The invention relates to the field of producing phenol from renewable sources, such as e.g. biomass in a suitable recombinant host.

Introduction

Phenol is current ly produced at several million tonnes per year from fossil raw materials, predominantly by the cumene process, i.e. a chemical process. Such fossil raw materials are not renewable as opposed to raw materials which are renewable, such as the renewable resource "biomass". A production process based on renewable resources, such as biomass, would achieve independence from fossil resources and would have the potential of saving large amounts of COj emissions.

Phenol is a chemical i ntermed iate used in i ndustry i n the product i on of phenolic resins, bisphenol A caprolactame and other chemicals. Phenol that is based on renewable resources, which is referred to here in the context of the invention as "biophenol", is strongly desired in order to reduce production cost and become independent of fossil resources. More importantly, chemical companies have committed themselves to reduce CO.; emissions both for their own processes as well as by increasing the use of renewable resources in their raw materials. Biophenol has a high potent ial of avoiding fossil resources and saving CO.> emissions, accordingly.

The production of chorismate in bacteria was described in Sprenger GA, "From scratch to value: engineering Escherichia coli wild type cells to the production of [.-phenylalanine and other fine chemicals derived from chorismate", Appl Microbiol Biotechnol 2007, 75 :739-749.

The product ion of phenol in bacteria has been described by Wierckx NIP et al., "Engineering of solvent-tolerant Pseudomonas putida S 12 for bioproduction of phenol from glucose", Appl Environ. Microbiol. 2005. 71 :8221-8227. However, the biosynthesis pathway used by Wierckx et. a I first builds up the molecule tyrosine from chorismate by incorporating an amino group from glutamate, and then breaks it down to phenol releasing the amino group again. This pathway comprises more reaction steps, and is energetically less favouorable than the method according to the invention due to the longer pathway which leads to the deaminaiion of glutamate. Using tyrosine as an intermediate for the biosynthesis of phenol is also a disadvantage because tyrosine is a strong inhibitor of the activity of the genes in the aromat ic amino acid pathway. The inhibition takes place at genome leve I (gene repression), transcriptome level and metabolome level (feedback inhibition). In contrast, the method according to the invention converts chorismate directly to phenol, as described below, which is less complex, energetically more efficient and has a higher potential of yielding high production rates of phenol.

In addition to the above pathway over tyrosine to produce phenol in bacteria, it is also known that phenol can be produced by chemical conversion of shikimic acid, wherein the shik imic acid is produced by fermentat ion (Gibson JM et a I., Benzene- Free Synthesis of Phenol, Angew. Chem. 2001 113(10): 1999-2002).

The biosynthesis of 4-hydroxybenzoate over tyrosine was also reported. Meinen et al. used the biosynthesis pathway over tyrosine (Meijnen J et al., Improved p-hydroxybenzoate production by engineered Pseudomonas putida S 12 by using a mixed-substrate feeding strategy. Appl Microbiol Biotechnol. 201 1. 90(3):885-93.).

WO2012063862 describes a method for producing phenol by fermentation. The pat hway over chorismate and 4-hydroxybenzoate was used by introducing the genes for a chorismate- pyruvate lyase and a 4-hydroxybcnzoatc decarboxylase in a Corynebacterium glutamicum strain. However, pheno l is toxic for Corynebacterium glutamicum and this strain stops to grow at low concentrations of phenol. The grow th phase and the product ion phase are therefore separated in a two-step process, wherein the second step is performed in a different medium than the first and the redox potential is lowered to -450 mV. Combined growth and production using only one fermentation vessel is not possible according to the inv ent ion reported in WO2012063862. For the same reason it is not possible to run a continuous fermentation with in-situ product removal where the biomass is regenerating itself by growth. Definitions

The term "host" within the meaning of the invention can comprise any host that is capable of producing chorismate, either naturally, only after transformat ion, or in addition to the natural iy present chorismate following transformation. A "host" according to the invention can be selected from the group consist ing of bacteria, yeast and fungi.

The term "genetic modification" within the meaning of the invention can comprise deletions as well as transformations. For example, a genetic modification can be a de leti on or a transformation that causes the host to overproduce chorismate. Such ov erproduction of chorismate in the host can be achieved by introducing one or more genetic modifications in the host. Accordingly, the host can comprise one or more genetic modifications to overproduce chorismate. These genet ic modifications can have the effect that the host is prod uc i ng chorismate at levels that are elevated, above the normal, endogenous physiological levels that are present in the host by nature.

The term "transformat ion ^" within the meaning of the invention comprises plasmid transformation as well as chromosomal transformation. I n plasmid transformation the transformed DNA is uncut and circular and therefore is held ext rac hro moso ma I ly in the host. I n chromosomal trans format ion the transformed DNA is cut and therefore linear and can thus be integrated into the chromosomal genome of the host to be transformed. Essentially, for chromosomal transformation, t he host can be transformed with a short piece of linear DNA that can be integrated into the chromosomal genome of the host to be transformed. The term "in-situ product recov ery ^" within the meaning of the inv ention refers to the remov al of phenol directly from the fermentation broth by using a suitable technique, while the fermenter broth is continued to be used in the ferment at ion. This may include circulating the fermentation broth through an external apparatus where the phenol is removed from the fermentation broth before the fermentation broth is partly recycled to the fermenter. The cel ls may or may not be retained in the fermenter in this case. Another option is to remov e the phenol from the fermentation broth while the fermenter broth remains in the fermenter.

Description The invention relates to a method for whole cell biosynthesis of phenol from bio mass as the starting material. Typically a source containing a signi ficant proportion of fermentable sugars can be used in the method according to the invent ion. These sugars can include polysaccharides such as di-saccharides, e.g. saccharose, or tri-saccharides, e.g. kestose, as well as C-6 monosaccharides such as glucose, fructose or mannosc and C-5 monosaccharides suc as xylose and arabinose. A microbial strain, preferably a bacterial strain or a yeast strain, that is capable of converting sugar to phenol would enable the production of phenol from a w ide range of renewable resources including sugar beet and sugar cane, starch-containing plants suc as corn, wheat and rye, as well as lignocellulose e.g. from straw, wood or bagasse.

Given the major disadvantages of the above methods known in the art that usually arc less efficient in energet ic terms and also more complex with regard to the synthesis of phenol, there has been a need in the art for an improved method for producing phenol. It has therefore been the problem of the invention to provide a method for producing phenol from renewable sources that avoids the disadvantages of methods known in the art that are less efficient in energetic terms and more complex with regard to the synt esis of phenol.

The invention has solved said problem by providing a met hod of generating a recombinant host strain for producing phenol as described herein. The invention has further solved said problem by prov iding a recombinant host strain capable of producing phenol as described herein. The inv ention has further solved said problem by providing a method of producing phenol in the recombinant host strain. The invention has further solved said problem by providing a method and associated recombinant strain comprising an oxygen-tolerant hydroxybenzoate decarboxylase that is not sensitive to oxygen and is fully active at normal redox condit ions, thus allow ing phenol production at aerobic conditions. The invention has further solved said problem by providing a host strain that is resistant to phenol, thus allow ing combined growth and phenol production at phenol concentrations high enough to enable production by cont inuous fermentation with the opt ion of in-situ product recovery.

