal volume of Reagent Alcohol (Baxter, McGaw Park, IL) was then added to each culture to solubilize any remaining DBT or 2-HBP.

A

1 ml sample was removed and cellular debris removed by -28centrifugation. The supernatant was analyzed for 2-HBP by the HPLC assay described above. The specific activity is calculated as mg of 2-HBP per liter/hours of incubation/OD 600 The results of the above assay is listed in Table 1.

TABLE 1: Biocatalytic Desulfurization Activity of Transformed Mutants STRAIN

OD

600 2-HBP Specific Activity (mg/l) (mg/l/hr/OD 6 0 0 IGTS8 2.89 3.94 0.45 GPE-362 1.53 0.00 0.00 GP-362 1 3 CPE-648 4.10 0.00 0.00 CPE648 (pTOXI-1) 3.84 15.84 1.37 CPE648 (pTOXI-2) 2.88 5.74 0.66 EXAMPLE 2: DNA SEQUENCING OF A DESULFURIZATION

ACTIVE

BIOCATALYST BY THE DIDEOXY METHOD FROM PLASMID PTOXI-1 Materials Bacterial strains and plasmids Plasmid pTOXI- 1 was used as the original source of DNA for sequencing. Escherichia coli strain JM109 was used as a host for subcloning and plasmid maintenance.

Plasmids pUC18 and pUC19 were purchased from Bethesda Research Laboratories (Bethesda,

MD).

.Enzymes and Reaents Restriction endonucleases were purchased from Bethesda Research Laboratories (BRL) and New England Biolabs (Beverly, MA). T4 ligase was purchased from BRL. A -29- Sequenase Version 2.0 DNA sequencing kit was purchased from United States Biochemical Corporation (Cleveland, OH). All enzymes and kits were used in accordance with manufacturer's recommendations.

Growth Media and conditions E. coli strain JM109 harboring plasmids was grown in L-broth (Difco) containing 100 g/ml ampicillin. Transformants were selected on L-plates supplemented with agar and containing 100 pg/ml ampicillin. E. coli strains were grown at 37°C.

Methods Plasmid DNA preparation from E. coli Plasmid DNA was prepared from E. coli via lysis by SDS (Maniatis, et The DNA was further purified through a polyethylene glycol precipitation before use in sequencing reactions.

Plasmid Subcloninq The following subclones of pTOXI-1 were generated by standard techniques to aid in DNA sequencing: 20 a) pMELV-1 (Figure 5) was derived by isolating the 6.7kb HinddIII/NdeI fragment from pTOXI-1 (shown in Figure 4) and ligating it to pUC-18 cut with HindIII/NdeI. JM109 cells harboring pMELV-1 were identified by plasmid isolation and restriction endonuclease analysis (Maniatis, et al. b) pSMELV-1A (Figure 6) contains the 1.6kb SphI/XhoI fragment of pMELV-1 subcloned into pUC-18.

c) pSMELV-2A (Figure 6) contains the 0.7kb BamHI/SacI fragment of pMELV-1 subcloned into pUC-18.

d) pSMELV-3A (Figure 6) contains the 3.5kb SacI/XhoI fragment of pMELV-1 subcloned into pUC-18.

e) pSMELV-4A (Figure 6) contains the 1.5kb SphI/BamHI fragment of pMELV-1 subcloned into pUC-18.

Dideoxy Sequencing from Plasmid DNA a) Denaturation. Prior to sequencing reactions, plasmid DNA must be denatured. This was accomplished by treatment with NaOH. The denatured DNA is then recovered by addition of salt and EtOH precipitation. Preferably, Vg of denatured plasmid DNA is used in each sequencing reaction. See manufacturer's recommendations with Sequenase Version 2.0 DNA sequencing kit (United States Biochemical Corporation).

b) Dideoxy sequencing. Chain termination dideoxy sequencing with Sequenase 2.0 was performed as described by the manufacturer Biochemical Corporation).

Sequencing of the cluster was initiated by priming subclones pMELV-1A, pMELV-2A, pMELV-3A, pMELV-4A with the 40 Universal Primer" defined as: 5'-GTTTTCCCAGTCACGAC-3' and the "Reverse Primer" defined as: 5'-AACAGCTATGACCATG-3'. The sequence was extended by synthesizing overlapping oligonucleotides to previously read sequence using the Gene Assembler Plus (Pharmacia, Piscataway, NJ). The synthesized oiigonucleotides were used as primers for continuing sequence reactions. Plasmid pMELV-1 was used as the template for all of the remaining sequences. DNA sequence was read from both strands of the plasmid clone to increase fidelity.

25 EXAMPLE 3: COMPLEMENTATION CLONING OF A DESULFURIZATION ACTIVE BIOCATALYST FROM A COSMID LIBRARY; o TRANSFECTION OF BIOCATALYST DNA INTO AN R. FASCIANS HOST ORGANISM Materials and Methods Bacterial strains, media and reagents Rhodococcus sp. Rhodococcus rhodochrous strain IGTS8, .obtained from the Institute of Gas Technology (Chicago, -31- IL) was used. UV1 is a mutant of IGTS8 that is unable to desulfurize DBT, described herein. R. fascians D188-5 (Desomer, et al., J. Bacteriol., 170:2401-2405, 1988) and R. rhodochrous ATCC13808 (type strain from ATCC) do not metabolize DBT. E. coli XL1-Blue (from Stratagene Cloning System, La Jolla, CA) is recAl lac thi endAl cvrA96 hsdR17 supE44 relAl proAB lacId lacZAM15 TnlO]. E. coli CS109 is W1485 thi supE E. coli S17-1 is a derivative of E. coli 294 and is recA thi pro hsdR- res- mod± [RP4-2- Tc::Mu-Km::TnZ] (Simon, et al., Plasmid vectors for the genetic analysis and manipulation of rhizobia and other gram-negative bacteria, p. 640-659. In A. Weissbach, and H. Weissbach Methods in enzymology, vol 118, Academic Press, Inc., Orlando, 1986).

Pseudomonas minimal salts medium (PMS) was prepared according to Giurard and Snell (Biochemical factors in growth, p. 79-111. In P. Gerhardt, R. G. E. Murray, R. N.

Costilow, E. W. Nester, W. A. Wood, N. R. Krieg, and G. B.

Phillips Manual of methods for general bacteriology, American Society for Microbiology, Washington, DC., 1981) and contained 0.2% glycerol, 40 mM phosphate buffer (pH 2% Hutner's mineral base, and 0.1% (NH 4 2

SO

4 PMS medium lacking sulfate was prepared with chloride salts in place of sulfate salts. Luria broth (LB) was 1% 25 bactotryptone, 0.5% yeast extract, and 1% NaCl. All liquid medium incubations were performed with shaking in water baths (New Brunswick Scientific, Edison, NJ).

Ampicillin (50 pg/ml) and tetracycline (12.5 pg/ml) were included as selective agents when required. Dibenzothio- 30 phene (DBT) was purchased from Fluka Chemical Corporation of Ronkonkoma, NY. DBT-sulfoxide was from ICN Biochemicals of Irvine, CA, and DBT-sulfone was obtained from Aldrich Chemical Company of Milwaukee, WI. Agarose was obtained from BRL.

-32- Plasmid vectors (Keen, et al., Gene 70:191-197, 1988) and pRF29 (Desomer, et al., 1988) served as sources of the Rhodococcus plasmid origin of replication.

Cosmid library construction High molecular weight DNA was isolated from IGTS8 by the method of Consevage et al, Bacteriol., 162:138- 146, 1985), except that cell lysis was accomplished in TE mM Tris-HC1, 1 mM EDTA, pH 8.0) containing lysozyme mg/ml) and SDS The DNA was partially digested with Sau3AI and fragments of approximately 20 kb were isolated after centrifugation through a sodium chloride gradient (Frischauf, et al., Digestion of DNA: size fractionation, p. 183-189. In S. L. Berge, and A. R. Kimmel (eds.), Methods in Enzymology, vol 152, Academic Press, Inc, San Diego, CA, 1987). These fragments were ligated into the BamHI site of pLAFR5 using standard procedures. In vitro packaging was performed using Gigapack Plus (Stratagene).

Packaged cosmids were transduced into E. coli S17-1.

DBT spray plate assay A spray plate assay for the identification of bacteria capable of modifying dibenzothiophene (DBT) was originally described by Kiyohara et al, (Appl. Environ.

Microbiol., 43:454-457, 1982) and modified by Krawiec 25 (Bacterial desulfurization of thiophenes: screening techniques and some speculations regarding the biochemical and genetic bases, p. 103-114. In G. E. Pierce Developments in Industrial Microbiology, vol 31, Society for Industrial Microbiology, Columbus, Ohio, 1990). The assay 30 was further modified for use with R. rhodochrous IGTS8 as follows. Cells from individual IGTS8 colonies were transferred to LB plates as small (0.5 cm) patches and were incubated at 30'C for 24 to 36 h. Large amounts of cells from these patches were transferred onto PMS-1% agarose l -33plates that lacked a source of sulfur. These plates were immediately sprayed with a 0.1% DBT solution in ethyl ether. The PMS-DBT plates were incubated at 30*C for a minimum of 18 hours and fluorescent products around the patches were detected by viewing under short-wave (254 nm) UV illumination.

Sulfur bioavailabilitv assay IGTS8 was incubated in PMS medium at 30'C for 24 to 48 h, the cells were pelleted by centrifugation, followed by two washes with sulfur-free PMS. Washed cells were inoculated into PMS that contained, as a sole source of sulfur, a 0.2% concentration of one of the following:

DBT,

DBT-sulfoxide, or DBT-sulfone. The inoculum was adjusted so that the beginning absorbance at 600 nm (A 600 was 0.02.

The culture was incubated at 30'C and growth was monitored at A 600 For cultures incubated with DBT, the supernatant was viewed at various intervals under short wave UV light to check for production of fluorescent products.

Plasmid isolation and hybridizations Cosmid DNA (pLAFR5) was isolated from E. coli as described by Ish-Horowicz and Burke (Nucl. Acids Res., 9:2989-2998, 1981), and from Rhodococcus species as described by Singer and Finnerty Bacteriol., 170:638-645, 1988). Large scale cosmid preparations were carried out 25 according to Birnboim and Doly (Nucl. Acids Res., 2:1413- 1423, 1979). DNA hybridization experiments were performed according to Southern Molec. Biol., 98:503-517, 1975).

DNA was labelled with 32 P-dCTP (Amersham), using the random primer method of Feinberg and Vogelstein (Anal. Bio- 30 chem., 137:266-267,.1984).

UV mutaqenesis of IGTS8- IGTS8 was incubated overnight in LB at 30°C and approximately 3000 colony forming units were spread onto -34fresh LB plates. These plates were immediately exposed to short wave UV light (254 nm) for 5 to 20 s at a distance of 3.5 cm. Plates were incubated at 30*C for 48 h or until colonies developed. Colonies from plates exhibiting >50% cell death were assayed for their ability to metabolize or desulfurize DBT, using the spray plate assay.

Electrotransformation of Rhodococcus R. rhodochrous IGTS8 and the UV1 mutant were transformed with plasmid DNA via electroporation (Gene Pulser, Biorad Laboratories, Inc, Hercules, CA). The bacteria were grown in LB for 24 to 48 h at 30*C, diluted to an A600 of 0.15 with fresh LB, and incubated at 30'C for an additional 4 h. Cells were collected by centrifugation and washed four to five times with 0.3 M sucrose and finally resuspended to -5 x 109 cells/ml in 0.5 M sucrose. To an ice cold 0.2 cm electroporation cuvette (Biorad), was added 40 pi of this bacterial solution. The cells were pulsed at 25 pF and 2.5 kV with the Pulse Controller at 800 ohms and were immediately diluted with 1 ml of LB containing 0.5 M sucrose. The cells were incubated at for 1 h, plated on LB agar plates plus appropriate antibiotics, and incubated at 30*C until colonies developed. When the plasmid carried the pRF29 Rhodococcus plasmid origin of. replication, colonies were visible after S. 25 48 h. In the absence of the pRF29 origin, colonies appeared after 4 to 5 days.

R. fascians D188-5 was transformed by electroporation in a similar manner but, due to its slower growth rate, it was incubated in LB overnight until it reached an A 600 of The cells were washed and resuspended in distilled Swater instead of sucrose. The Pulse Controller was set at S 400 ohms and the recovery period after electroporation was in LB for 4 h before plating onto selective media. Successful transformation of R. fascians D188-5 with E. coli plasmids required that the DNA be methylated in vitro beforehand, using the CpG methylase, SssI (New England Biolabs, Inc., Beverly, MA).

Gas chromatography and mass spectroscopy Cells were incubated overnight in LB medium at 300C and 100 il was used to inoculate 50 ml of PMS minimal medium. The culture was incubated at 30'C for 4 days, washed twice with sulfur-free PMS and the pelleted cells were inoculated into 50 ml of PMS that contained 0.1% DBT as the sole source of sulfur. These cells were incubated at 30'C for 24 h and the supernatant was stored frozen at For assays involving R. fascians D188-5, incubation times were increased 2 to 3-fold.

Sample preparation and chemical analyses were performed as described (Olson, et al., Energy Fuels, submitted, 1993). Briefly, each sample supernatant (-50 ml) was thawed and residual insoluble material was removed by centrifugation. The cleared supernatant was acidified with HC1 to pH 1.0 and then extracted three times with ml of ethyl acetate. Insoluble material from the centrifugation step was also extracted with ethyl acetate. The ethyl acetate extracts were combined, dried over anhydrous calcium chloride, filtered, and ethyl acetate was removed by rotary evaporation. A known amount of internal standard (octadecane in chloroform solution) was added to the 25 sample, which was then analyzed by GC/FID (gas chromatography/flame ionization detection) and GC/FTIR/MS (gas chromatography/Fourier transform infrared/mass spectrometry). In some samples, the acidic components in the ethyl acetate extract or in the post-extraction aqueous layer were methylated by treating with an ether solution of diazomethane.

The analyses were performed on a serially interfaced GC/FTIR/MS system as previously described (Diehl, et al., i Spectros. Int. J .,8:43-72, 1990, Olson and Diehl, Anal.

35 Chem., 59:443-448, 1987). This system consisted of the -36- Finnegan ion trap (ITD 800) operated with the AGC on and the Nicolet 20SXB Fourier transform infrared spectrometer.