In particular, the inv ent ion has solved said problem by prov iding a method of generating a recombinant host strain for producing phenol, comprising the steps of: a) prov iding a host comprising chorismate (CHO), b) transforming said host with a first nucleic acid sequence comprising ubiC ( SEQ ID NO: 1) encoding chorismate lyase, and c) transforming said first transformant with a second nucleic acid sequence encoding an oxygen- tolerant 4-hydroxybenzoate decarboxylase, thereby generating a recombinant host that is capable of producing phenol under aerobic conditions,

wherein step b) and step c) are carried out simultaneously or sequentially.

Step a) of the method can make use of chorismate (CHO) that is present in the host. Chorismate is a key intermediate in the implemented pathway (see Figure 1 and Figure 2). It is a shared precursor for all three aromatic amino acids and is therefore naturally present in ail organisms capable of producing aromatic amino acids, which includes all common microorganisms. Intracellular chorismate can therefore be produced from all fermentable sugars.

Step b) of the method provides the host with a first nucleic acid sequence, preferably a gene, the product of which converts chorismate ( CHO) to 4-hydroxybenzoate (4-HB). ubiC (SEQ I D NO: 1) encodes the enzyme chorismate lyase that converts chorismate to 4-hydroxybenzoate (4-H B ), thereby generating a recombinant host that overexpresses chorismate lyase (see Figure 2 and Figure 3). Thus, in a preferred embodiment of the invention, the first nucleic acid sequence is SEQ ID NO: 1.

Step c) of the method according to the invent ion addit ionally provides t he host with a second nucleic acid sequence, preferably a gene cluster, the product of w hich converts 4- hydroxybenzoate (4-H B ) to phenol, by introducing a nucleic acid into the host encoding an oxygen-tolerant 4-hydroxybenzoate decarboxylase (see e.g. Figure 2 and Figure 3). Said nucleic acid in step b) can comprise the gene cluster hbdBCD (SEQ ID NO 2) that expresses 4- hydroxybenzoate decarboxylase. The gene cluster hbdBCD of step c) can be derived from the E.coli strain E. coli 01 1 1 : B4. Thus, in a further embodiment of the method according to the invention, said second nucleic acid sequence comprises the gene cluster hbdBCD, as defined in SEQ ID NO 2. The enzyme encoded by SEQ I D NO: 2 is a 4-hydroxybenzoate decarboxylase that is oxygen-tolerant. Thus, in a preferred embodiment of the invention, said second nucleic acid sequence is SEQ ID NO: 2. Thus, the invention has solved the above problem by prov iding a method of generating a recombinant host strain for producing phenol, comprising the steps of: a) prov iding a host comprising chorismate (CHO), b) transforming said host with a first nucleic acid sequence comprising ubiC (SEQ ID NO: 1) encoding chorismate lyase that converts chorismate ( CHO) to 4-hydroxybenzoate (4-HB), and c) transforming said first transformant with a second nucleic acid sequence encoding an oxygen- tolerant 4-hydroxybenzoate decarboxylase that converts 4- yd ro xy be nz a t e (4-H B) to phenol, thereby generating a recombinant host that is capable of producing phenol under aerobic condit ions,

wherein step b) and step c) are carried out simultaneously or sequentially.

In a preferred embodiment of the invention, said first nucleic acid sequence is SEQ ID NO: 1. In a further embodiment of the method according to the invention, said second nucleic acid sequence comprises the gene cluster hbdBCD, as defined in SEQ ID NO 2. The enzyme encoded by SEQ I D NO: 2 is a 4-hydroxybenzoate decarboxylase that is oxygen-tolerant. This, in a preferred embodiment of the invent ion, said second nucleic acid sequence is SEQ I D NO: 2.

Thus, the min imum requirements for producing phenol in a recombinant host according to the invention are the presence of chorismate in the host and the first nucleic acid sequence comprising ubiC (SEQ I D NO: 1) and the second nucleic acid sequence encoding an oxygen- tolerant 4-hydroxybenzoate decarboxylase that preferably is hbdBCD, as defined in SEQ I D NO: 2. The chorismate present in the host can be the endogenous chorismate that is produced naturally by the host, or it can be chorismate that is overproduced by the host, if said host comprises one or more genetic modificat ions to overproduce chorismatc.

The first and second nucleic acid sequences transformed into the host in steps b) and c) can be on the same plasmid, on different plasm ids or on the chromosome (chromosomal integrat ion, e.g. Example 5). If the first and second nucleic acid sequences transformed into the host are on the same plasmid then step b) and step c) of the method can be carried out simultaneously (e.g. see Example 1). If the first and second nucleic acid sequences transformed into the host are on different plasmids then step b) and step c) of the method can be carried out sequentially. For example, the gene ubiC (SEQ I D NO: 1) can be transformed on a first plasmid into the host, and the gene cluster hbdBCD ( SEQ I D NO: 2) can be transformed on a second plasmid into the host (see e.g. Example 3). In one embodiment, the host strain is transformed with ubiC and hbdBCD that are present on the same plasmid (e.g. see Example 1, pJFl 19 ubiChbdB CD). In another embodiment, the host strain is transformed with ubiC (SEQ I D NO: 1) on the first plasmid pJ F l 19 and with hbdBCD ( SEQ I D NO: 2) on the second plasmid ACYC (e.g. sec Example 3).

The technical advantage of the method of the invent ion over the prior art methods described above is that the economic feasibility of phenol production in a large scale production facility is improved. The invent ion allows a one-step conversion of sugar into phenol in a single vessel using continuous fermentation with in-situ product removal at aerobic conditions. This improves the sugar yield and the space time yield significantly. Compared to a fed-batch fermentation a cont inuous fermentation has a much better overall sugar yield since less sugar is needed to generate bio mass which in case of a fed-batch fermentation needs to be generated for every new batch. The overall space-time yield is improved since no time is lost between the production phases as would be the case in a fed-batch fermentation ( e.g. for harvesting the product, cleaning and sterilizing the fermenter, generating the biomass). Furthermore, the one-step conversion allows production using only one fermentat ion vessel. That reduces the complexity of the process and the capital expenditure for the production facility. Compared to the synthesis pathway over tyrosine reported by Wierckx et a I., the implemented synthesis pathway is less complex, more energy efficient and avoids large intracellular concentrations of tyrosine which would inhibit the biosynthesis pathway to phenol. As a result the method according to the invention is much more efficient, since it is able to achieve a better sugar yield ( due to the energy efficiency) and a better space-time-yield (due to the shorter patliway and the avoidance f tyrosine as an intermediate), as compared to the methods known in the art.

In a further embodiment of the method according to the invention the host of step a) can overproduce chorismate (CHO). Such overproduction of chorismate in the host can be achieved by introducing one or more genetic modifications in the host. Accordingly, in a further embodiment of the method of the invention, the host can comprise one or more genetic modifications to oveiproduce chorismate. These genetic modifications have the effect that the host is producing chorismate at levels that are elevated, above the normal, endogenous physiological levels. Since more substrate is provided for the subsequent reactions in step b)

(chorismate, CHO, to 4-hydtoxybcnzoate, 4-HB, by the ubiC gene product ) and in step c) (4-HB to phen l by the hbdBCD gene product ), more end product, i.e. phenol, is produced.

Such one or more genetic modifications can comprise a deletion of one or more oityrR, pheA and tyrA that can be introduced into the host.

The TyrR protein, encoded by the gene tyrR, represses the expression of several of the genes in the common part of the aromatic amino acid pathway by binding to recognition sequences referred to as TyrR boxes. The TyrR protein is modulated by the presence of aromatic amino acids. In particular, the presence of tyrosine and ATP allows it to self-associate into a hexamer w hich can also bind to w eak TyrR boxes some of w hich overlap the promotors of the genes in the aromatic amino acid pathw ay. In some cases the mechanism of repression involves exclusion of the RNA polymerase from the promotors, w hile in others it interferes with the ability of bound RNA polymerase to form open complexes or to exit the promotors. By deleting tyrR the regulatory effects caused by TyrR can be avoided completely, as shown in Figure 2. The deletion f tyrR can be achieved by using the λ red recombinase according to the protocol by Datsenko and Wanner, as described in Example 1. Here, tyrR can be deleted by using the tyrR: :FRT-kan cassette that is shown in SEQ ID NO:3, and as described in Example 1.

The gene pheA encodes for a bifunctional enzyme which catalyses the conversion of chorismate to prephenate (chorismate mutase) as well as the conversion of prephenate to kcto- phcnylpyruvate. The gene tyrA also encodes for a bifunctional enzyme which also catalyses the conversion of chorismate to prephenate (chorismate mutase) as well as the conversion of prephenate to 4-hydroxyphenylpyruvate (prephenate dehydrogenase). By deleting both the pheA and the tyrA gene the pathway fr m chorismate to phenylalanine and tyrosine can be completely inactivated since all chorismate activity is removed as well as the prephenate dehydratase and the prephenate dehydrogenase activity, as shown in Figure 2. Deletion of pheA and tyrA can be achieved by using the pheAtyrA: :FRT-CAT cassette that is shown in SEQ ID NO:4, as also described in Example 1.

In further embodiments of the invention, one, two or all three of tyrR, pheA and tyrA can be deleted in the host strain used. In a preferred embodiment of the invention, all three of the genes tyrR, pheA and tyrA are deleted in the host (AtyrRpheAtyrA), so that chorismate is overproduced. One example of such a recombinant strain carrying all three deletions is E.coti BW25113 AtyrRpheAtyrA that is listed in Table 1. The generation of the strain E. coli BW25113 AtyrR SpheAtyrA is described in Example 1. In a further embodiment of the method said one or more genetic modifications to overproduce cliorismate can comprise a transformation with one or more ofaroG (SEQ ID NO: 9), aroG ^fbr (SEQ ID NO: 10), aroB (SEQ ID NO: 11) and aroL (SEQ ID NO: 12). The host can be transformed individually with each one ofaroG (SEQ ID NO: 9), aroG ^fbr (SEQ ID NO: 10), aroB (SEQ ID NO: 11) and aroL (SEQ ID NO: 12). The host can also be transformed with each combination of roG (SEQ ID NO: 9), roG' ^?,r (SEQ ID NO: 10), aroB (SEQ ID NO: 11) and aroL (SEQ ID NO: 12) so that an optimal overproduction of chorismate is achieved.

The reactions catalysed by the gene products ofaroG (SEQ ID NO: 9), aroC "(SEQ ID NO: 10), aroB (SEQ ID NO: 11) and aroL (SEQ ID NO: 12) are depicted in Figure 2. They all result in an overproduction of chorismate in the host.

The gene product ofaroG (SEQ ID NO: 9) catalyses the reaction from E4P to DAHP, as shown in Figu e 2. aroC " ^' (SEQ ID NO: 10) encodes for the same enzyme, except for a G to A mutation which makes the enzyme resistant to feedback inhibition, as reported by Kikuchi et al (Kikuchi, Y., Tsujimoto, K., Kurahashi, O. (1997) Applied and Environmental Microbiology 63 761-762) and shown in Figure 6.

The gene product of am B (SEQ I D NO: 11) catalyses the reaction from DAHP to 3DQ, as shown in Figure 2. The gene product oiaroL (SEQ I D NO: 12) catalyses the reaction from SH I to SH I P, as shown in Figure 2.

The transformation of one or more of these genes results in an overproduction of chorismate in the host. Thus, in one embod iment of the method of the invention the host can comprise one or more genetic modifications to overproduce chorismate, w herein said one or more genetic modificat ions can be a transformation with one or more oiaroG (SEQ I D NO: 9), aroCf" ^" (SEQ I D NO: 10), aroB (SEQ I D NO: 1 1) and aroL ( SEQ I D NO: 12).

The transformation of the host with one or more ofaroG ( SEQ I D NO: 9), aroG ^Jbr (SEQ I D NO: 10), aroB ( SEQ I D NO: 1 1) and aroL ( SEQ I D NO: 12) can be done as a single transformation step, wherein the genes that are transformed into the host are on the same piasmid. However, these transformations can also be performed in such a way that the genes that are introduced into the host are on separate plasmids or integrated directly on the chromosome.

In a further embodiment of the method said one or more genetic modifications that can be present in the host can comprise a deletion of one or more oityrR, pheA and tyrA and can further comprise a transformat ion with one or more of roG ( SEQ I D NO: 9), aroG^ ^r ( SEQ I D NO: 10), aroB ( SEQ I D NO: 1 1) and aroL ( SEQ I D NO: 12).

I n a part icularly preferred embodiment of the method, the host of step a) is subjected to deleting all three of the genes tyrR. pheA and tyrA thereby generating the host AtyrRpheAtyrA, preferably E.coli BW251 13 AtyrRpheAtyrA that is listed in Table 1 and described in Example 1, so that chorismate is ov erproduced. The host AtyrRpheAtyrA can subsequently be transformed with ubiC ( SEQ I D NO: 1) in step b) and with hbdBCD ( SEQ I D NO: 2) in step c), simu latenously or sequent ially, thereby generating a host AtyrRpheAtyrA transformed with ubiC and hbdBCD. One example of such a strain is E.coli BW251 13 AtyrRpheAtyrA transformed with ubiC and hbdBCD, as listed in Table 1, and as further described in Example 1 (E. coli

BW25113 AtyrR ApheAtyrA).

In a further part icularly preferred embodiment of the method, the host of step a) can be subjected to deleting all three of the genes tyrR, pheA and tyrA, thereby generating the host AtyrRpheAtyrA, so that chorismate is overproduced, which is then transformed with aroL (SEQ ID NO: 12 ) and with ubiC (SEQ ID NO: 1) in step b) and with hbdBCD (SEQ ID NO: 2) in step c) of the method according to the invention, thereby generating a host AtyrRpheAtyrA transformed with aroL, ubiC and hbdBCD. One example of such a strain is E.coli BW25113 AtyrRpheAtyrA transformed with aroL, ubiC and hbdBCD, as listed in Table 1, and as further described in Example 2.