Gas chromatography was conducted with a 30 m x 0.32 mm column (1.0 pm phase thickness) with a 2.0 ml/min helium carrier flow rate measured at 330'C. On-column injections were utilized for sample introduction because the sulfoxides and sulfones are thermally unstable and they decompose in split or splitless injectors (Vignier, et al., J.

High Resol. Chromatoqr. Chromator. Commun., 6:661-665, 1983). The oven temperature program was as follows: injection, followed by increases in temperature at rates of 20°C/min to 80°C, 5'C/min to 200*C, 10'C/min to 330°C, and hold for 5 min. GC/FID analyses were performed with a HP 5880A with a similar column and program for flow rate and oven temperature.

Results Isolation of a Dsz- Mutant of R. rhodochrous IGTS8 When cloning from a foreign bacterial genus into E.

coli, not all genes are expressed nor are all protein products active. To assure that cloned desulfurization genes would be expressed in the host cell, a mutant of R.

rhodochrous IGTS8 that could no longer desulfurize DBT was isolated. Using this mutant as a cloning recipient would insure that the cellular environment was appropriate for 25 gene expression and protein function, thereby allowing screening for cloned desulfurization genes by complementation.

R. rhodochrous IGTS8 was mutagenized by exposure to UV light, and 1000 survivors were screened for the ability to produce a UV fluorescent product in the DBT spray plate assay. Three potential desulfurization negative mutants were identified and then re-evaluated in the sulfur bioavailability assay. Two mutants (designated UV1 and UV23) could not use DBT or DBT-sulfone as sole sources of sulfur 35 and thus appeared to be Dsz-. When'grown in the presence -37of DBT, mutant UV1 could not metabolize DBT to 2-HBP or to any other potential intermediate, as measured by GC/MS analysis. Therefore, strain UV1 was considered to be Dszand was used as the host for complementation studies to identify clones that carried desulfurization genes.

Cosmid cloning of desulfurization genes DNA from Dsz+ source organism IGTS8 was used to construct a library in the cosmid vector, pLAFR5. This library was transduced into E. coli S17-1 and plasmids were isolated from approximately 25,000 colonies. These cosmids were electroporated into R. rhodochrous UV1, a Dsz- mutant of IGTS8, with an efficiency of -300 transformants/pg DNA. Various numbers of UV1 transformants were pooled and incubated for 18 hours at 30'C, after which the cells were washed twice and resuspended in sulfate-free PMS. Approximately 7 x 108 pooled cells were inoculated into 100 ml of PMS with DBT as the sole source of sulfur. A predicted product.of the DBT desulfurization reaction is 2-HBP, which is fluorescent when exposed to UV light. Therefore, batch cultures were grown at 30°C and the supernatants were observed for fluorescence. Approximately 3300 UV1 transformants were screened in four sepa- S: rate batches. In one batch (representing -600 transformants) a UV fluorescent product appeared in the supernatant 25 after five days' incubation. Individual colonies were isolated and twelve of these continued to produce a fluorescent product when exposed to DBT.

Attempts to recover cosmid DNA from these isolates failed, so Southern hybridizations were performed to determine if the cosmids had become integrated into the chromosome of strain UV1. Chromosomal DNA was isolated from seven transformants and digested with EcoRI. After agarose electrophoresis and blotting, the fragments were hybridized with 32 P-labelled probes derived from 35 In all transformants tested, pLAFR5 probes hybridized to a

I

-38- DNA fragment -20 kb in size. Vector derived probes did not hybridize to the control IGTS8 genome. Therefore, the desulfurization positive cosmid clones had apparently integrated into the chromosome of strain UV1.

Since the plasmids had integrated into the chromosome, the genomic DNA connected to either side of the plasmid cloning site must represent R. rhodochrous IGTS8 sequences that were able to complement the Dsz- mutation in strain UV1. (This would be true regardless of whether the mode of integration was by homologous or illegitimate recombination.) Sequences were recovered that flanked the inserted plasmid from three desulfurization positive transformants by digesting genomic preparations with EcoRI or BamHI. These enzymes cut pLAFR5 once in the polylinker region so that an intact sequence of pLAFR5 could be recovered, linked to a neighboring chromosomal fragment from IGTS8. The digested DNA was ligated to itself (at a concentration of -20 ng/pl) and was transformed into E.

coli S17-1. Sixteen tetracycline resistant colonies were obtained, seven from the BamHI digestion and nine from the EcoRI digestion. Restriction enzyme analysis revealed that all the EcoRI-rescued clones contained a 2.1 kb fragment of IGTS8 DNA. The BamHI-rescued clones contained :a 1.65 kb fragment from IGTS8.

25 The 2.1 kb IGTS8 DNA from the EcoRI rescue experiment was used as a template to make labelled DNA probes, which were hybridized to colony lifts of the original, intact cosmid library in E. coli. Of 5000 colonies, 17 hybridized with the IGTS8 probes. Cosmid DNA was isolated from 30 each clone and transformed into strain UV1. Three of the seventeen DNA preparations complemented the Dsz- phenotype.

A restriction map for this region was constructed, using EcoRI and HindIII. Probes from the 2.1 kb IGTS8

DNA

hybridized to the 4.5 kb EcoRI fragment. All cosmid clones that conferred the Dsz+ phenotype contained the -39entire 4.5 kb EcoRI fragment and portions of the 4.5 kb EcoRI-HindIII and 18 kb EcoRI fragments. These results indicated that the desulfurization genes lay within a kb region.

Subcloninq the desulfurization genes The 4.5 kb EcoRI and the 4.5 kb EcoRI-HindIII fragments were subcloned into pLAFR5, but neither fragment complemented the Dsz- mutation of strain UVl. The 9.0 kb EcoRI fragment from GE1-H, the 15.0 kb EcoRI-HindIII fragment from GE1-C, and the 18 kb EcoRI fragment from GE1-K were subcloned into pLAFR5 to yield the plasmids pSAD60-28, pSAD48-12, and pSAD56-6, respectively. When transformed into UV1, all three produced UV fluorescent products from DBT in the spray plate assay, consistent with the localization of the Dsz+ phenotype as determined by restriction mapping. Construction of additional subclones from this region narrowed the location of the relevant genes to a 6.5 kb BstBI fragment.

Nature of the mutation in strain UVl Genomic blots of EcoRI digested IGTS8 and UV1 DNA were hybridized with probes produced from the 2.1 kb EcoRI-rescued fragment of IGTS8. No hybridization was detected to UVl DNA, indicating that the UV1 mutation is a large deletion and not a simple point mutation.

A Rhodococcus plasmid origin of replication increases transformation of UV1 Electroporation of-UVl with pSAD48-12 typically resulted in a low transformation efficiency (-550/pg

DNA)

and only about 50% of. the transformants exhibited the Dsz+ phenotype (presumably because DNA had been lost or rearranged during recombination with the chromosome). To improve the transformation efficiency, a 4.5 kb HindIII fragment from pRF29 was cloned into the HindIII site of pSAD48-12, resulting in pSAD74-12. This 4.5 kb fragment contains a Rhodococcus plasmid origin of replication, which allowed pSAD74-12 to replicate as a plasmid in strain UV1. This clone transformed UV1 with an efficiency of greater than 104 transformants/pg DNA. Nearly 100% of these transformants exhibited the Dsz+ phenotype. Unfortunately, the yield of plasmid prepared directly from UV1 was so poor that DNA from minipreparations could not be visualized on agarose gels. However, plasmid isolated from UV1 could be used to transform E. coli S17-1, from which large amounts of the plasmid were prepared.

The Dsz+ phenotype is not expressed in E. coli S17-1 E. coli S17-1 was transformed with pSAD48-12 and desulfurization activity was measured with the spray plate assay. No positive colonies were identified. It was possible that the E. coli polymerase could not recognize the IGTS8 promoter(s) in pSAD48-12, so the IGTS8 DNA was placed under control of the E. coli lac promoter. The kb EcoRI-HindIII IGTS8 fragment from pSAD48-12 was subcloned into the pBluescript vectors, SK and KS', so that the IGTS8 fragment was cloned in both orientations with respect to the lac promoter. Neither clone expressed the Dsz+ phenotype in E. coli XL1-Blue. It is not yet known whether this stems from poor transcription or translation 25 of the cloned genes or whether the overproduced proteins are inactive in E. coli S17-1.

The Dsz+ gene orr enes are expressed in R. fascians Since the cloned genes were either not expressed or produced inactive proteins in E. coli, efforts were initi- S: 30 ated to express the genes in other Rhodococcus species.

R. fascians D188-5 exhibited no desulfurization in the DBT spray plate assay or in the sulfur bioavailability assay.

Initial attempts to transform R. fascians with the desulurization positive plasmid, pSAD74-12 were unsuccessful. furization positive plasmid, pSAD74-12 were unsuccessful.

-41- Other Rhodococcus species are known to have endogenous restriction systems that cleave DNA at Sall-like restriction sites. Since pSAD74-12 contained multiple Sail recognition sequences, CpG methylase, SssI, was used to methylate pSAD74-12 in vitro. With methylated pSAD74-12 DNA, transformants of R. fascians D188-5 were obtained with an efficiency of about 7 x 103 transformants/Vg

DNA.

These transformants displayed the Dsz+ phenotype in the spray plate assay and GC analysis of liquid medium supernatant revealed the formation of 2-HBP from DBT.

Efforts to transform pSAD74-12 into a second species, R. rhodochrous ATCC13808 were ineffective, despite the use of unmethylated or CpG-methylated plasmid. It is possible that the electroporation conditions for ATCC13808 were not optimal, though a wide range of conditions was tested. It seems more likely that ATCC13808 has a restriction system that is not inhibited by CpG methylation.

2-HBP is the major desulfurization product The predominant metabolite produced from DBT by R.

rhodochrous IGTS8 is 2-HBP, with small amounts of 2'hydroxybiphenyl-2-sulfinic acid (DBT-sultine) and 2'hydroxybiphenyl-2-sulfonic acid (DBT-sultone) also identified by GC/MS analysis (Olson, et al., Energy Fuels in 'press, 1993). These products were also produced by IGTS8 25 in this work (Table Neither R. fascians D188-5 nor R.

rhodochrous Dsz- mutant UV1 produced these products from DBT. However, when R. fascians D188-5 was transformed with plasmid pSAD74-12 and when the R. rhodochrous UVi mutant was transformed with plasmid pSAD104-10, these bacteria produced products from DBT that were identical to those identified for R. rhodochrous IGTS8 (Table In particular, 2-HBP was produced in large quantities, indicating that carbon-sulfur bond specific desulfurization of was mediated by products of genes cloned from IGTS8.

-42- One subclone, pSAD90-11, carried a DNA fragment that was supposedly identical to that cloned into pSAD104-10, but the two plasmids differed in the results they produced when introduced into R. rhodococcus UV1. In the plate assay, the surface film of DBT disappeared from the vicinity of colonies that contained pSAD104-10, producing a clear zone, and a fluorescent halo appeared around those colonies. On the other hand, when cells contained 11, no fluorescent products were produced but a zone of DBT clearing did form around each colony. GC/MS analysis showed that no 2-HBP was produced by cells containing pSAD90-11, but that a significant amount of DBT-sultone did accumulate (Table The sultone does not accumulate in the parent strain, UV1 (data not shown). These observations imply that when the 9.0 kb EcoRI fragment was subcloned into pSAD90-11 the DNA was damaged so as to inactivate the gene(s) encoding .the enzyme(s) that convert the sultone to 2-HBP. This suggests that at least two enzymes are involved in desulfurization and that the sultone may be an intermediate in the pathway. This result is consistent with the kinds of metabolites detected in the original isolate, R. rhodochrous IGTS8 (Olson, et al., 1993).

*oo *.e

S

S S S S S S 55 5 55 5* 55 4- Table 2 M~etabolites produced from DBT by Rhodococcus species transformed with subclones derived from R. rhodochrous IGTS8.

Metabol ltea Rhodococcus species (plasmid) R. rhodochrous Uvi Uvi R. fascians IGTS8 (pSAD104-lo)b (pSAD9O-ll) c D188-5 (pSAD- DBT +e DBTO 0 0 0 DBT0 2 0 0 0 0 DBT-sultone DBT-sultine 0 or trace 0 trace 2-HBP 0..

a Products are: DBT, dibenzothiophene; DBTO, dibenzothiophene 5-oxide (sulfoxide);

DBTO

2 dibenzothiophene (sulfone); DBT-sultone, 2 '-hydroxybi- -44phenyl-2-sulfonic acid (detected as dibenz[c,e][1,2]oxathiin 6,6-dioxide); DBT-sultine, 2'-hydroxybiphenyl-2-sulfinic acid (detected as dibenz[c,e][1,2]oxathiin 6-oxide); dibenzothiophene sulfone; 2-HBP, 2hydroxybiphenyl (Krawiec, pg. 103-114. In G. E. Pierce Developments in Industrial Microbiology, vol 31, Society for Industrial Microbiology, Columbus, Ohio, 1990).

b 9.0 kb EcoRI DNA fragment from IGTS8 subcloned into pLAFR5, plus the origin of replication from pRF29.

c Mutated 9.0 kb EcoRI DNA fragment from IGTS8 subcloned into pLAFR5, plus the origin of replication from pRF29.

d 15.0 kb EcoRI-HindIII DNA fragment from IGTS8 sub- 15 cloned into pLAFR5, plus the origin of replication from pRF29.

e Presence of metabolites is reported in relative amounts from very large amounts to very small trace amounts.

S.

S.

20 IGTS8 cannot use DBT-sulfoxide as a sulfur source *5 R. rhodochrous IGTS8 was incubated in minimal medium with one of the following as the sole source of sulfur: DBT, DBT-sulfoxide, or DBT-sulfone. IGTS8 was incapable of utilizing the sulfur supplied by DBT-sulfoxide but grew well in the presence of DBT or DBT-sulfone. DBT-sulfoxide was not toxic to cells when grown in a rich medium

(LB).

Therefore, either IGTS8 cannot transport or otherwise act on DBT-sulfoxide, or else DBT-sulfoxide is not a true intermediate of the desulfurization pathway.

EXAMPLE 4: DNA SEQUENCING OF A 9763 NUCLEOTIDE

ECORI-

SAU3AI FRAGMENT CONTAINING THE GENE OR GENES FOR THE DESULFURIZATION BIOCATALYST OF IGTS8

BY

THE METHOD OF SANGER ET AL.