In a particularly preferred embodiment, the genetic modification to overproduce chorismate comprises a transformation with aroC " ^' ( SEQ ID NO: 10), aroB ( SEQ ID NO: 1 1 ), aroL ( SEQ I D NO: 12), ubiC ( SEQ I D NO: 1) and hbdBCD ( SEQ I D NO: 2). Thus, one particularly preferred embodiment o the method according to the inv ention can generate the recombinant strain AfyrR ApheAtyrA transformed with aroG ^fbr (SEQ ID NO: 10), aroB (SEQ I D NO: 1 1), aroL ( SEQ I D NO: 12) and ubiC ( SEQ ID NO: 1) and hbdBCD ( SEQ I D NO: 2).

In another particularly preferred embodiment, the genet ic modification to overproduce chorismate comprises a transformation with aroC ^Jbr ( SEQ I D NO: 10), ubiC (SEQ ID NO: 1) and hbdBCD ( SEQ ID NO: 2). Thus, one particularly preferred embodiment of the method according to the inv ention can generate the recombinant strain AtyrR ApheAtyrA transformed with aroG ^fbr ( SEQ I D NO: 10) and ubiC ( SEQ I D NO: 1) and hbdBCD (SEQ I D NO: 2).

In another particularly preferred embodiment, the genetic modification to overproduce chorismate comprises a transformation with aroCr ¹ " ^" ( SEQ I D NO: 10), aroL (SEQ ID NO: 12), ubiC ( SEQ I D NO: 1) and hbdBCD ( SEQ I D NO: 2). Thus, one particularly preferred embodiment of the method according to the invention can generate the recombinant strain AtyrR ApheAtyrA transformed with aroG ^/br (SEQ I D NO: 10), aroL (SEQ ID NO: 12) and ubiC (SEQ I D NO: 1) and hbdBCD (SEQ I D NO: 2).

The host that can be used for the method according to the invention can be selected from the group consist ing of bacteria, yeast and fungi. In a preferred embodiment, the bacterium is an Escherichia coli strain. The Escherichia coli strain can be selected from the group consisting of f. coli BW25 1 13. E. coli DHlOb, and E. coli I J 1 10. In a particularly preferred embodiment of the method according to the invention, E.coli BW25113 AtyrRpheAtyrA is used, as listed in Table 1 and as described in Example 1.

In a further embodiment of the method according to the invention the host can be a phenol- resistant host, preferably a phenol-resistant bacterium, more preferably a phenol-resistant Pseudomonas putida strain, more preferably Pseudomonas putida S12 and most preferably Pseudomonas putida S12 ApheApobA.

The transformation steps b) and c) that are performed in the method according to the invention can comprise plasmid transformation or chromosomal transformation. In plasmid transformation the transformed DNA is uncut and circular and therefore is held extrachromosomally in the host. In cliromosomal transformation the transformed DNA is cut and therefore linear and can thus be integrated into the chromosomal genome of the host to be transformed.

The invention further prov ides a recombinant host strain obtainable by the method according to the invention, as described above.

In one embodiment, the recombinant host strain comprises chorismate and further comprises a first nucleic acid sequence comprising ubiC (SEQ ID NO: 1) and a second nucleic acid sequence encoding an oxygen-tolerant 4-hydroxybenzoate decarboxylase, wherein the recombinant strain is capable of producing phenol under aerobic condit ions. I n a further embodiment of the host strain according to the invention said second nucleic acid sequence comprises hbdBCD, as defined in SEQ I D NO: 2.

In a further embodiment, the recombinant host strain can overproduce chorismate. Such overproduction of chorismate in the host can be achieved by introducing one or more genetic modificat ions in the host. Accordingly, in a further embodiment, the host strain can comprise one or more genetic modifications to overproduce chorismate. These genetic modifications have the effect that the host is producing chorismate at a higher rate than normal. Since substrate is provided at a higher rate for the subsequent reactions in step b) (chorismate, CHO, to 4- hydroxybenzoatc, 411 B, by the ubiC gene product ) and in step c) (4-hydroxybenzoate, 411 B, to phenol by the hbdBCD gene product ) the end product, i.e. phenol, is produced at a higher rate.

In one embodiment, the recombinant host strain comprises one or more genetic

modifications, wherein said genetic modification comprises a deletion of one or more oityrR, pheA and tyrA. These genes and their gene products, as well as their deletion, have been described above.

In further embodiments of the invention, one, two or all three oityrR, pheA and tyrA can be deleted in the recombinant host strain of the invention. In a preferred embodiment of the invention, ail three of the genes tyrR, pheA and tyrA are deleted in the host, so that chorismate is overproduced (AtyrRpheAtyrA). One example of such a recombinant strain carrying all three deletions is E.coli BW251 13 AtyrRpheAtyrA that is listed in Table 1, and as described in Example 1.

In a further embodiment, the recombinant host strain comprises one or more genetic modifications, wherein said genetic modification comprises a transformation with a nucleic acid sequence comprising one or more of aroG ( SEQ I D NO: 9), aroG^" ^" ( SEQ I D NO: 10), aroB ( SEQ I D NO: 1 1) and aroL (SEQ I D NO: 12). These genes and their gene products have been described above. In a further embodiment, the recombinant host strain comprises one or more genetic modifications, wherein said genetic modification comprises a deletion of one or more oityrR, phe.4 and tyrA; and further comprises a transformation with one or more oiaroG (SEQ ID NO: 9), aro(/"(SEQ ID NO: 10), aroB (SEQ ID NO: 11) and aroL (SEQ ID NO: 12). These genes and their gene products have been described above.

in a particularly preferred embodiment of the recombinant host strain of the invention, the genetic modification to overproduce chorismate comprises a transformation with aroG^' ^" (SEQ ID NO: 10), aroB (SEQ ID NO: 11), aroL (SEQ ID NO: 12), ubiC (SEQ ID NO: 1) and hbdBCD (SEQ I D NO: 2). Thus, one particularly preferred embodiment of the recombinant host strain according to the invention is the recombinant strain AtyrR ApheAiyrA transformed with aroG ^Jbr (SEQ ID NO: 10), aroB (SEQ ID NO: 11). aroL (SEQ ID NO: 12) and ubiC (SEQ ID NO: 1) and hbdBCD (SEQ ID NO: 2).

In another particularly preferred embodiment of the recombinant host strain of the invention, the genetic modification to overproduce chorismate comprises a transformation with aro " ^" (SEQ ID NO: 10), aroL (SEQ ID NO: 12), ubiC (SEQ ID NO: 1) and hbdBCD (SEQ ID NO: 2). Thus, one particularly preferred embodiment of the recombinant host strain according to the invention is the recombinant strain AtyrR ApheAtyrA transformed with aro '" ^" (SEQ ID NO: 10), aroL (SEQ ID NO: 12) and ubiC (SEQ ID NO: 1) and hbdBCD (SEQ ID NO: 2).

In another particularly preferred embodiment of the recombinant host strain of the invention, the genetic modification to overproduce chorismate comprises a transformation with aroG ^fbr (SEQ ID NO: 10), ubiC (SEQ ID NO: 1) and hbdBCD (SEQ ID NO: 2). Thus, one particularly preferred embodiment of the method according to the inv ention can generate the recombinant strain AtyrR ApheAtyrA transformed with aroG ^fbr (SEQ ID NO: 10) and ubiC (SEQ ID NO: 1) and hbdBCD (SEQ ID NO: 2).

In a further embodiment, the recombinant host strain can be selected from the group

consisting of bacteria, yeast and fungi. In a preferred embodiment, the bacterium is an Escherichia coli strain. In a further embodiment, said Escherichia coli strain can be selected from the group consisting of E. coli BW251 13, E. coli DHlOb, and E. coli LJ110. In a particularly preferred embodiment of the recombinant host strain according to the invention,

E.coli BW25113 AtyrRpheAtyrA is used, as listed in Table 1, and as described in Example 1.

In a further embodiment of the recombinant strain according to the invention, said host can be a phenol-resistant host, preferably a phenol-resistant bacterium, more preferably a phenol-resistant Pseudomonas putida strain, more preferably Pseudomonas putida S12 and most preferably Pseudomonas putida S12 ApheApobA.

The deletion oipheA in Pseudomonas putida, as indicated by ApheA, inactivates the conversion of chorismate to prephenate (the chorismate mutase reaction, CHO to PREPH, see Figure 2) thereby inact ivat ing the pathway to phenylalanine and tyrosine. Un like E. coli,

Pseudomonas putida does not possess two isoenzymes for the chorismate mutase reaction, so it is sufficient to delete the pheA gene in Pseudomonas putida to inactivate the tyrosine and phenylalanine pathway. The gene pobA catalyses the conversion of 4-hydroxybenzoate to protocatechuate (4-hydroxybenzoate hydroxylase) which enables Pseudomonas putida to degrade 4-hydroxybenzoate. Delet ing pheA and pobA, as ind icated by ApheApobA, thus increases the availability of 4-hydroxybenzoate for phenol production, as the competing pathways are inact iv ated. E. coli does not hav e a hydroxybenzoate hydroxylase, so there is no corresponding deletion to be made if E. coli is used as the host.

The inv ent ion further prov ides a method of producing phenol in a recombinant host comprising the steps of a) provid ing a recombinant host strain accord ing to the invention, as described above, and b) incubating said recombinant host strain under fermentation conditions thereby producing phenol.

I n a further embodiment of the method according to the inv ent ion method, the phenol production can be induced. Such induct ion phen l production can be in the absence or in the presence of oxygen (O2). Phenol product ion can be induced by the presence or the absence of a specific chemical compound or by a change in a physical condition. For instance the presence of an inducer may activate the transcription of certain genes in the biosynthesis pathway to phenol (e.g. IPTG acting on the lac operon to express genes located on a pi asm id ), or the absence of, for instance, tyrosine may activate the expression of genes involved in the aromatic amino acid pathway from which phenol is derived. A change in a physical condit ion such as temperature, pH or 0.> concentration may also activate the expression of genes involved in phenol synthesis.

In a further embodiment, the method of producing phenol in a recombinant strain can further comprise the step c) of harvesting the produced phenol from the recombinant host strain.

Step b) of the method of producing phenol in a recombinant host according to the invention can be performed as a batch fermentation, as a fed-batch fermentation or as a continuous fermentation.

In the context of the invention, batch fermentation refers to a fermentation method in which the complete fermentation medium is provided at the start of the fermentation. The product is harv ested at the end of the fermentation (i.e. phenol).

In the context of the inv ent ion, fed-batch fermentation refers to a fermentation method in which a part of the fermentation medium is provided at the start of the fermentation, and a part is fed to the fermenter during the fermentation. The product is harvested at the end of the fermentation (i.e. phenol).

In the context of the invention, continuous fermentation refers to a fermentation method in which substrate is added and the product (i.e. phenol) is removed continuously during the fermentation.

In a further embodiment of the method of producing phen l in a recombinant strain, the fermentation conditions of step b) can comprise aerobic conditions. That means that both the reactions in the shake flasks as well as the fermenter can be performed under aerobic conditions. Such aerobic conditions can be implemented e.g. by gassing the shake flasks and/or fermenter with air. I n a further embodiment of the method of producing phenol in a recombinant strain, the fermentation cond it ions can comprise the presence of a raw sugar cane juice, wherein said raw sugar cane juice can preferably comprise a high concentration of 1 -kestose. One example of this embodiment is shown in Example 6. Such sugar cane juice can be used as the substrate in such fermentation and can comprise, e.g glucose 14 g/1, fructose 24 g/1, sucrose 130 g/1, kestose 1 19 g/1 and nystose 5 g/1, as measured by 11 PLC.

It will be apparent to those skilled in the art that various modifications can be made to the methods and recombinant host strains of the invention. Thus, it is intended that the present invention covers such modifications and variations, provided they come within the scope of the appended claims and their equivalents.

Figures, Tables and Sequences

Figure 1 shows the biosynthesis pathway from glucose to phenol.

E4P = erythrose-4-phosphate

3 D = 3-dehydroquinate

3 DS = 3-dehydroshikimate

SHI = shikimate

SH I 3 P =shikimate-3-phosphate

ESH I3 P = 5-enolpyruvyl-shikimate-3 -phosphate

CHO = chorismatc

411 B = 4- li yd roxybe nzoat e

PR EPH = prephenate

Figure 2 shows the metabol ic engineering by the method of the invention of the cel lular reaction network in a host in order to create a recombinant host strain according to the invention that overproduces phenol. Figure 3 shows the final two react ions in the phenol synthesis pathway that are catalysed by the ubiC (SEQ I D NO: 1) gene product ( CHO to 4-H B ) and by the hbdBCD ( SEQ I D NO: 2) gene product hydroxybenzoic acid decarboxylase ( 4-H B to phenol). Figure 4 shows the HPLC analysis of the fermentation broth, as described in Example 1.

Figure 5 shows the HPLC analysis of the fermentation broth, as described in Example 2.

Figure 6 shows the genes aroG (SEQ I D NO: 9), aro f" (SEQ I D NO: 10), aroB ( SEQ I D NO: 1 1), aroL ( SEQ I D NO: 12) and ubiC ( SEQ I D NO: 1), which were all synthesised with restriction sites ( underlined ) for Bglll and Barnl l I to allow integration into the plasmid. The ATG start codon and the TAA stop codon are in bold. The G to A feedback resistance mutation in aroG ^fbr ( SEQ I D NO: 10) is bold and underl ined. Figure 7 shows the gene cluster hbdBCD (SEQ ID NO: 2), which was isolated from E. coli 01 1 1 : B4 and integrated into the same plasmid that is already carrying ubiC by the appropriate restriction enzymes plasmid. The gene cluster has the composition hbdB: 0.6 kbp, hbdC: 1.4 kbp, hbdD: 0.2 kbp, as indicated. Figure 8 shows the HPLC analys is of the fermentat ion broth, as described in Exam le 3.