A 9763 nucleotide EcoRI-Sau3AI fragment containing the gene or genes responsible for the Dsz+ phenotype was isolated from the IGTS8 source organism. The DNA sequence of this fragment was determined from both strands of DNA using the dideoxy chain-termination method of Sanger et al.

(1977),

D

NA sequencinq with chain-termination inhibitors 74 PROC. NATL. ACAD. SCI. USA 5463-5467, a modified T7 DNA polymerase (USB) and [a- 35 S]-dCTP (Amersham). Deletion 20 clones for DNA sequencing were constructed in pBluescript 6. (Stratagene) using exonuclease III and the methods of b@ Henikoff (1984), Unidirectional diestion with exonuclease III creates tareted breakoints for DNA seuencing 28 GENE 351-359.

Sequences from 141 individual deletion clones were used to reconstruct the entire 9763 nucleotide fragment.

.4 -46- Computerized sequence assembly was performed using DNA InspectorII (Textco, Hanover, NH). The DNA sequence was determined independently for each strand of DNA, but the entire 9763 nucleotide fragment was not completely sequenced on both strands. The sequence determined from one strand of DNA covered 95% of the 9763 nucleotide sequence.

On the other DNA strand, 96% of the sequence was determined. The sequence was determined from at least two independent deletion clones for the entire 9763 nucleotide fragment.

EXAMPLE 5: FURTHER RESOLUTION OF THE SEQUENCE OF PTOXI-1 AND OPEN READING FRAMES (ORFS) ENCODED THEREIN; DSZ+ PROMOTER ENGINEERING; EXPRESSION OF THE DSZ+ PHENOTYPE. IN A HETEROLOGOUS HOST ORGANISM; MAXICELL ANALYSIS OF DESULFURIZATION

GENE

EXPRESSION PRODUCTS Organization of the desulfurization cluster Sequencing of pTOXI-1, the results of which are set forth below in the Sequence Listing, predicted three nearly contiguous open reading frames (ORFs) on one strand of the clone (Figure The sizes of each ORF are predicated as 1359 bases (bps 786-2144) for ORF 1, 1095 bases (bps 2144- 3238) for ORF 2 and 1251 bases (bps 3252-4502) for ORF 3.

Subclone analysis described below has revealed that ORFs 1, S 25 2 and 3 are required for the conversion of DBT to 2-HBP and that all of the genes encoded by these ORFs are transcribed -47on a single transcript as an operon. All subclones described below are maintained in E. coli Rhodococcus shuttle vector pRR-6. Activity of each subclone was determined by growing transformants of Dsz- strain CPE-648 in a rich media (RM) for 48 hours. 1 ml of the culture was used to inoculate 25 ml BSM supplemented with greater than 100 pM DBT or DBT-sulfone. Cultures were assayed for desulfurization products after 48 120 hours. A diagram of each of the subcloned fragments is shown in Figure 8.

In subsequent studies, the subclones were grown in rich media with chloramphenicol, then crossed into BSM supplemented with 100 pM of either DBT or DBT-sulfone.

Cultures were shaken at 30oC for 2-5 days and assayed for desulfurization products by HPLC.

A. pENOK-1: A subclone was constructed which contains the 4.0 kb SIhl fragment of pTOXI-1. This fragment spans e ORFs 1 and 2 but truncates ORF 3. Analysis of pENOK-1 containing transformants revealed the production of no 0.

products when incubated with DBT. However these trans- 20 formants were capable of producing 2-HBP from DBT-sulfone.

B. pENOK-2: A suclone which contains the 3.6 kb SacI .fragment of pTOXI-1 was constructed. This fragment contains ORFs 2 and 3 but truncates ORF 1. Analysis of pENOK- 2 transformants revealed no production of any desulfurie 25 zation products from either DBT or DBT-sulfone. The lack S-

I

-48of any activity detectable from either ORFs 2 or 3 suggests that the ORFs are arranged as an operon with transcription mediated from a single upstream promoter. Presumable, this promoter has been removed in this subclone.

C. pENOK-3: A 1.1 kb XhoI deletion mutation of pTOXI-1 was constructed. Both ORFs 1 and 2 are truncated. ORF 3 remains intact. Transformants harboring pENOK-3 show production of DBT-sulfone from DBT. No production of 2-HBP is detected from either DBT or DBT-sulfone. It should also be noted that at the nucleotide level, a deletion of this type would not result in a polar mutation. The sequence predicts an in-frame splicing of-ORFs 1 and 2 which would produce a hybrid protein that is presumably inactive.

However, by avoiding stop codons, the putative single mRNA transcript remains protected by ribosomes allowing for translation of ORF 3. The ability of the ORF-3 product to produce DBT-sulfone from DBT demonstrates that DBT-sulfone is a true intermediate in the carbon-sulfur bond specific biocatalytic desulfurization pathway of IGTS8.

20 D. pENOK-11: The 3.4 kb NcoI fragment from pTOXI-1 was subcloned into a unique NcoI site of pRR-6. This fragment contains all of ORFs 2 and 3 but truncates the 5' end of ORFl. Transformants with pENOK-11 demonstrated no desulfurizing-specific enzymatic activity towards DBT or DBT- 25 sulfone. This indicates essential coding regions bordering -49this fragment. This is consistent with the predication that the entire cluster is expressed on a single transcript as discussed for subclone pENOK-2. Again, the promoter for gene transcription is not present in this subclone. Subclone pENOK-13 (below) corroborates this prediction.

E. pENOK-13: A subclone of pTOXI-1 was constructed which had a 2.6 kb SPhI-XhoI deletion. This subclone only contains an intact ORF 3. ORF 1 is lost completely and ORF 2 is truncated. This subclone showed no desulfurizing-specific enzymatic activity towards DBT or DBT-sulfone. This result should be compared with the phenotype of pENOK-3 which demonstrated production of DBT-sulfone from DBT.

Because pENOK-13 differs from pENOK-3 by the additional deletion of the smaller SphI/XhoI fragment, this would indicate an element in the 1.6 kb SphI/XhoI fragment which is essential for gene expression. Because sequencing has revealed no significant ORF's contained in this region, it is postulated that a promoter element may be present in this region.

20 F. pENOK-16: A subclone of pTOXI-1 was designed which eliminates nearly all unnecessary sequences from the desulfurization cluster. This construct contains the 4 kb BstBI-SnaBI which presumably contains all essential sequence for complete desulfurization in that in contains all S25 of ORFs 1, 2 and 3 as well as 234 bases of upstream se- 25 of ORFs 1, 2 and 3 as well as 234 bases of upstream sequence. The 3' SnaBI site lies 80 base pairs beyond the termination of ORF 3. CPE-648 harboring this plasmid was capable of converting DBT and DBT-sulfone to 2-HBP. pENOK- 16 thus represents the smallest amount of the cluster yet observed which demonstrates the complete desulfurization phenotype.

G. pENOK-18: This subclone contains a NsiI-BfaI fragment of pTOXI-1. The NsiI site is 23 bp downstream of the predicted start site of ORF 1. CPE-648 harboring this subclone lacks desulfurization activity on both DBT and DBT-sulfone. This subclone most likely eliminates the promoter region and truncates the first structural gene.

H. pENOK-Nsi: To help further elucidate the start site of ORF 1, a subclone was made in which a 4 bp deletion is introduced at the unique Nsil site which is 23 bp downstream of the predicted start site of ORFl. The mutation was generated by cutting with NsiI and blunting the ends with T4 DNA Polymerase. If the NsiI site is within the first structural gene this frameshift mutation would cause 20 an early stop signal in ORF 1. Transformants of pENOK-Nsi were capable of producing DBT-sulfone from DBT. However, no production of 2-HBP was detected indicating that the mutation had disrupted an essential structural gene.

In subsequent studies, due to the clear expression of 25 the ORF-3 encoded oxidase, in this clone, it was considered -51likely that the ORF-2 product would also be expressed.

Accordingly, ORF-2 alone is incapable of further metabolism of DBT-sulfone.

I. pENOK-19: A subclone of pTOXI-1 was constructed which contains a deletion from the NotI site, which is in the earlier part of ORF 2, to the SnaBI which is after ORF 3.

This subclone should demonstrate the activity of ORF 1 alone. CPE648 transformants harboring this subclone displayed no enzymatic activity towards DBT or DBT-sulfone.

The results of pENOK-Nsi and pENOK-19, taken together, suggest that the ORF-I and ORF-2 products must be simultaneously expressed in order to further metabolize DBT-sulfone.

J. pENOK-20: In order to evaluate the function of ORFs 2 15 and 3 separately from ORF 1, DNA spanning ORFs 2 and 3 was S. amplified by the Polymerase Chain Reaction (PCR). Primers RAP-1 (5'-GCGAATTCCGCACCGAGTACC-3', bps 2062-2082) and RAP- 2 (5'-ATCCATATGCGCACTACGAATCC-3' bps 4908-4886) were synthesized with the Applied Biosystems 392 DNA/RNA Synthes- 20 izer. Nucleotides in bold were altered from the template sequence in order to create restriction sites for subcloning; thus primer RAP- 1 contains an EcoRI site, and primer RAP-2 contains an NdeI site. Amplification was carried out with the GeneAmp Kit (Perkin Elmer Cetus) which utilizes -52the Taq polymerase and the Perkin Elmer Cetus 9600 Thermocycler. Parameters were as follows: Template: pMELV-1 Plasmid DNA 0.2 or 2.0 ng Primers: RAP-1 0.5 or 0.2 pM RAP-2 0.5 or 0.2 pM Cycles: IX@ 96°C 2 min 96'C 30 sec 52°C 30 sec 72*C 2 min Amplification yielded-the predicted 2846 bp fragment.

In order to express the amplified fragment harboring ORFs 2 and 3, it was ligated to the XbaI/EcoRI fragment of the chloramphenicol resistance gene promoter from Rhodococcus fascians (Desomer et al.: Molecular Microbiology (1992) 6 2377-2385) to give plasmid pOTTO-1. Ultimately, a blunt end ligation was used for the subcloning of the amplified product due to the fact that ligation using the engineered restriction sites was unsuccessful. This fusion S.was ligated to shuttle-vector pRR-6 to produce plasmid 20 pENOK-20. CPE648 transformants of pENOK-20 were grown in Sthe presence of DBT and 25 pg/ml chloramphenicol for promoter induction. All transformants converted DBT to DBTsulfone presumably through the activity of the ORF 3 as o demonstrated in subclone pENOK-3. The inability to further 25 process DBT-sulfone with the presence of ORF 2 suggests that the product of ORF 2 alone is incapable of using

DBT-

sulfone as a substrate. This is consistant with results -53obtained from pENOK-Nsi, and suggests that ORF-2 alone is incapable of using DBT-sulfone as a substrate.

Assignment of Gene Products of ORFs 1. 2 and 3 Based on the foregoing subclone analyses, functions have been tentatively assigned to each of the ORFs present within the pTOXI-1 sequence. ORF 3 can be identified as responsible for an oxidase capable of conversion of DBT to DBT-sulfone. Subclone pENOK-3 demonstrates this activity very clearly. ORFs 1 and 2 appear to be responsible for conversion of DBT-sulfone to 2-HBP. This aryl sulfatase activity is evidenced in subclone pENOK-1. However subclones pENOK-19 and pENOK-20 indicate that neither ORF 1 or ORF 2 alone is capable of any conversion of the intermediate DBT-sulfone. This suggests that the protein products of ORFs 1 and 2 work together to cleave both of the carbon-sulfur bonds. Presumably, this is achieved through a heterodimer arrangement of the proteins, or through a regulatory function of one protein on the other. The results of paralell investigations, presented in Example 3, suggested that ORF-1 encodes an enzyme that converts

DBT--

sulfone to DBT-sultone. Lengthy incubations of CPE-648 harboring pENOK-19 (intact native promoter and ORF-I) have shown neither the depletion of DBT-sulfone nor the production of any new products. This is contrary to indi- 25 cations derived from Example 3.

.ooooi -54- Alternative Promoter Screening Increasing the specific activity of desulfurization is a significant objective of the studies described herein.

One approach to accomplishing this goal is to replace the original promoter with one that can produce both higher and constitutive expression of the desulfurization gene cluster. Because there are so few reported and characterized Rhodococcus promoters, random genomic libraries have been prepared and screened for promoter activity in two systems.

In one, the reporter is the chloramphenicol resistance gene used in the above-discussed plasmid constructions. In the other, the desulfurization cluster itself is used as a reporter.

Promoter Screening Example A. Chloramphenicol Resistance Reporter.

As also described below, partially digested Rhodococcus genomic DNA has been cloned upstream of a promoterless.chloramphenicol resistance gene. The resulting libraries were then transformed into Rhodococcus which are subjected to chlorarnphenicol selection. Four apparent promoter elements were rescued by pRHODOPRO-2, although plasmid could be isolated from only one of these, possibly due to vector instability. The stable plasmid RP2-2A has been subjected to sequence analysis. Technical problems 25 have been observed with restriction enzyme treatment of the NarI cloning site used in these vectors. Unfortunately, the NarI enzyme demonstrates severe site-selectivity and does not appear to digest the vector well. New vectors have been constructed in order to alleviate this problem, although a lack of convenient and unique restriction sites slowed the progress of these studies. A recent observation on the Rhodococcus replication origin will aid in constructing a more effective promoter probe, as discussed below.

Recently, the 1.4kb BglIIfragment was removed from pRR-6, and the ends were blunted and religated to produce pRR-12 (Figure which contains no BalII sites. Desomer et al. (Molecular Microbiology (1992) 6 2377-2385) reported that this region was needed for plasmid replication. Thus, it was surprising that this construct was capable of producing Cmr transformants, indicating that this region was not essential for plasmid replication in the strain of organisms used for the present studies. This observation forms the conceptual basis for construction of a vector that will utilize a synthesized BglII site for 20 cloning the random genomic fragments. BllII accepts

DNA

digested by Sau3A, an effective and frequent cutter of IGTS8 DNA. These constructs are expected to allow for the production of better, more representative random genomic libraries.

a -56- Promoter Screening Example B: Desulfurization Cluster Reporter.