Figure 9 shows the HPLC analysis of the fermentation broth, as described in Example 5.

Figure 10 shows the HPLC analysis of the fermentation broth of shake flasks, as described in Example 6.

Table 1 shows a list of the E.coti strains, plasmids and genes used according to the invention. Table 2 shows the results of Example 1 and Example 2.

SEQ ID NO: 1 shows the sequence oiubiC. The native gene ubiC, of which this sequence was derived, is found in the NCBI data base under the accession number CP000948, position 4350225 to 4350722. ubiC encodes c orismatc lyase, which catalyses the reaction from chorismate (CHO) to 4-hydroxybenzoate (4-HB), as shown in Figure 2 and in Figure 3.

SEQ ID NO: 2 shows the sequence of the gene cluster hbdBCD derived from E. coli

01 11 :B4. The gene cluster has the composition hbdB: 0.6 kbp, hbdC: 1.4 kbp, hbdD: 0.2 kbp, as depicted in Figu re 7. The sequence can be found in the NCBI data base under the accession number NC 01 364 position 3434167 to 343 1 06. The gene cluster encodes 4- hydroxybenzoate decarboxylase that catalyses the reaction from 4 - h y d ro xy be nz at c (4-HB ) to phenol, as shown in Figure 2 and Figure 3.

SEQ ID NO: 3 shows the sequence of the tyt R:: FRT-kan cassette, which was used to delete tyrR.

SEQ ID NO: 4 shows the sequence of the pheAtyrA: : FRT-C AT cassette, which was used to delete pheA tyrA .

SEQ I D NO: 5 shows the sequence of the knockout primer tyrR, the 5' primer, which as used for deleting tyrR. SEQ I D NO: 6 shows the sequence of the knockout primer tyrR, the 3 ' primer, which as used for deleting tyrR.

SEQ I D NO: 7 shows the sequence of the knockout primer pheAtyrA, the 5' primer, which as used for deleting pheAtyrA.

SEQ I D NO: 8 shows the sequence of the knockout primer pheAtyrA, the 3' primer, which as used for deleting pheAtyrA.

SEQ ID NO: 9 shows the sequence oiaroG. The native gene aroG, of which this sequence was derived, is found in the NCBI data base under the accession number CP000948, position 837448 to 838500. The aroG gene product which catalyses the reaction from E4P to DAHP, as shown in Figure 2. SEQ I D NO: 10 shows the sequence of aroG ^fh . The sequences as per SEQ I D NO: 7 differs from SEQ ID NO: 9 by the fact that G is changed to A at position 436, thereby generating the fbr (feedback resistance) mutant ofaroG. SEQ ID NO: 1 1 shows the sequence oiaroB. The native gene aroB, of which this sequence was derived, is found in the NCBI data base under the accession number CP000948, position 3613165 to 3614253. The aroB gene product catalyses the reaction from DAMP to 3DQ, as shown in Figure 2. SEQ I D NO: 12 shows the sequence iaroL. The native gene aroL, of which this sequence was derived, is found in the NCBI data base under the accession number CP000948, position 344960 to 345484. The aroL gene product catalyses the reaction from SH I to SHI3P, as shown in Figure 2. SEQ I D NO: 13 shows the sequence of an insertion cassette with a Ptac promotor and a ribosome binding site upstream of the aroBaroG ^fbr sequence, and a FRT flanked

chloramphenicol resistance downstream of the aroBaro f" ^' sequence as well as a transcription terminator, as described in Example 5. SEQ I D NO: 14 and SEQ I D NO: 15 show the sequences of the amplification primer pair that was used to amplify the insertion cassette, as shown in SEQ I D NO: 13.

SEQ ID NO: 16 shows the sequence of the cloning site on the chromosome after integration of the cassette in the fuc locus between the fucP and fuel genes, as described in Example 5. This sequence includes the chromosomal DN.A sequences directly upstream and downstream of the cassette. The cassette is found between position 5148 and 91 12 of SEQ ID NO: 16.

SEQ I D NO: 17 and SEQ I D NO: 18 show the sequence of the primer pair used for testing for successful chromosomal integration of the cassette, as described in Example 5. Mutants with defects in the fuc locus are not able to grow on L-fucosc as a carbon source and will form pale colonies on MacConkey medium containing 1% fucose (contrary to wild type ceils which will form red colonies, as described by Albermann et ai. 2010). After selection on agar plates containing chloramphenicol, the positive colonies were further tested on MacConkey medium with 1% fiicose. The fncose negative colonies were further tested with by PGR using primers SEQ I D NO: 17 for the 5'-test and S EQ ID NO 18 for the 3 '-test.

Examples

Example 1: Creating the recombinant strain E. coli BW25113 AtyrR ApheAtyrA

pj F 1 19i(biChbdBCD and using it for phenol production by fermentation A novel biosynthesis pathway for synthesizing phenol from sugar v ia chorismate (CHO) was identified (see Figure 1 and Figure 2). This pathway was implemented in an E. coli strain and the further genetic modifications necessary to create a phenol producing strain were performed (see Figure 2). In this way the whole cell biosynthesis of phenol was made possible. The created recombinant strain produced between 0,5 mM and 1 ,5 mM phenol in shake flask experiments and up to 5 mM in a 1 litre bioreactor fermentation. A list of the E.coli strains, piasmids and genes used is provided in Table 1.

The following genetic modifications were performed to overproduce chorismate in addition to the endogenous chorismate of the host strain E. coli BW25113 : the genes pheA and tyrA were deleted in order to remove the chorismate mutase reaction that consumes chorismate so that chorismate is overproduced. These deletions make the strain auxotrophic towards phenylalanine and tyrosine. In addition, the regulatory gene tyrR that encodes an aporepressor for the expression of the Tyr regulon was deleted. The corepressor of tyrR is either tyrosine or phenylalanine plus tryptophan. a. Creating the host strain E. coli BW25113 AtyrR ApheAtyrA

The knock out deletion of the genes pheA, tyrA and tyrR genes was performed by using recombinat ion by the phage λ red recombiiiase according to the method of Datsenko and Wanner ( Datsen ko, K.A. , Wan ner, B.L., One-step inactivat ion of chromosomal genes in Escherich ia col i K- 1 2 using PGR products (2000), PNAS, 97 6640-6645). pheA and tyrA are located right next to each other on the chromosome and were inactivated in one step. tyrR is located on a different place on the chromosome and was inactivated in a second step by the same method. The insertion cassettes used for the disruptions contained an antibiotic resistance gene flanked by FRT ( flippa.se recognition target) sites. Sequences homologous to regions adjacent to the gene to be inactivated were located at either end of the cassettes as described by Datsenko and Wanner. The insertion cassettes were amplified by PGR (polymerase chain reaction) from template plasmids with the antibiotic resistance gene and the FRT sites. The PGR primers contained the homologous regions and a priming site. The insertion cassette pheAtyrA::¥KT- CAT used for pheAtyrA disruption had a chloramphenicol resistance and is given by SEQ ID NO: 4. The primers used for the amplification of the pheAtyrA knockout cassette (plie A t y r A : : F R T - C A T cassette, SEQ I D NO: 4) are given by SEQ ID NO: 7 and SEQ ID NO: 8. The plasmid pCO l -FRT-GAT was used as template plasmid for the pheA yrA : : FRT-G AT cassette. The insertion cassette n vA'::FRT-kan used for tyrR disruption had a kanamycin resistance and is given by SEQ I D NO: 3. The primers used f r the amplification of the tyrR knockout cassette ( vrR::FRT-kan, SEQ I D NO: 3) are given by SEQ I D NO: 5 and SEQ I D NO: 6. The plasmid pCOl-FRT-kan was used as template plasmid for the m7?:: FRT-kaii cassette.

The tyrR ::FRT-kan cassette was integrated into the chromosome of :, coli BW251 13 by the λ red recombinase, thus yielding the strain E. coli BW25113 AtyrR. Cells in which the disrupt ion had been successful were selected on agar plates containing kanamycin. Furthermore, control PGR was carried out to demonstrate the successful disruption. In the same way, the cassette pheA tyrA : : F RT-C AT was integrated into the chromosome of E. coli BW251 13 AtyrR by the λ red recombinase, thus yielding the strain E. coli B W25113 AtyrR ApheAtyrA (see Table 1). Cells in which the disruption had been successful were selected on agar plates containing chloramphenicol. Control PGR was again used to demonstrate the successful disruption of pheAtyrA. In addit ion, the phenylalanine and tyrosine auxotrophy of the created mutants was checked. The mutants were only able to grow in the presence of phenylalanine and tyrosine. In both chromosomal integrations the helper plasmid pKD46 was used to express the λ red recombinase. The antibiotic resistances were removed from E. coli B W25113 AtyrR ApheAtyrA by expressing a flippase using the helper plasmid pCP20. b. Creating the strain E. coli BW25113 AtyrR ApheAtyrA pJFl ubiChbdBCD

The plasmid pj F 1 1 (Furste, J. P., Pansegrau, W . , Frank , R . , B locker. H .. Scholz, P..

Bagdasarian, M., Lanka, E. (1986) Molecular Cloning of the Plasmid RP4 Primasc Region in a Multi-Host-Range tacP Expression Vector. Gene 48 1 19- 131) was chosen as a vector for the over-expression oi ubiC and hbdBCD.

The gene ubiC was designed with 5 ' restrict ion sites for Ndel and Bgli l and 3 ' restrict ion site for Bamll l and synthesized by the company Geneart ( part of Life Technologies ). The deliv ery plasmid pMA was amplified in E. coli DH 10b and the gene cut off by Ndel and Bamll l restriction enzymes. Preparative agarose gel electrophoresis was used to purify the gene. A clon ing site on pJF I 1 was opened by a sequential digest with first Ndel and then Baml l . The ubiC gene and the vector were ligated by T4 ligase (see Figure 6 and SEQ ID NO: 1). The successful integrat ion of the ubiC gene on the pj F 1 19 plasmid was checked by digesting pj F 1 1 9 hiC by Ndel, Baml l l d igestion and analyzing the DNA fragments by agarose gel electrophoresis. The protein expression profile was checked by expressing pJFI \9ubiC in E. coli DH 10b and subsequent SDS-PAGE analysis of the cell extract. The activity of the gene product UbiC was demonstrated by incubat ing raw enzyme extracts of : ^' . coli DH 10b pJ F I \ 9 hiC with chorismate and measuring the resu lt ing 4-hydroxybenzoate product ion by HPLC. Finally, the plasmid pJF I 19wWC was sequenced by the company Qiagen and the detected sequence was aligned to the original ubiC gene sequence and controlled for homology. All analyses confirmed the successful integration of ubiC into pJ F I 19.

The gene cluster hbdBCD was amplified by PGR from the chromosome of the strain E. coli O l l l : B4 (ATCC 33780) using primers with restriction sites for H indl l l EcoRI. The primer product was then incorporated on the plasmid pUC19 by digesting pUC19 with H indl l l/EcoR I and ligating the primer product with the opened pUC19 using T4 ligase (see Figure 7 and SEQ ID NO: 2).

The plasmid p\JCl9hbdBCD was then digested with Drdl. The result ing DNA fragment of approx. 3 kb contained the hbdBCD gene cluster. This fragment was purified by preparative agarose gel electrophoresis. The fragment was incorporated into pJF 119ubiC by digesting this vector with Drdl and ligating the fragment with the vector using T4 ligase. The correct cloning of hbdBCD was checked by digesting p J F 1 \9ubiChbdBCD with bgil l and Rsrl l and separating the result ing fragments using agarose gel electrophoresis. The observed bands matched the expected DNA fragments. The activity of the gene product HbdBCD was demonstrated by incubating raw enzyme extracts of E. coli DH 10b pJF l \ 9hhdBCD with 4-hydroxybenzoate and measuring the resulting phenol production by HPLC.

The pJF l 1 piasmid contains an ampiciilin resistance which is used for selection. A second variant of the piasmid pJF 1 19uhiChhdBCD was made by exchanging the ampiciilin resistance for a kanamycin resistance. This was done by digesting pJ F l 1 9hhdBCD with BspHI to remove the ampiciilin resistance and then ligating the linear pJF1 19hbdBCD fragment with a gene for kanamycin resistance. Finally. pJF l 19 hiChhdBCD was transferred into E. coli BW25113 AtyrR ApheAtyrA by electroporation to create the strain E. coli BW25113 AfyrR ApheAtyrA pJF l \ 9uhiChhdBCD (see Table 1). Both a variant with ampiciilin resistance on the piasmid as well as a variant with kanamycine resistance on the piasmid was created. c. Producing phenol from sugar with the strain E. coli BW25113 AfyrR ApheAtyrA pJF119ubiChbdBCD

E. coli BW25113 AtyrR ApheAtyrA pJF 1 19uhiChhdBCD with ampiciilin resistance was grown in a 10 ml shake flask culture and in a 1 1 bioreactor / fermenter, both under aerobic conditions. Such aerobic conditions were implemented by gassing the shake flask and/or bioreactor / fermenter with air.