Vector pKAMI has been used as a second direct "shotgun" approach to finding a suitable alternative promoter (Figure 10). An NdeI site was engineered upstream of the promoterless Dsz cluster to serve as the site of insertion of random genomic DNA (from strains GPE-362, CPE-648 and IGTS8) fractionated by NdeI and the compatible 4bp cutters MseI and BfaI. Originally, this ligation mixture was directly transformed into GPE-362 cells, which were then used en masse to inoculate 250 ml BSM DBT. These efforts were undertaken with the goal of amplifying a superior Dsz+ strain due to its ability to utilize DBT as the sole source of sulfur. To date, 14 transformations of this type have been done. Of these, all but 2 have resulted in producing Dsz+ cultures. Eleven individual clones have been isolated Sand characterized. These are capable of low-level (0.6 1.0 mg/L 2-HBP/OD 60 constitutive expression of the desulfurization trait. Restriction analysis of plasmids 20 isolated from these eleven has revealed that all but one (KB4-3) are simple rearrangements of the pKAMI backbone resulting in.gratuitous expression from vector borne promoters. Many of the rescued plasmids show identical restriction patterns although originating from separate 25 ligations, suggesting an inherent vector instability. It i appears as if, with this type of selection, rearrangements -57of pKAMI that utilize a vector promoter sequence are strongly selected.

The above-described selection procedure has thus given way to a promoter screen geared to minimize the plasmid rearrangement. In this procedure, the pKAMI/genomic library is first amplified in E. coli, then the individual JM109 colonies are pooled together. The plasmids are extracted, and used to transform Dsz- strain GPE-362.

Instead of using en masse enrichment, the GPE362 transformaxt4 are plated to Rich Media chloramphenicol for selection of plasmid containing cells. Resulting colonies are replica-plated to BSM agarose DBT plates, then checked for desulfurizatton activity by UV fluorescence production. Over 7,000 GPE-362 transformants have been screened in this fashion. Thirty-six have been isolated from these which produce UV fluorescence on BSM

DBT

plates. Current efforts focus on the identification and I. characterization of the engineered plasmids borne by these 36 transformants.

Alternative Promotor Enineering S* The close physical arrangement of the three ORFs of pTOXI-1 does not provide sufficient space for promoters for either ORFs 2 or 3. This fact, coupled with the results of the subclone analysis in which intact ORFs 2 and 3 provided no activity (see pENOK-2, pENOK-11, and pENOK-13), suggested that this cluster of genes is organized as an operon -58with only one promoter for expression of the three genes.

Given that the desulfurization trait of IGTS8 is repressed by sulfate (Kilbane and Bielaga, Final Report D.O.E. Contract No. DE-AC22-88PC8891 (1991), it is possible that the operon promoter is tightly controlled by sulfur levels.

With the elucidation of the molecular arrangement of the desulfurization cluster, alternative promoters can be rationally engineered to eliminate the sulfur repression, increase expression of the desulfurization genes and thereby increase the specific activity of the Dsz* trait.

Examples of potential alternative promoters include other known and described promoters such as the chloramphenicol resistance gene promoter from Rhodococcus fascians (Desomer et al.: Molecular Microbiology (1992) 6 (16), 2377-2385), the nitrile hydratase gene promoter from Rhodococcus rhodochrous (Kobayashi, et al.: Biochimica et Biophysica Acta, 1129 (1991) 23-33), or other strong promoters isolated from Rhodococcus sp. by "shot-gun" promoter probing. Other potential-alternative promoters include those 20 from other Gram positive organisms such as Corvnebacterium, Bacillus, Streptomyces, and the like.

Promoter Engineering Example A: Expression from the chloramphenicol resistance gene promoter from Rhodococcus fascians.

25 pSBG-2 (Figure 11). The promoterless desulfurization cluster was isolated from pTOXI-1 as a 4.0 kb DraI/SnaBI -59fragment and ligated to a unique blunted AflII site of pRR- 6. This ligation inserted the cluster downstream of the chloramphenicol resistance gene promoter and upstream of the resistance structural gene. Thus, messenger RNA (mRNA) transcription should proceed through the Dsz gene cluster and proceed on to the resistance gene. However, original selections of transformants on chloramphenicol did not yield transformants, suggesting poor transcriptional readthrough. Dsz+ transformants harboring the plasmid were selected first through sulfur bioavailability assays and secondarily on chloramphenicol plates. Unlike IGTS8, pSBG- 2 transformants are capable of converting DBT to 2-HBP in BSM media supplemented with 20 mM Na 2

SO

4 which demonstrates the removal of sulfate repression by promoter replacement. Specific activity of transformants was measured between 0.9 and 1.7 mg 2-HBP//OD 600 /hr for a 16 hr culture in a rich media (RM) supplemented with 25 pg/ml ±chloramphenicol.

pSBG-3. The Rhodococcus origin of replication was removed from pSBG-2 by elimination of the 4.0 kb Xbal fragment. Without the.origin, transformation is obtainable only through integration. CPE-648 transformants with this plasmid were selected on RM chloramphenicol and replica-plated onto BSM DBT plates. Colonies were obtained 25 which produced 2-HBP, as detected by fluorescence after 18 hr of incubation at oooee Individual expression Of each ORF Recently, studies have been initiated to express the three ORFs separately, each engineered with an alternative promoter. These studies are expected to elucidate the following: First, any potential rate limiting steps in the desulfurizafion process will be identified and overcome.

Potential polarity effects of operon expression, i.e.

poorer expression of downstream ORFs 2 and 3, may be causing such rate limitations. Also, given the unresolved issue of the individual functions of ORFs 1 and 2, these studies are expected to demonstrate reconstitution of DBT-sulfone to 2-HBP conversion by the separate expression of ORFs 1 and 2.

All ORFs were isolated through PCR amplification and subsequent subcloning. A typical Shine-Dalgarno sequence and a unique cloning site for alternative promoters has been engineered upstream of each ORF. Stop codons in all reading frames have been engineered downstream of each ORF to prevent read-through. Additionally, convenient flankng 20 restriction sites for mobilization of the promoter/ORF fusions have been added to each primer. The primers used for amplification of each ORF are listed below. In-frame stop codons are marked with an asterik Sequences identical to pTOXI-I template DNA are shown in bold.

a ORF1UP: EcoRI NdeI Start ORFiDOWN: Stop XbaI 3 '-GTACTGTTCGGCGCAGCTGGGACTAGATCTTAAGC-5.S Sto-p* Stop EcoRI ORF2UP: Bq.'III 3 1 EcoRI NdeI Start ORF2 DOWN: StopBarIII 3 1 -CGGAGTTAGCGGTGGTATCTLCTAGACTTAAGC-5, Stop* Stop BcqlII ORF3UP: MseI GGTTCTTACATATGAGGACAGACCATCACATGTCCTA3 EcoRI NdeI Start ORF3DOWN: 0. 0 MseI 3 1* 3'GACTCCTAGACTCCGCATTTCTTAGC-51 :::Stop* Stop' Stop EcoRI Cycling parameters were: 1 x 960C 2.0 rai x 960C 30 sec 0 C 30 sec 720C 1.0 min Each ORF has been successfully amplified and subcloned into pUC-19 NdI as EcoRI fragments. Alternative promoters -62will be ligated into the unique Ndel sites, and the fusions will be moved to Rhodococcus-E. coli shuttle vector pRR-6 for expression in Rhodococcus.

Heterologous Expression of the Dsz+ Trait In order to determine whether plasmid pTOXI-1 contained all of the genetic material necessary for the Dsz+ trait, heterologous expression of pTOXI-1 was attempted in Rhodococcus fascians, a related organism which does not metabolize DBT (Dsz-) and in E. coli, a non-related organism which is also Dsz-.

A. Rhodococcus fascians (ATCC 12974), a Dsz- strain, was transformed with pTOXI-1. A single transformant demonstrated UV fluorescence on BSM DBT plates, and further analysis by HPLC clearly indicated production of 2-HBP when DBT was provided as a substrate. Thus pTOXI-1 contains sufficient information to convert a heterologous Dsz- strain to the Dsz+ phenotype.

B. E. coli strain JM109 was also transformed with pTOXI-1 and was incubated with each of the substrates DBT and DBT- 20 sulfone in either a minimal media (BSM) or a rich media (Luria Broth). In no case was production of 2-HBP observed by HPLC analysis. The inability of E. coli to express the desulfurization genes was not unexpected as gram positive genes are not universally expressible in E. coli without promoter replacement.

In order to replace the promoter of the desulfurization cluster, a 4.0 kb DraI/SnaBI fragment was isolated from pTOXI-1. This fragment contains all of the necessary structural genes but lacks the promoter sequences. This promoterless desulfurization cluster was ligated to E. coli expression vector pDR540 (Pharmacia, Piscataway, NJ) cut with i BamHI and ends made blunt with Klenow. The construction fuses the tac promoter to the desulfurization cluster. The -63tac promoter is under control of the lactose repressor and is repressed in a laci q host such as JM109. Expression from the tac promoter is inducible by the addition of isopropyl

B-D-

thiogalactopyranoside (IPTG). Transformants of JM109 harboring pDRDsz grown in Luria Broth at 30*C demonstrate the Dsz+ phenotype when incubated with DBT and induced with IPTG.

A

specific activity as high as 1.69 mg 2HBP/1/OD 600 /hr has been observed with pDRDsz. Activity is greatly diminished when transformants are grown at 37*C. The highest level of activity has been observed at lhr post induction.

The above-described expression of the Dsz+ trait in both a related and non-related heterologous host indicates that pTOXI-1 carries all of the genetic information required for conversion of DBT to 2-HBP.

Successful expression in E. coliprovided a workable system in which the proteins encoded by the desulfurization cluster could be identified and characterized. Total protein from Dsz+ cells of JMI09 (pDRDsz) was isolated and examined on denaturing acrylamide gels. No novel bands could be detected with Coomassie stain. Cellular fractionation of proteins into periplasmic, cytosolic and membrane components were also analyzed by Coomassie stained gels. Again, no novel bands were detected. Without any purification, the :newly expressed proteins were apparently levels too low to 25 easily detect and resolve from background.

Maxicell Analysis of.E. coli harborin DRDsz Proteins encoded by genes on plasmid DNA can be specifically radiolabeled in UV-irradiated cells of E. coli (Sancar, S et A. Journal of Bacteriology. 1979, p. 692-693). This 30 technique is known as Maxicell Analysis. Briefly, a recA strain of E. coli 9eS JM 10 9 which harbors a plasmid is grown in M9CA medium (Maniatis etal.) to a density of 2 x 108 cells/ml. Continuously stirred cells were then subjected to UV exposure from a Mineralight Lamp Model UVG-254 (Ultrovilet Products, Inc., San Gabriel, CA) at a distance of 10 cm f -64for a fluence rate of 0.5 Joules-*m 2 s- 1 Cells were exposed for either 60, 90 or 120 seconds. The cells were then incubated at 37°C for 16 hours after which they were then washed with M9 buffer and suspended in minimal medium lacking sulfate. After 1 hour of starvation at 37'C, 3 S]methionine (>1000 Ci/mmol) (NEN Research Products, Boston, MA) was added at a final concentration of 5 pCi/ml and incubation was continued for 1 hour. Cells were collected by centrifugation and proteins isolated through a boiled cell procedure (Maniatis, et Proteins were separated on an acrylamide gel.

After the run, the gel was dried and subjected to autoradiography for 3 days.

Maxicells of JM109 harboring vector pDR540 showed only vector marker galactokinase protein. Maxicells of JM109 harboring vector pDRDsz showed the presence of three novel protein bands of sizes which correlated well with the predicted molecular weights of the three proteins responsible for the Dsz+ trait, as predicted by open reading frame analysis (see Table 3).

Table 3 Open Reading Predicted Measured Frame Size (kDa) Size (kDa) ORF-1 49.5 49.5 ORF-2 38.9 33.0 S 25 ORF-3 45.1 45.0 Data obtained from Maxicell analysis thus indicated S. that the three .predicted open reading frames of pTOXI-1 encode three structural genes which constitute the desulfurization phenotype.

30 The relative intensity of the three novel bands is reflective of both the number of methionine residues and the level of translation for each of the proteins. Clearly, ORF-2 with only 1 Met gives the faintest band. In addition to the incorporation of only a single Met residue, E. coli may process the single terminal methionine, further reducing the amount of labelled protein. Therefore, the low intensity of the ORF-2 band most likely does not strictly suggest a low level of protein translation.

Interestingly, the ORF furthest from the promoter (ORF-3) appears to be present at levels comparable to ORF-1, indicating no polar effects in this operon when expressed in E. coli. It is expected that more significant information regarding protein levels will be obtained from a similar Maxicell analysis of a Rhodococcus sp. host containing plasmid pTOXI-I. Additionally, the presence of an ORF-I/ORF-2 heterodimer, postulated above, may be observable under non-denaturing conditions.

As required by 37 C.F.R. Section 1.821(f), Applicant's Attorney hereby states that the content of the "Sequence Listing" in this specification in paper form and the content of the computer-readable form (diskette) of the "Sequence Listing" are the same.

EQUIVALENTS

Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the 25 invention described herein. These and all other such equivalents are intended to be encompassed by the following claims.

-66- SEQUENCE LISTING GENERAL INFORMATION:

APPLICANT:

NAME: Energy BioSystems Corporation STREET: 3608 Research Forest Drive B-7 CITY: The Woodlands STATE: TX COUNTRY: US ZIP: 77381 TELEPHONE: 713-364-6100 TELEFAX: 713-364-6110 (ii) TITLE OF INVENTION: Recombinant DNA Encoding A Desulfurization Biocatalyst (iii) NUMBER OF SEQUENCES: (iv) CORRESPONDENCE ADDRESS: ADDRESSEE: Hamilton, Brook, Smith and Reynolds, P.C.

STREET: Two Militia Drive CITY: Lexington STATE: Massachusetts COUNTRY: U.S.A.

ZIP: 02173 COMPUTER READABLE FORM: MEDIUM TYPE: Floppy disk COMPUTER: IBM PC compatible OPERATING SYSTEM: PC-DOS/MS-DOS SOFTWARE: PatentIn Release Version #1.25 (vi) CURRENT APPLICATION DATA: APPLICATION NUMBER: FILING DATE:

CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION: NAME: Brook, David E REGISTRATION NUMBER: 22,592 REFERENCE/DOCKET NUMBER: EBC92-03A (ix) TELECOMMUNICATION INFORMATION: TELEPHONE: 617-861-6240 TELEFAX: 617-861-9540 INFORMATION FOR SEQ ID NO:1: SEQUENCE CHARACTERISTICS: S" LENGTH: 5535 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) -67- (ix) FEATURE: NAME/KEY:

CDS

LOCATION: 790. .2151 (ix) FEATURE: NAME/KEY:

CDS

LOCATION: 3256. .4506 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1: GCATGCACGT CGCGCCGACG CATTTGCGCG CACGGCTCCG

GGCAGTTCTC

AGGCACGGAT GGGCACCCTC2 AAACCCTCGG

CTACAGCCCT

TTTGCACGCT ACGTGGCGAC

C

GTGTAGCGCC GCTTAACGGG I1 CCGCAGGTAG TCGACCACCC

C

TTGCTCGCAG ATCGGCTGAT

G

GCATGGCGTC CGGTGCATAC

A

ACGTCGACTC GATGCGCCGT

G

CAAGGATAGA GATTCGAAGG

A

AGGCGGCCAJA GTCATCGGCA

C

GGCCTGAATT TGGCTGGTGG

A

GGGGGTGGGG GTGAGACTGC

T

GCATACGCG ATG ACT CAA C Met Thr Gin G 1 TCG GCC GGC AAT GTG ACT Ser Ala Gly Asn Val Thr is GCG TCG AAT GAC TTT CTG Ala Ser Asn Asp Phe Leu 30 35 ACT CTG GAG CGC GGC AAG Thr Leu Giu Arg Gly Lys s0 CCC GTC GAG GAC AGC TAC Ala Val Giu Asp Ser Tyr 65 GCC GGG CAG GGT GCA GTC Gly Gly Gin Gly Ala Val

!.ACGAACTCA

AGACATTGG

;CGGCTGGAC

'GCGCACGGC

TTCCCGCAG

~TTGCCGATC

~CGACGATCT

CGAGTGAGA

.CCTCGGATC

CGTCACCGT

GCATTGAAA

TAGCGACAG

CCCAAACCAC

AAGCTCATGC

TGACGCTGGA

GGGACATCGG

CGGTCGGAGG

GACGTGGTCG

AACCAGATCG

TCCTTTGTGG

GACCCAAATG

CACCTTGACC

TCAGGTGAAG

GAATCTAGCC

GCCGATAGCG

GCGACGCATC

GGTCCGACCC

CCAGCTGGCT

TGATCGACCG

ACGGGACACG

ACGGTTTTGA

TGCTTGGCTA

CGGACGGCCG

CGACGTGCCC

TTTAACGGTG

ATGATTGACA

GCGGCGCTG

ATCCTCGCCG

CGGATCGACC

GACGTGTGTG

TGCCCCTCCT

TTAGGGTCAT

CTCGCGATTG

GCGTCGGTCA

TTGACCTCGA

GCAGCGGCGA

CGTGGTTCAA

GGCACACCCC

TTTAAAGGAC

120 180 240 300 360 420 480 540 600 660 720 780 828 876 924 972 1020 1068 0* a a a.

a a.

a.

a a a .a a a a.

a. *a a a A.A OGA CAA ATG CAT CTG CCC GGT TTC TTC in Arg Gin Met His Leu Ala Giy Phe Phe 5 CAT GCA CAT COG GCG TGG CGG CAC ACC GAC His Ala His Gly Ala Trp Arg His Thr Asp 20 TCG GGG AAG TAC TAC CAA CAC ATC GCC CGT Ser Gly Lys Tyr Tyr Gin His Ile Ala Arg 40 TTC GAT CTG TTG TTT CTG CCT GAC GGG TTG Phe Asp Leu Leu Phe Leu Pro Asp Gly Leu 55 GOG GAO AAC CTG GAC ACC GGT GTC GGC CTG Gly Asp Asn Leu Asp Thr Gly Val Gly Leu 70 GCC TTG GAG CCC GCC ACT GTG GTC GCA ACC Ala Leu Giu Pro Ala Ser Val Val Ala Thr 85 ATG CCC GC GTG ACC GAG CAC CTG GGT OTT GGG GCA ACC ATT TCG GCG 1116 -68- Met Ala Ala Val Thr Giu His Leu Gly Leu Gly Ala Thr Ile Ser Ala 100 105

ACC

Thr 110

CAG

Gin

GAC

Asp

CCC

Ala

TGG

Trp

GTG

Val 190

TG

Trp

GAG

Glu

GCC

Al a

GTG

Val

GGG

Gly 270

CTC

Leu

AGT

Ser

GGC

Gly

TAC

Tyr

TTG

Leu

GCT

Ala

CGC

Arg

AAC

Asn 175

TTC

Phe

CTG

Leu

CCG

Pro

GGG

Gly

ATG

Met 255

CGC

Arg

GC

Cly

CTG

.Leu

ATC

I le

TAT

Tyr

TCA

Ser

GAA

Glu

TAT

Tyr 160

AGC

Ser

GCC

Ala

AAT

Asn

GTG

Val

AAG

Ly s 240

CAG

Gln

GAT

Asp

GAA

Glu

GTC

Val

AAC

Asn 320

CCC

Pro

GGG

Cly

GCG

Ala 145

GAC

Asp

TGG

Trp

GAT

Asp

GTG

Val

ATC

Ile 225

TGG

Trp

CC

Ala

CCC

Pro

AGC

Ser

CAT

His 305

CTG

Leu

CCC

Pro

GGT

Gly 130

CC

Arg

CGC

Arg

GAC

Asp

CCC

Pro

CC

Arg 210

CTG

Leu

GCC

Ala

ACC

Thr

GAT

Asp

CAG

Gin 290

CCC

Pro

GC

Ala

TAT

Tyr 115

CCC

Arg

AAC

Asn

GCC

Ala

GAG

Glu

C

Ala 195

CGA

Gly

CAG

Gin

GAG

Glu

TAC

Tyr

CAG

Gin 275

CC

Ala

GAA

Glu

GC

Ala

CAC

His

GTG

Val

TTC

Phe

CAT

Asp

CAC

Asp 180

AAG

Lys

CCT

Pro

CC

Ala

GCC

Ala

CAG

Gin 260

ACC

Thr

CTG

Val

CTC

Val

TAC

Tyr

CTT

Val

TCC

Ser

GC

Cly

GAG

C iu 165

CC

Ala

CTG

Val

CTC

Leu

GCC

Gly

GTC

Val 245

GCC

Gly

AAA

Lys

CCA

Ala

GCA

Gly

CCT

Pro 325

CCT

Al a

TGG

Trp

ATT

Ile 150

TTC

Phe

CTC

Leu

CAC

His

CAG

Gin

CTG

Leu 230

TTC

Phe

ATC

Ile

ATC

Ile

CAG

Gin

CTG

Leu 310

CTC

Leu

CCC

Arg

AAC

Asn 135

AAT

Asn

TTC

Leu

CTG

Val1

TAC

Tyr

GTA

Val 215

TCG

Ser

ACT

Ser

AAA

Lys

TTC

Phe

GAA

Glu 295

TCG

Ser

GAC

Asp

CTG

Val 120

GTC

Val

CAG

Gin

GAA

Giu

CTG

Leu

GTC

Val 200

CCC

Pro

CCC

Pro

CTT

Leu

CC

Ala

ACC

Thr 280

CCA

Arg

ACC

Thr

ACT

Thr

TTC

Phe

GTC

Val

CAT

His

C

Ala

CAC

Asp 185

GAT

Asp

CCT

Arg

CCC

Arg

CCA

Ala

GAG

Glu 265

CC

Ala

CTC

Leu

CTA

Leu

CCG

Pro

CC

Ala

ACC

Thr

CTG

Leu

GTC

Val 170

AAG

Ly s

CAC

His

TCA

Ser

CGT

Gly

CCC

Pro 250

GTC

Val

GTG

Val

GAA

Glu

TCC

Ser

ATC

Ile 330

ACC

Thr

TCG

Ser

GAA

Giu 155

AAC

Lys

C

Ala

CAC

His

CCT

Pro

CCC

Arg 235

AAC

Asn

GAC

Asp

ATG

Met

TAT

Tyr

ACT

Ser 315

AAG

Lys

CTC

Leu.

CTC

Leu 140

CAC

His

AAA

Lys

CC

Ala

CCC

Gly

CAC

Gin 220

CC

Arg

CTC

Leu

CCT

Ala

CCC

Pro

CTC

Leu 300

CAC

His

GAC

Asp

CAT

Asp 125

AAC

Asn

GAC

Asp

CTC

Leu

GC

Gly

GAG

Giu 205

CCT

Gly

TTC

Phe

GAG

C lu

GCG

Ala

CTA

Val 285

AAC

Asn

ACC

Thr

ATC

Ile 1164 1212 1260 1308 1356 1404 1452 1500 1548 1596 1644 1692 1740 1788 1836

C

a.

C

a C C C. C C. C a CTG CCC CAT CTC CAG CAT CCC AAT CTC CCC ACC CAA CTC CAC ATG TTC Leu Arg 335 Asp Leu Gin Asp Arg 340 Asn Val Pro Thr Gin Leu His Met Phe 345 -69ccc Ala 350

CGC

Arg CCC GCA ACG CAC Ala Ala Thr His

AGC

Ser 355

GTG

Val GAA GAG CTC ACG Glu Giu Leu Thr GCG GAA ATG GGT Ala Glu Met Gly

CGG

Arg 365 TAT GGA ACC Tyr Gly Thr

AAC

Asn 370

GAC

Asp GGG TTC GTT Gly Phe Val

CCT

Pro 375

CAC

His TGG GCC GGT Trp Ala Gly ACC GGG Thr Gly 380 GAG CAG ATC Giu Gin Ile CAT GGT TTC Asp Cly Phe 400 TTC GTC GAC Phe Val Asp

GCT

Ala 385

ATC

Ile GAG CTG ATC Glu Leu Ile

CGC

Arg 390

TTC

Phe TTC GAG GGC Phe Glu Gly ATC TCT CCC Ile Ser Pro CTG CCG GGC Leu Pro Gly

TC

Ser 410

GGC

Gly GGC CCC GCG Cly Ala Ala 395 TAC GAC GAG Tyr Asp Giu TAC TTC CC Tyr Phe Arg CAG OTO GTT Gin Val Val 415 ACC GAG Thr Giu

CCG

Pro 420 CTG CAG GAT Leu Gin Asp

CC

Arg 425 TAC CAG GOC Tyr Gin Gly 430

CCA

Pro AAC ACT Asn Thr 435 CAA CCT Gin Pro CTG CGC GAC CAC TTG OCT CTG CGC GTA Leu Arg Asp His Leu Gly Leu Arg Val 440 445 TCA TGACAAGCCG CGTCGACCCC GCAAACCCCG Ser CAA CTG CAA Gin Leu Gin

GGA

Gly 450

S

S

SS

59 S.

S

S

.5

S

S.

*5 S

S

*5 S S 5.55 S 5 5* S.

S

*559

GTTCAGAACT

ACGCTCTGCT

TCCTCAGCGG

TTGGGGGTGA

TACTCGGCAT

TCACAGCGGC

TCCTGCGCGG

CGCTGGGCTC

TGGACGACGT

TCGAAGAATC

CCGCGGTCTT

AGTTGCAAGC

ACCCCAGTGT

GACTGGCGA

CCAGCCTGCA

CCGACTTCCA

CGATTCCGCC

CACGGCATCG

CCAGCAGGGC

GATCCCGCCA

CACCCCGCTC

CGCCGACCTT

CCAGCTGGGC

GTGGGAGGCG

CGAGCTGGTG

GGCGACCGTC

GGCCAGCGGA

CACCGGGGCC

GTGGACGGTC

CGCGGCCGTC

CGCCGCGAAC

GCAGCGTCTG

ATCCGCGACA

GAATCGGGCT

ACGGTTCATT

CTGCTCAGCG

TTGGGGCGCC

GCCGGACGTC

GACTACCTCG

CGCGCCTTGT

CCGATCAGCA

AAGGGTGCGG

GACGTTGACG

CGCCCAGTCG

AGCAGCGGGC

GACGCCGGGC

CTGGGCGTAT

GTTCCACGCC

CACTGACCTA

TCCTCGACC

TCACCTACGA

AGGGGTTGCG

AGGGCTTCTT

GAATCGGCGT

AGTTGGATCC

TGCACACCCT

GTCCTGGTGT

ACCTCTTTCC

CCCTGTACAG

TGGATCTCGG

TGGTTCGCCA

TGTGGGCACG

CGACCGGAGC

TGGATCACGA

CAGCAACTGC

CGCCGGCATC

CCAGCCTGCC

GGCACCTGGG

TGTCCGCGAC

CTCGGCCTCG

CTGGCGGCAA

TGAGCACGGT

CGATGTTCCC

CGATGTCGCC

TTGCCTGCCC

CCTCGATCAG

GCGACCTGGC

CGATCATTCC

AGTAGGCCAG

CGCCCTCCC

CCGGTACCCA

GAACTCGACG

TACACCCGTT

CGCACGCGTC

GACAGCCCGA

GCAATTCGCA

ACGCTGGTAG,

GAACTGGGTG

GCTGAGCAGC

CGCGGTCAGC,

TCGGCCGGGG

CGCAATGCCT

CTTGTTCAAC

GACGCGGTGA

CGCTTCGGCG

CTCCTGGAGC

1884 1932 1980 2028 2076 2124 2178 2238 2298 2358 2418 2478 2538 2598 2658 2718 2778 2838 2898 2958 3018 3078 3138 G CACACAG CA ATTCCTG CTC ACCAACAACT TGCTGCAGGA GGGCGGCTCC GGAATTTCTG AACAACAGCC TCAATCGCCA ACCCGTCGCC CTCGATCAGT CCGATAGGAA CATCCGC ATG ACA Met Thr CTG TCA CCT GAA AAG Leu Ser Pro Glu Lys

I

GAC

Asp

GCC

Ala

CGC

Arg

GAA

Giu

CGC

Arg

CAC

His*

GAA

Giu

AAT

Asn

GCC

Ala 145

TTC

Phe

GTC

Val

CCG

Pro

GGC

Gly

GTC

Val

AAC

Asn

ACC

Thr

GAA

Giu 50

TAC

Tyr

GAA

Glu

CTC

Leu

GAA

Giu

GCC

Ala 130

ACC

Thr

TGC

Cys

CAG

Gin

ACA

Thr

ATG

Met 210

GAG

Glu

GAT

Asp

GCC

Ala 35

GAC

Asp

GGC

Gly

ATC

Ile

ACC

Thr

CAC

His 115

TCC

Ser

CCG

Pro

AGC

Ser

GAT

Asp

TCG

Ser 195

CGG

Arg

CCT

Pro

CCC

Pro 20

GTC

Val

CTG

Leu

GGC

Gly

GCG

Ala

AAC

Asn 100

CTG

Leu

AGC

Ser

ACC

Thr Gcc Gly

GAT

Asp 180

CGG

Arg

CAG

Gin

GAC

Asp 5

GTC

Val

GAG

Glu

CGC

Arg

TGG

Trp

GCA

Ala

GCC

Ala

TAC

Tyr

GAG

Glu

GAA

Glu

GCC

Ala 165

TCT

Ser

GCT

Ala

ACC

Thr

GAA

Glu

GCG

Ala

CGT

Arg

GCG

Ala

GGC

G ly 70

GCC

Ala

CCG

Pro

ACC

Thr

AAC

Asn

GAC

Asp 150

AAG

Lys

CCG

Pro

GGC

Gly

GAC

Asp

GTG

Val

GTT

Val

GAT

Asp

AGC

Ser 55

GCA

Ala

GAT

Asp

ATG

Met CAd Gin

AAC

An 135

GGC

Gly

GGG

Gly

CAG

Gln

GTT

Val

AGC

Ser 215

CTG

Leu

CAG

Gin

GCC

Ala

CGC

Arg 40 G CG Ala

GAC

Asp

GGA

Gly

ATC

Ile

ATO

Ile 120

AGC

Ser Gd C Gly

TCG

Ser

CAG

Gln

ACG

Thr 200

GGT

Gly

GGC

Gly

CGT

Arg 25

GCC

Ala

CTG

Leu

TGG

Trp

TCT

Ser

GAA

Giu 105

GCG

Ala

CAC

His

TAC

Tyr

GAC

Asp

GGT

Gly 185

CCC

Pro

TCC

Ser

GCG

Ala 10

GGG

Gly

GGG

Gly

CTG

Leu ccc Pro

TTG

Leu 90

CTG

Leu

CAG

Gin

GTG

Val

GTG

Val1

CTG

Leu 170

GCG

Ala

AAC

Asn

ACG

Thr

CCC

Pro

CTA

Leu

GGT

Gly

TCG

Ser

ACC

Thr 75

GGA

Gly

ATC

Ile

AAC

Asn

CTG

Leu

CTC

Leu 155

CTG

Leu

ATC

Ile

GAC

Asp

GAG

Asp

AAC

An CAC GTT CGA CCA CGC GAC GCC GCC His Val Arg Pro Arg Asp Ala Ala

GCC

Ala

TCG

Ser

CTC

Leu 60

GCC

Al a

CAC

His

GGC

Gly

AAC

An

GAC

Asp 140

AAT

An

TTC

Phe

ATT

Ile

GAC

Asp

TTC

Phe 220

GCC

Ala

GAA

Giu

GCA

Ala 45

CTC

Leu

ATC

Ile

CTG

Leu

TCG

Ser

TGG

Trp 125

TGO

Trp

GGC

Gly

GTG

Val

GCT

Ala

TGG

Trp 205

CAC

His

TTC

Phe

AAG

Lys 30

ACA

Thr

GTC

Val1

GAG

Giu

TTC

Phe

CAG

Gin 110

TGG

Trp

AAG

Lys

ACG

Thr

TTC

Phe

GCC

Ala 190

GCC

Ala

AAC

Asn

GTT

Val

TGG

Trp

GCC

Ala

CCG

Pro

GTC

Val

GGA

Gly

GAA

Glu

ACC

Thr

GTC

Val

AAG

Lys

GGC

Gly 175

GCT

Ala

GCC

Ala

GTC

Val

CTC

Leu

CGA

Arg

GAG

Giu

CGC

Arg

GTC

Vai

TAC

Tyr

CAA

Gin

GGA

Gly

AGC

Ser

CAC

His 160

GTC

Val

ATC

Ile

ATC

Ile

AAG

Lys

CC

Ala 3198 3255 3303 3351 3399 3447 3495 3543 3591 3639 3687 3735 3783 3831 3879 3927 3975 :-,too 0 0*00 0 9 06 66* 9. 9 *00.0' Go**:9 9 .Ooa -71- 225

TTC

Phe ATA CAA TCC Ile Gin Ser 230

CGC

Arg 235

GCG

Ala

GAG

G iu 245

GTC

Val GGC AGC CTC Gly Ser Leu

TTC

Phe 250

GCG

Ala CCC ATA GCG Pro Ile Ala ATC TTC GC Ile Phe Ala GCC AGO GAG Ala Arg Glu 275 ATT CAA CAG Ile Gin Gin

AAC

Asn 260

TAC

Tyr TAT CTG GGG Tyr Leu Giy

ATC

Ile 265

GCG

Ala CAC GGC GCA His Gly Ala

CTC

Leu 270

CCG

Pro 240 CAA TTG Gin Leu 255 GAT GCC Asp Aia GCC GGT Ala Gly ACC CGT ACC Thr Arg Thr AGG CCC TGG Arg Pro Trp

ACA

Thr 285

TCC

Ser GCA ACC GAG Ala Thr Glu 290 TTC ACC Phe Thr

GAT

Asp 295

GGA

Gly TAC ACC ATC Tyr Thr Ile

CC

Arg 300

GCC

Ala TAC OCT GAG Tyr Gly Giu ATC GCA TTG Ile Ala Leu 305

CAC

His

CAG

Gin 310

GTG

Val GCT GAC GCC Ala Asp Ala

GCC

Ala 315

GAC

Asp CGT GAA GCG Arg Giu Ala

GCC

Ala 320 CTG CTG CAG Leu Leu Gin

ACG

Thr 325

CTG

Leu TGG GAC AAG Trp Asp Lys

GGC

Gly 330

TCG

Ser GCG CTC ACC Ala Leu Thr CCC GAG Pro Glu 335 GAC CGC GGC Asp Arg Gly ACC AAC GCC Thr Asn Ala 355 GCG CGC GGA Ala Arg Gly

GAA

Giu 340

GCC

Ala ATG GTG AAG Met Val Lys

GTC

Val 345

AGC

Ser GGA GTC AAA Giy Val Lys CTC AAC ATC Leu Asn Ile

AGC

Ser 360

TAC

Tyr GGC GTC TTC Gly Val Phe

GAG

Glu 365

TTC

Phe GCO TTG GCC Ala Leu Ala 350 GTG ATC GGC Val Ile Gly TOO CGC AAC Trp Arg Asn 4023 4071 4119 4167 4215 4263 4311 4359 4407 4455 4503 4556 4616 4676 4736 4796 4856 4916 ACA CAT CCC Thr His Pro

S.

S

S. 0@S S OS .5 S S

S

5* t.

S.

5*

S

S.

S S

S.

6*SS S S *5 6 5* 6 6 St..

SSceee 0 S Sd 0 5 S

GTG

Val 385

GTC

Val

TCC

Ser 370 CO C Arg

AGO

Arg 375

CAC

His GGT TTC GAC Gly Phe Asp ACC CAC TCC Thr His Ser

CTG

Leu 390

TTG

Leu GAC CCG GTG Asp Pro Val

CGC

Arg 380

TAC

Tyr AAG ATC GCC Lys Ile Ala

GAC

Asp 400

ACC

Thr GGC AAG CAC Gly Lys His

ACC

Thr 405 AAC GGT CAA Asn Gly Gin

TAC

Tyr 410 ATT CCC GGC TTC Ile Pro Gly Phe 415 CGC GGCCGAGTCG TGAGGATCTG AGGCGCTGAT CGAGCCGAG GCCACCG

CGA.ATCOCCC

CAAGACCTGT

GCGGGTGAGA

CGGACCAGCG

TCAGCTTGTA

TTCGCAACCG

GCCGATACTC

GGATGAGGGT

TGTCGATGGG

CGACCAGATG

GGCCATGGCA

CACGGTCGCA

AGCTTCTCCA

GTTTCAGGCG

TCOTTCGAGC

GGGGGCGTTG

ATCCCGCTGC

AACGTCGATT

TACGTACGGG

ACCTCCGTTT

AGCTTGCCTT

ACCGCCGCCA

GACGTOACCC

CGATCGGATT

TGCACACAAG

CGCTTGATTC

TGTGGAACAC

GCGGGCTTGT

AGGGCCCTTG

CGTAGTGCGC

GAGATATTGT

GTCGGGCTCA

CGCGCCGGCA

GCOGCGTCGA

GTOACCTTGG

AAGTGGATCC

-72-

AGTGCTCGGC

CGACCCCCTT

CCTTGTCGAT

CCGACTTGGG

GGGTATCCGG

CCAGCACCGG

CTGTGGACTC

CGCCGATCAT

CGTCGACCCA

CCTCGTCCTG

TGCCGAGGAA

CGGGTACCG

GGGGTACTTC

GTCCTCGGCG

CAGCGCAGCA

AAACACCTCC

GGCGGTCATC

CCGGAACCCA

CACGAGCAAG

TAGGTACACG

CCATTGCGCG

CTGCTCCAA

TAGAACTCCA

GCGCCGTAAT

TTGTAGATTT

AGCACATTGG

CGCAGTGCCT

CCGCGGCGTC

TCGGTGAGCG

CACAGCTTCT

AA.ATCGGTGC

GTCAGCCGGG

GGAGCACGTC

CTACCTCGAA

CCCGCATCGC

CCTGCTTGTG

TCCAGAACCC

GGCATGAGCG

CGACOAGCTC

CCTGCTCCAG

CCGAGAGATC

TGATCGTCGA

GGCGTCGTCG

GGCCTTGATC

CGCGGTCGCA

AAACCAGCAG

CAGCGCCCCA

CAGCAGATCA

CTTGCGGCCG

GCGGACATTG

CCGGTCGGCG

GGCCGACAGC

ACGATCTTGG

GCGACCTGGG

CCTGGATGAG,

CGCTGTTCAC

TCACCGACGG

GCCCACGACT

TCGGCGCGGA

AGATGGATGC

AAGGCCTTCG

CCGGCACCAG

4976 5036 5096 5156 5216 5276 5336 5396 5456 5516 5535 INFORMATION FOR SEQ ID NO:2: SEQUENCE CHARACTERISTICS: LENGTH: 454 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: Met

I

Asn Asp Thr Gin Gin Val Thr His 20 Phe Leu Ser Arg 5 Ala Gin Met His Leu Ala Gly Phe Phe Ser 10 Arg Ala Gly His Gly Ala His Thr Asp a a Ala Ser Asn Thr Leu Glu Gly Lys Tyr 35 Arg Gly Lys Tyr Phe His Ile Ala Phe Asp Leu 50 Asp Ser Leu 55 Leu Leu Pro Asp Gly Gly Tyr Gly Asp Asp Thr Gly dly Val 75 Leu Ala Val Giu Leu Gly Gly Gin so Ala Val Ala Leu 85 Leu Glu Pro Ala Ser Val Thr Pro Pro Gly Gly 130 Ala Arg Glu His 100 Gly Leu Gly Ala 105 Phe Val Val Ala 90 Thr Ile Ser Ala Thr Leu Thr Met Ala Ala Tyr His Val Ala 125 Arg Val Ser Trp Asn Phe Gly Ile Arg Val 120 Asn Val Ala Asp 125 Asn Thr Tyr Tyr 110 Gin Leu Ser Asp Ala Glu Val Thr Ser 135 Asn Gln His Leu Glu Asp Ala Arg Tyr -73- 160 Asp Arg Trp Asp Asp Pro Val Arg 210 Ile Leu 225 Trp Ala Ala Thr Pro Asp Ser Gin 290 His Pro 305 Leu Ala Leu Gln Thr His Thr Asn 370 Ala Asp 385 Ile Ile Gin Val Gin Gly Gin Gly 450 Ala Asp Giu 165 Giu Asp Ala 180 Ala Lys Val 195 Gly Pro Leu Gin Ala Gly Giu Ala Vai 245 Tyr Gin Giy 260 Gin Thr Lys 275 Ala Vai Ala Giu Vai Gly Ala Tyr Pro 325 Asp Arg Asn 340 Ser Giu Giu 355 Vai Gly Phe Glu Leu Ile Ser Pro Ala 405 Val Pro Val 420 Asn Thr Leu 435 Gin Pro Ser Phe Leu His Gin Leu 230 Phe Ile Ile Gin Leu 310 Leu Val Leu Val1 Arg 390 Phe Leu Leu Giu Val Leu Tyr Val 200 Val Pro 215 Ser Pro Ser Leu Lys Ala Phe Thr 280 Giu Arg 295 Ser Thr Asp Thr Pro Thr Thr Leu 360 Pro Gin 375 His Phe Leu Pro Gin Asp Ala Asp 185 Asp Arg Arg Ala G iu 265 Ala Leu Leu Pro Gin 345 Ala Trp Giu Gly Arg 425 Val 170 Lys His Ser Gly Pro 250 Val Vai Giu Ser Ile 330 Leu Giu Ala Gly Ser 410 Gly Lys Ala His Pro Arg 235 Asn Asp Met Tyr Ser 315 Lys His Met Gly Gly 395 Tyr Tyr Lys Ala Gly Gin 220 Arg Leu Ala Pro Leu 300 His Asp Met Gly Thr 380 Ala Asp Phe Leu Gly Giu 205 Gly Phe Giu Ala Val 285 Asn Thr Ile Phe Arg 365 Gly Ala Giu Arg Val 445 Trp Val 190 Trp Glu Ala Val1 Gly 270 Leu Ser Gly Leu Ala 350 Arg Giu Asp Phe Thr 430 Asn Ser 175 Phe Ala Leu Asn Pro Val Gly Lys 240 Met Gin 255 Arg Asp Gly Glu Leu Vai Ile Asn 320 Arg Asp 335 Ala Ala Tyr Gly Gin Ile Gly Phe 400 Val Asp 415 Glii Tyr I. p p p. p p p Arg Asp His Leu Gly Leu Arg 440 Pro Gin Leu -74- INFORMATION FOR SEQ, ID NO:3: SEQUENCE CHARACTERISTICS: LENGTH: 416 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:3: Met Thr Leu Ser Pro Giu Lys Gin His Val Arg Pro Arg Asp Ala Ala Asp Ala Arg Giu Arg His Giu Asn Ala 145 Phe Val Pro Gly Val 225 Phe Asn Thr Glu Tyr Glu Leu Glu Ala 130 Thr Cys Gin Thr Met 210 Giu Ile Asp Ala Asp Gly Ile Thr His Ser Pro Ser Asp Ser Arg Pro Gin Pro Val Leu Gly Ala As n 100 Leu Ser Thr Giy Asp 180 Arg Gin Asp Ser Val1 Glu Arg Trp Ala Ala Tyr Giu Glu Ala 165 Ser Ala Thr Glu Giu 245 Ala Arg Al a Gly 70 Ala Pro Thr Asn Asp 150 Lys Pro Gly Asp Val 230 Arg Val1 Asp Ser 55 Al a Asp Met Gin Asn 135 Gly Giy Gin Val Ser 215 Leu Gly Ala Arg 40 Ala Asp Gly Ile Ile 120 Ser Gly Ser Gin Thr 200 Gly Gly Ser Arg 25 Ala Leu Trp Ser Glu 105 Ala His Tyr Asp Gly 185 Pro Ser Ala Leu Gly Gly Leu Pro Leu 90 Leu Gin Val Vali Leu 170 Ala Asn Thr Pro Phe 250 Leu Giy Ser Thr 75 Gly Ile Asn Leu Leu 155 Leu Ile Asp Asp Asn 235 Ala Ala Ser Leu Ala His Gly Asn Asp 140 Asn Phe Ile Asp Phe 220 Ala Pro Glu Ala Leu Ile Leu Ser Trp 125 Trp Gly Val Ala Trp 205 His Phe Ile Lys Trp Thr Aia Val Pro Glu Val Phe Gly Gin Glu 110 Trp Thr Lys Val Thr Lys Phe Gly 175 AiaAla 190 Ala Ala Asn Val Val Leu Ala Gin Arg Giu Arg Val Tyr Gin Gly Ser His 160 Val Ile Ile Lys Ala 240 Leu 255 Ile Phe Ala Asn Val Tyr Leu Gly Ile Ala His Gly Ala Leu Asp Ala 270 260 265 Ala Arg Glu 275 Ile Gin Gin Tyr Thr Arg Thr Gin 280 Pro Ala Arg Pro Trp Thr 285 Ser Pro Ala Gly Tyr Gly Giu Ala Thr Giu 290 Phe Thr Asp 295 Gly Tyr Thr Ile Arg 300 Ala Ile Ala Leo 305 His Gin 310 Val Aia Asp Ala Ala 315 Asp Arg Glu Ala Ala 320 Leu Leu Gin Thr 325 Leu Trp Asp Lys Gly 330 Ser Ala Leu Thr Asp Arg Gly Thr Asn Ala 355 Ala Arg Gly Met Val Lys Giy Vai Lys Pro Giu 335 Leo Ala Ile Gly Ala Leo Asn Ile Ser 360 Tyr Gly Val Phe Glu 365 Phe Thr His Pro 370 Val Arg Arg 375 His Gly Phe Asp Arg 380 Tyr Trp Arg Asn Thr His Ser 385 Val Leo 390 Leu Asp Pro Val Lys Ile Ala Asp 400 Thr Gly Lys His Thr 405 Asn Giy Gin Tyr 410 Ile Pro Gly Phe 415 INFORMATION FOR SEQ ID NO:4: SEQUENCE CHARACTERISTICS: LENGTH: 5535 base pairs TYPE: nucleic acid STRANDEDNESS: double TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: NAME/KEY: CDS LOCATION: 2148. .3245 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: GCATGCACGT CGCGCCGACG CATTTGCGCG CACGGCTCCG AGGCACGGAT GGGCACCCTC AACGAACTCA CCCAAACCAC AAACCCTCGG CTACAGCCCT CAGACATTGG AAGCTCATGC TTTGCACGCT ACGTGGCGAC GCGGCTGGAC TGACGCTGGA GTGTAGCGCC GCTTAACGGG TGCGCACGGC GGGACATCGG CCGCAGGTAG TCGACCACCC CTTCCCGCAG CGGTCGGAGG TTGCTCGCAG ATCGGCTGAT GTTGCCGATC GACGTGGTCG

GGCAGTTCTC

GCCGATAGCG

GCGACGCATC

GGTCCGACCC

CCAGCTGGCT

TGATCGACCG

ACGGGACACG

GCGGCGCTGG

ATCCTCGCCG

CGGATCGACC

GACGTGTGTG

TGCCCCTCCT

TTAGGGTCAT

CTCGCGATTG

120 180 240 300 360 420 -76- GCATGGCGTC CGGTGCATAC ACGACGATCT AACCAGATCG ACGGTTTTGA GCCTCGGTCA

ACGTCGACTC

CAAGGATAGA

AGGCGGCCAA

GCCTGAATT

GGGGGTGGGG

GCATACGCGA

GTGACTCATG

AAGTACTACC

CCTGACGGGT

GGCGGGCAGG

ACCGAGCACC

GCTCGGGTGT

ACCTCGCTCA

GCCCGCTATG

GACGAGGACG

CACTACGTCG

TCACCTCAGG

GCCGGGAAGT

ACCTACCAGG

ATCTTCACCG

GAATATCTCA

GGCATCAACC

CAGGATCGGA

CTCACGCTGG

GCCGGTACCG

GATGGTTTCA

GTGGTTCCGG

CGCGACCACT

GATGCGCCGT

GATTCGAAGG

GTCATCGGCA

TGGCTGGTGG

GTGAGACTGC

TGACTCAACA

CACATGGGGC

AACACATCGC

TGGCCGTCGA

GTGCAGTCGC

TGGGTCTTGG

TCGCGACGCT

ACGACGCTGA

ACCGCGCCGA

CCCTCGTGCT

ATCACCACGG

GTGAGCCGGT

GGGCCGAGGC

GCATCAAAGC

CCGTGATGCC

ACAGTCTGGT

TGGCGGCGTA

ATGTCCCGAC

CGGAAATGGG

GGGAGCAGAT

TCATCTCTCC

TTCTGCAGGA

TGGGTCTGCG

GCGAGTGAGA

ACCTCGGATC

CCGTCACCGT

AGCATTGAAA

TTAGCGACAG

ACGACAAATG

GTGGCGGCAC

CCGTACTCTG

GGACAGCTAC

CTTGGAGCCG

GGCAACCATT

CGATCAGTTG

AGCGCGCAAC

TGAGTTCTTG

GGACAAGGCG

GGAGTGGCTG

GATCCTGCAG

CGTCTTCAGT

CGAGGTCGAC

GGTACTCGGC

CCATCCGGAA

CCCTCTCGAC

GCAACTGCAC

TCGGCGCTAT

CGCTGACGAG

GGCCTTCCTG

TCGCGGCTAC

CGTACCACAA

TCCTTTGTGG

GACCCAAATG

CACCTTGACC

TCAGGTGAAG

GAATCTAGCC

CATCTGGCCG

ACGGACGCGT

GAGCGCGGCA

GGGGACAACC

GCCAGTGTGG

TCGGCGACCT

TCAGGGGGTC

TTCGGCATTA

GAAGCGGTCA

GCCGGCGTGT

AATGTGCGCG

GCCGGCCTGT

CTTGCACCCA

GCTGCGGGGC

GAAAGCCAGG

GTGGGACTGT

ACTCCGATCA

ATGTTCGCCG

GGAACCAACG

CTGATCCGCC

CCGGGCTCCT

TTCCGCACCG

CTGCAAGGAC

TGCTTGGCTA

CGGACGGCCG

CGACGTGCCC

TTTAACGGTG

ATGATTGACA

GTTTCTTCTC

CGAATGACTT

AGTTCGATCT

TGGACACCGG

TCGCAACCAT

ACTATCCCCC

GGGTGTCCTG

ATCAGCATCT

AGAAACTCTG

TCGCCGATCC

GACCTCTGCA

CGCCCCGGGG

ACCTCGAGGT

GCGATCCCGA

CGGTGGCACA

CGACGCTATC

AGGACATCCT

CCGCAACGCA

TGGGGTTCGT

ACTTCGAGGG

ACGACGAGTT

TTGACCTCGA

GCAGCGGCGA

CGTGGTTCAA.

GGCACACCCC

TTTAAAGGAC

GGCCGGCAAT

TCTGTCGGGG

GTTGTTTCTG

TGTCGGCCTG

GGCCGCGGTG

GTATCACGTT

GAACGTCGTC

GGAACACGAC

GAACAGCTGG

CGCGAAGGTG

GGTACCGCGT

TCGGCGCTTC

GATGCAGGCC

TCAGACGAAA

GGAACGACTG

CAGTCACACC

GCGGGATCTG

CAGCGAAGAG

TCCTCAGTGG

CGGCGCCGCG

CGTCGACCAG

540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2156 2204 a a a 9 a. a.

a 9 a *aaaa.

a AGTACCAGGG CAACACTCTG AACCTTC ATG ACA AGC Met Thr Ser 1 CGC GTC GAC CCC GCA AAC CCC GGT TCA GAA CTC GAT TCC GCC ATC CGC Arg Val Asp Pro Ala Asn Pro Gly Ser Glu Leu Asp Ser Ala Ile Arg

GAO

Asp

GCA

Ala

CTC

Leu

TAC

Tyr

CGG

Arg CG C Arg 100

GAO

Asp

CTG

Leu

ACG

Thr

CTT

Leu

AGCC

Ser 180

ACC

Thr

GCG

Ala

TGG

Trp

ACA

Thr

TCG

Ser

AGC

Ser

ACC

Thr

GCA

Ala

CAG

Gln

CTT

Leu

CGC

Arg

CTG

Leu

GAG

Glu 165

AGT

Ser

GTC

Val

GTG

Val

GC

Ala

CTG

Leu

GAA

Glu

GGC

Gly

CGT

Arg

CCT

Pro

GGC

Gly

GCC

Ala

GC

Gly

GTA

Val1 150

CAC

His

CCT

Pro

AAG

Lys

TTG

Leu

GGG

Gly 230

ACC

Thr

TCG

Ser

CAG

Gin

TTT

Phe

GGG

Gly

TTC

Phe

GGA

Gly

CAG

Gln 135

GCG

Ala

GGT

Gly

GGT

Gly

GGT

Gly

GCC

Ala 215

GAG

G iu TAC AGC Tyr Ser 25 GGC TTC Gly Phe CAG GGC Gln Gly GGG GGT Gly Gly 0GC AOG Arg Thr TTT GTC Phe Val 105 CGT OGA Arg Arg 120 CTG GGC Leu Gly CTG GGC Leu Gly GAA CTG Glu Leu GTC GAT Val Asp 185 GCG GAC Ala Asp 200 AGC GGA Ser Gly TTG CAA Leu Gin

AAC

Asn

CTC

Leu

ACG

Thr

GAG

Glu

CGT

Arg 90

OGC

Arg

ATC

Ile

GAO

Asp

TOG

Ser

GGT

Gly 170

GTT

Val

CTO

Leu

GAO

Asp 000 Ala

TGO

Oys

GAO

Asp

GTT

Val

ATO

Ile 75

OTA

Leu

GAO

Asp

GO

Gly

TAO

Tyr

TGG

Trp 155

GTG

Val 000 Pro

TTT

Phe

GTT

Val1

ACC

Thr 235

COG

Pro 000 Ala

CAT

His 60

COG

Pro

CTO

Leu

GAO

Asp

GTO

Val

CTO

Leu 140

GAG

Giu

GAO

Asp

GOT

Ala

COO

Pro

GAO

Asp 220

GG

Gly -77-

GTA

Val 000 Ala 45

TTO

Phe

CCA

Pro

GGC

Gly

AGO

Ser

TOG

Ser 125

GAG

Glu 000 Ala

GAO

Asp

GAG

Glu

OAT

Asp 205 000 Ala 000 Ala 000 AAC Pro Asn 30 GO ATC Gly Ile ACC TAO Thr Tyr CTG OTO Leu Leu ATO ACC Ile Thr OOG ATO Pro Ile 110 GOC TOG Ala Ser TTG OAT Leu Asp CGO 000 Arg Ala GTO GAG Val Oiu 175 CAG CTC Gin Leu 190 GTO GC Val Ala CTG TAO Leu Tyr CGO CCA Arg Pro

GOT

Ala

GAA

Giu

GAO

Asp

AGO

Ser

COG

Pro

ACA

Thr

GCA

Ala

COO

Pro

TTG

Leu 160

CTG

Leu

GAA

Giu

CGC

Arg

AOT

Ser

GTG

Val 240

CTG

Leu

CTC

Leu

CAG

Gin

GAG

Glu

CTO

Leu 000 Ala

ATT

Ile

TG

Trp 145

TTG

Leu

GTG

Val1

GAA

Glu

GT

Gly

TG

Trp 225

GTG

Val

OTO

Leu

GAO

Asp

COT

Pro 000 Gly

TTG

Leu 000 Ala cGC Arg 130

CG

Arg

CAC

His

CG

Pro

TOG

Ser

CAG

Gin 210

CTG

Leu

GAT

Asp

ACG

Thr

GTO

Val

GC

Ala

TTG

Leu

GG

Gly 000 Ala 115

ATO

Ile

CAA

Gin

ACC

Thr

ATO

Ile 000 Ala 195

CC

Ala

COO

Pro

CTO

Leu 2252 2300 2348 2396 2444 2492 2540 2588 2636 2684 2732 2780 2828 2876 2924 4 *444*a a 000 OTO OAT GAG 000 AAT 000 TAO 000 AOT GTG TOG ACO GTO AGO AGO Gly Leu 245 Asp Giu Arg Asn Al a 250 Tyr Ala Ser Val Trp 255 Thr Val Ser Ser -78-

GGG

Gly 260

CC

Ala CTG GTT CGC CAG Leu Val Arg Gin

CGA

Arg 265

CTG

Leu CCT GGC CTT GTT Pro Gly Leu Val

CAA

Gin 270

CAT

His CGA CTG GTC GAC Arg Leu Val Asp

GCG

Ala 275 GTC GAC GCC Val Asp Ala

GGG

dly 280

GCG

Ala TGG GCA CGC Trp Ala Arg

GAT

Asp 285

TCG

Ser TCC GAC GCG Ser Asp Ala GTG ACC Val Thr 290 AGC CTC CAC Ser Leu His GGC TTC GCC Gly Phe Gly 310 GAC GCC CTC Asp Ala Leu

GCC

Ala 295

CC

Ala AAC CTG GCC Asn Leu Cly

GTA

Val 300

CGT

Arg ACC GGA CCA Thr Cly Ala GAC TTC CAG Asp Phe Gin

CAG

Gin 315

CGC

Arg CTG GTT CCA Leu Val Pro

CGC

Arg 320

CTG

Leu GTA CCC CAG Val Gly Gin 305 CTG CAT CAC Leu Asp His CTC ACC AAC Leu Thr Asn GCC CTC CTG Ala Leu Leu 325 AAC TTG Asn Leu

GAG

Glu 330

GTC

Val1 ACA CAd CAA Thr Gin Gin

TTC

Phe 335 CTG CAG GAA Leu Gin Giu 340

TTT

Phe

CCC

Pro 345

CTC

Leu CCC CTC GAT Ala Leu Asp CAG TGG CC GCT CCC GAA Gin Trp Ala Ala Pro Glu 350 355 TAGGAACATC CGCATGACAC CTG AAC AAC Leu Asn Asn

ACC

Ser 360 AAT CCC CAC Asn Arg His

CGA

Arg 365 C C

C

C. C C.

TCTCACCTGA

TTG CCCGTGC

GTTCGGCAAC

CGCGCGAATA

TCGCGGCAGC

TGATCGAACT

ACAACTGGTG

TCAGCGCCAC

GCGGCGCCAA

AGCAGGGTGC

ACGACTGGGC

TCAAGGTCGA

AATCCGAGCG

TGGGGATCC

CCTGGACACC

GTGAGTTCAC

AAAGCAGCAC

GCTAGCCGAA

AGCCGAGCGC

CGGCGGCTGG

CGATGGATCT

GATCGGCTCG

GACCGGAAAT

CCCGACCGAA

GGGGTCGGAC

GATCATTCCT

CGCCATCGGC

GCCTGACGAA

CGGCAGCCTC

GCACGCCGCA

GGCCGGTATT

CATCGCATTG

GTTCGACCAC

AAGTGGCGAG

GAAGACCTGC

GGCGCAGACT

TTGGGACACC

CAGGAACAAG

GCCTCCAGCG

GACGGCGGCT

CTGCTGTTCG

GCCGCTATCC

ATCCCACA

GTGCTGCGCG

TTCGCGCCCA

CTCGATGCCG

CAACAGGCAA

CAGGGAGCTG

GCGACGCCGC

CCACCCCCGT

GCGCGAGCGC

GCCCACCC

TGTTCGGATA

AAGAACACCT

AGAACAACAG

ACGTGCTCAA

TGTTCGGCGT

CGACATCGCG

CCGACAGCGC

CGCCCAACGC

TAGCGCAATT

CCAGCACTA

CCCACCATCC

ACGCCGCCGC

CGACAACGAT

CGACCGTGAT

GCTGCTGTCG

CATCGAGGTC

CCACCTCACC

GTACACCCAG

CCACCTGCTC

TGGCACGAAG

CGTCCAGGAT

GGCTCGCGTT

TTCCACGGAC

CTTCGTTCTC

GATCTTCGCC

CACCCGTACC

CTACACCATC

CCGTGAAGCG

CCCGTCGCGC

CGCGCCGCGGG

CTCCTCGTCC

GTCCGCGAAA

AACCCCCCGA

ATCCCGCAGA

GACTGGAAGG

CACTTCTGCA

GATTCTCCC

ACGCCCAACG

TTCCACAACG

GCCTTCATAC

AACGTCTATC

CAGCCGAGC

CGCTCCTACG

CCCCACCTC

2972 3020 3068 3116 3164 3212 3262 3322 3382 3442 3502 3562 3622 3682 3742 3802 3862 3922 3982 4042 4102 4162 4222 4282 TGCAGACGGT GTGGGACAAG GGCGACGCGC TCACCCCCGA GGACCGCGGC GAACTGATGG -79-

TGAAGGTCTC

TCTTCGAGG7

GCAACGTGCG

AGCACACCTT

TGATCGAGGC

TCCATACGTA

GGCGACCTCC

GAGCAGCTTG

GTTGACCGCC

CTGCGACGTG

GATTCGATCG

TCCAGGAGCA

TAATCTACCT

ATTTCCCG CA

TTGGCCTGCT

GCCTTCCAGA

CGTCGGCATG

AGCGCGACGA

TTCTCCTG CT

GTGCCCGAGA

CGGGTGATCG

GGGAGTCAAA

GATCGGCGCG

CACCCACTCC

GAACGGTCAA

CGAGGCCACC

CGGGTGCACA

GTTTCGCTTG

CCTTTGTGGA

GCCAGCGGGC

ACCCAGGGCC

GATTCGTAGT

CGTCGGCGTC

CGAAGGCCTT

TCGCCGCGGT

TGTGAAACCA

ACCCCAGCGC

AGCGCAGCAG

GCTCCTTGCG

CCAGGCGGAC

GATCCCGGTC

TCGAGGCCGA

GCGTTGGCCA

CGCGGAACAC

CTGCACGACC

TACCCGATTC

GCGCGGCCGA

CAAGGAGATA

ATTCGTCGGG

ACACCGCGCC

TTGTGCGGCG

CTTGGTGACC

GCGCAAGTGG

GTCGACGATC

GATCGCGACC

CGCACCTGGA

GCAGCGCTGT

CCCATCACCG

ATCAGCCCAC

GCCGTCGGCG

ATTGAGATGG

GGCGAAGGCC

CAGCCCGGCA

*CCAACGCCGC

*ATCCCAGGTA

CGGTGTCCTA

CCGGCTTCAC

GTCGCGAATC

TTGTCAAGAC

CTCAGCGGGT

GGCACGGACC

TCGATCAGCT

TTGGTTCGCA

ATCCAGTGCT

TTGGCGACCG

TGGGCCTTGT

TGAGCCGACT

TCACGGGTAT

ACGGCCAGCA

GACTCTGTGG

CGGACGCCGA

ATGCCGTCGA

TTCGCCTCGT

CCAGTGCCGA

CCTCAACATC

CGGTTTCGAC

CAAGATCGCC

CTCCTGAGGA

GCCCGCCGAT

CTGTGGATGA

GAGATGTCGA

AGCGCGACCA

TGTAGGCCAT

ACCGCACGGT

CGGCCdGGTA

CCTTGGGGTA

CGATGTCCTC

TGGGCAGCGC

CCGGAAACAC

CCGGGGCGGT

ACTCCCGGAA

TCATCACGAG

CCCATAGGTA

CCTGCCATTG

GGAACTGCTC

AGCAGCGGCG

CGCTTCTGGC

GACGTCGGCA

TCTGAGGCGC

ACTCAGCTTC

GGGTGTTTCA

TGGGTCGTTC

GATGGGGGGC

GGCAATCCCG

CGCAAACGTC

CCGGTAGAAC

CTTCGCGCCG

GGCGTTGTAG

AGCAAGCACA

CTCCCGCAGT

CATCCCGCGG

CCCATCGGTG

CAAGCACAGC

CACGAAATCG

CGCGGTCAGC

CAA

4342 4402 4462 4522 4582 4642 4702 4762 4822 4882 4942 5002 5062 5122 5182 5242 5302 5362 5422 5482 5535 p p p a p.

a p p a p

PPP.

p 0P p a. p.

a p 9 INFORM4ATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 365 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID Met Thr Ser Arg Val Asp Pro Ala Asn Pro Gly Ser Glu Leu Asp Ser 2 5 10 Ala Ile Arg Asp Thr Leu Thr Tyr Ser Asn Cys Pro Val Pro Asn Ala 25 Leu Leu Thr Ala Ser Glu Ser Gly Phe Leu Asp Ala Ala Gly Ile Glu 35 40 Leu Asp Val Gin Pro Ala Glu Gly Leu Leu Leu Gly Ala Ala Ala 115 Ile Arg Ile 130 Trp Arg Gin 145 Leu Tyr Arg Arg 100 Asp Leu Thr Ser Thr Ala Gin Leu Arg Leu Gly Arg 70 Pro Gly Ala Gly Val Gin 55 Phe Gly Phe Gly Gin 135 Ala Gin Gly Arg Phe Arg 120 Leu Leu Gly Gly Thr Va1 105 Arg Gly Gly Thr Glu Arg 90 Arg Ile Asp Ser Val His 60 Ile Pro I 75 Leu Leu Asp Asp I Gly Val S

I

Tyr Leu G 140 Trp Giu A 155 Phe Pro ;ly ier er lu la Thr Leu Ile Pro 110 Ala Leu Arg Tyr Leu Thr Ile Ser Asp Ala Asp Ser Pro Thr Ala Pro Leu 160 Leu His Thr Leu Glu His Gly Giu Leu *c 9 .9 a. a 9 *9*a 4 4** a Va Git Gly Trp 225 Val Val Vai Ala Va1 305 Leu Leu Ala L Pro 1 Ser Gin 210 Leu Asp Ser Asp Val 290 ly 4 Asp Thr 2 Pro C Ile Ala 195 Ala Pro Leu Ser Ala 275 Thr Ser 180 Thr Ala Trp Gly Gly 260 Ala Ser 165 Ser Val Val Ala Leu 245 Leu Val Leu Pro Lys Leu Gly 230 Asp Val Asp His Gly 310 Leu Leu Asn 2 Gly Gly Ala 215 Glu Glu Arg Ala Ala 295 Val Ala 200 Ser Leu Arg Gin Gly 280 Ala Asp 185 Asp Gly Gin Asn Arg 265 Leu Asn Gly 170 Val Leu Asp Ala Ala 250 Pro Trp Leu Val Pro Phe Val Thr 235 Tyr Gly Ala Gly Asp Ala Pro Asp 220 Gly Ala Leu Arg Val 300 Asp Glu Asp 205 Ala Ala Ser Val Asp 285 Ser Val Gin 190 Val Leu Arg Val Gin 270 His Thr Glu 175 Leu Ala Tyr Pro Trp 255 Arg Ser Gly Leu Glu Arg Ser Va1 240 Thr Leu Asp Ala Gln Gly Phe Ala Asp Phe Gin Gin Arg Leu Val Pro Arg 315 320 iis ,sn ,lu 355 Asp Asn 340 Phe Ala 325 Leu Leu Ala Aln ksn Leu Glu Ser 360 Leu Pro 345 Leu Glu 330 Val Asn Arg Ala Arg Thr Leu His Gin Asp Arg 365 Gin Gin 350 Phe 335 Trp Leu Ala

Claims (15)

1. A method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule of Rhodococcus origin wherein said transformed microorganism expresses a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the microorganism mixture under conditions sufficient to bring about the catalytic cleavage of organic carbon-sulfur bonds, whereby the Io organic sulfur content of the fossil fuel is significantly reduced.

2. The method of claim 1 wherein the fossil fuel is petroleum.

3. The method of claim 1 wherein the recombinant DNA molecule is derived from a strain of Rhodococcus sp. ATCC 53968.

4. A method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: 2 (ORF-1), comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, S: whereby the organic sulfur content of the fossil fuel is significantly reduced.

5. A method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule 2: 5 which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: 3 (ORF-3), comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and S: b) incubating the fossil fuel and the transformed microorganism mixture under 0s conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, whereby the organic sulfur content of the fossil fuel is significantly reduced.

6. A method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic A [lI:\DayLib\L.IBA]2872.doc:TLT 82 sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID NO: (ORF-2), comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under Sconditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, whereby the organic sulfur content of the fossil fuel is significantly reduced.

7. A method of desulfurizing a fossil fuel which contains organic sulfur molecules through the use of a transformed microorganism containing a recombinant DNA molecule which encodes a biocatalyst capable of desulfurizing a fossil fuel which contains organic sulfur molecules, wherein the biocatalyst comprises the protein set forth in SEQ. ID. Nos: 2 (ORF-1), 3 (ORF-3) and 5 (ORF-2), comprising the steps of: a) contacting the fossil fuel with the transformed microorganism; and b) incubating the fossil fuel and the transformed microorganism mixture under conditions sufficient to bring about the oxidative cleavage of organic carbon-sulfur bonds, is whereby the organic sulfur content of the fossil fuel is significantly reduced.

8. A method for desulfurizing dibenzothiophene, comprising the step of contacting dibenzothiophene with a transformed microorganism containing a recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more reactions in the desulfurization of dibenzothiophene, wherein said recombinant nucleic acid molecule has the sequence of a nucleic acid molecule isolated from Rhodococcus, or a complement of said isolated nucleic acid molecule.

9. A method for desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a recombinant nucleic acid molecule which encodes an enzyme which catalyzes the 2: conversion of dibenzothiophene to dibenzothiophene sulfone, wherein the enzyme comprises an active fragment of the enzyme having the amino acid sequence shown in ORF-3 (SEQ ID NO: 3).

10. A method for desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more steps in the conversion of dibenzothiophene sulfone to 2 -hydroxybiphenyl, wherein the enzymes comprise an active fragment of the enzyme having the amino acid sequence shown in ORF-1 (SEQ ID NO: 2).

11. A method for desulfurizing an organic sulfur compound, comprising the step of contacting the organic sulfur compound with a transformed microorganism containing a K' N ~1 [I:\DayIb\LIBA]2872.doc:TLT 83 recombinant nucleic acid molecule which encodes one or more enzymes which catalyze one or more steps in the conversion of dibenzothiophene sulfone to 2-hydroxybiphenyl, wherein the enzymes comprise an active fragment of the enzyme having the amino acid sequence shown in ORF-2 (SEQ ID NO:

12. A method for desulfurizing an organic sulfur compound by contacting the organic sulfur compound with a transformed microorganism, substantially as hereinbefore described with reference to any one of the examples.

13. A fossil fuel desulfurized by the method of any one of claims 1-7.

14. Desulfurized thiophene when desulfurized by the method of claim 8. 0

15. A desulfurized organic sulfur compound when desulfurized by the method of any one of claims 9-12. Dated 8 November, 1999 Energy Biosystems Corporation Patent Attorneys for the Applicant/Nominated Person j SPRUSON FERGUSON *O S 0 Of [I:\DayLib\LIBA]2872.doc:TLT S So IJ S [I :\DayLib\LIB1A]2872.doc:TLT