The fermentation medium used for the shake flask culture was based on an E. coli medium published by Riesenberg et a I ( Riesenberg D., Schulz V., Knorre, W.A., Poll I, 11- D., Korz, D., Sanders E.A., Ro, A., Deckwcr, W-D. (1991) High cell density cultivation of Escherichia c li at controlled specific growth rate. Journal of Biotechnology 20 17-27). It contained the following compounds: 13.3 g/L KH ₂ P0 ₄ , 4 g/L (NHj b Oj. 1.7 g/L Citrate, 2.5 g/1 Liiria Broth, 1 7.5 g/L Glucose, 0.024 g/L [.-Phenylalanine, 0.016 g/L L-Tyrosine, 10 ml/1 trace clement solution, 0.6 g/L MgS0 ₄ * 7 H >0. 0.2 mM CaCl ₂ * 2 H >0, 0.1 g/1 ampiciilin

The fermentation medium used in the bioreactor contained the following components: 15.5 g/L

KH2PO4, 4.67 g/L (NH ₄ ) ₂ P0 ₄ , 1.98 g/L Citrate, 17.5 g/L Glucose, 0.5 g/L Thiamine, 0.037 g/L L-Phenyialanine, 0.024 g/L L-Tyrosine, 10 ml/1 trace element solution, 0.6 g/L MgSQt * 7 H2O, 0.2 mM CaC!> * 2 H >0, 0. 1 g/ ^' l ampici llin

The trace element solution had the following composition: 75 mg/L Fe(III)Citrat * 1120. 3.75 mg/L H3BO3, 18.75 mg/L Mn(II)Ci ₂ * 4 FLO, 10.5 mg/L EDTA (Titriplex III), 1.88 mg/L CuC12 * 2 H ₂ 0, 3.13 mg/L Na ₂ Mo0 ₄ * 2 H ₂ 0, 3.13 mg/L Co(II)Cl ₂ * 6 lf>(), 10 mg/L Zn Acetate * 2 I ).

In the bioreactor the concentration of dissolved oxygen was controlled at p0 ₂ = 5 % and the pH was controlled at pH = 7.0 by addition of a 25% NFL solution. That means that aerobic conditions were used, wherein the aerobic conditions were implemented e.g. by gassing the shake flasks and/or fermenter/bioreactor with air.

The expression f the genes on pj F 119 was induced by adding I PTG to the culture once the bacteria had reached their exponential growth phase.

As a negative control shake flask fermentations were performed with the strains /:. coli

BW25 1 1 3 AtyrR ApheAtyrA pJF l 1 9Δ and E. coli BW251 13 AtyrR ApheAtyrA

pUC 19hhdBCD. pJF l 1 9 signifies a p J F 1 19 plasmid without any added genes. The same shake flask fermentation medium was used for the negative controls and the fermentation procedure was identical including the addition IPTG.

The shake flask fermentation with E. coli BW25 1 1 3 AtyrR ApheAtyrA pJF l \ 9 hiChhdBCD yielded 1 .4 mM phenol after 24 hours. The fermentations in the bioreactor yielded 4,9 mM phenol after 40 hours. No phenol could be detected in the cultures of the negative controls (see Table 2). The concentrations were determined by HPLC-UV using a gradient method of 30 minutes and UV detection at 280 nm. The phenol peaks in the samples had identical retention times and UV- spectra to a phenol standard solution (see Figure 4). HPLC-MS was used to detect the mass of phenol as a further confirmat ion of phenol production (see Table 2).

Example 2: Creating the recombinant strain E. coli BW25113 AtyrR ApheAtyrA pj F 1 19aroLubiChbdBCD and using it for phenol production by fermentation The host strain E. coli BW251 13 AtyrR ApheAtyrA was developed as described in Example 1 .

The gene aroL ( SEQ I D NO: 12) was designed with a 5' restriction sites for Ndei and Bgl l l and 3 ' restrict ion site for BamHI and synthesized by the company Geneart (part of Life

Technologies), see also Figure 6. It was then cloned in the plasmid pJF l 19 by the same method as described for the cloning oi ubiC in Example 1 to create the plasmid pJF l \ 9aroL (see Table 1). ubiC (SEQ ID NO: 1) was cloned onto the created plasmid pJ F 1 1 9arol. downstream of the awl. gene. pJF l \ 9a rot was digested with BamHI and ubiC was then cut off the delivery plasmid by digesting with Bgll l and BamHI. ubiC was ligated with the opened vector using T4 ligase to yield the plasmid pJFl 19 aroLubiC. The correct integration of aroL and ubiC on pJ F 1 19 was checked by digest ing with Ndei . This yielded two fragments; the aroL gene (0.5 kb) and the vector with ubiC (5.8 kb). A wrong orientation of ubiC would yield different fragments. One fragment would be the entire insert aroLubiC (1.0 kb) and the other the vector without insert (5.3 kb). The insert was also sequenced by Qiagen and the result aligned to the expected sequence. Both analyses confirmed the correct integration oiaroL and ubiC in pJ F l 1 .

The hbdBCD gene cluster (SEQ I D NO: 2) was cloned on the pJF 1 19aroLubiC plasmid by the method described in Example 1 and transformed into the strain E. coli BW251 13 AtyrR ApheAtyrA to create E. coli BW251 13 AtyrR ApheAtyrA pJFl \ 9aroLubiC (see Table 1). The fermentation with E. coli BW251 13 AtyrR ApheAtyrA pJFl \ 9 aroLubiC was done in a shake flask using the same method and the same fermentation medium as described above in Example 1.

This shake flask fermentat ion yielded 0.59 ni phenol after 24 hours. No phenol could be detected in the cultures of the negative controls (see Table 2). The concentrations were determined by H P LC-UV using a gradient method of 30 minutes and UV detection at 280 nm. The phenol peaks in the samples had identical retention t imes and UV-spcctra to a phenol standard solution (see Figure 5). H PLC- MS was used to detect the mass of phenol as a further confirmation of phenol production (see Table 2).

Example 3: Creating the recombinant strain coli BW25113 AtyrR ApheAtyrA p.) F 1 19ubiC pACYChbdBCD and using it for phenol production by fermentation

The strain E. coli BW251 13 AtyrR ApheAtyrA pW \ \ 9ubiC was created as described in Example 1.

The gene cluster hbdBCD was amplified by PGR from the chromosome of the strain E. coli 01 1 1 :B4 (ATCC 33780) and cloned on plasmid pUC19 as described in Example 1. The plasmid pUC 19hhdBCD was digested with EcoRI and H indi 11 to release the hbdBCD gene cluster. This DNA. fragment was purified by preparative agarose gel electrophoresis. The plasmid pACYC was digested with EcoRI and H indi 11. The linearized vector was purified by preparative agarose gel electrophoresis and ligated with the hbdBCD fragment. The resulting plasmid pACYC hbdBCD was transformed into /:. coli DH 10b and selected on agar plates containing chloramphenicol (pACYC contains a chloramphenicol resistance). The correct incorporation of hbdBCD on pACYC was controlled by digesting with EcoRI and H indi I I and analysing the resulting fragments by agarose gel electrophoresis. This analysis confirmed the correct cloning of ^' hbdBCD on pACYC. The plasmid pACYChbdBCD was transformed into the strain E. coli BW251 13 AtyrR ApheAtyrA and a control digestion with EcoRI and H indi 11 was repeated with plasmid from the transformed strain. This analysis again confirmed the correct cloning of hbdBCD on pACYC. The plasmid pj F 1 1 9 hiC, described in Example 1, was transformed into E. coli BW251 13 AtyrR ApheAtyrA pACYChbdBCD to yield the strain E. coli BW251 13 AtyrR ApheAtyrA pJFl \9ubiC pACYChbdBCD.

A fermentation ith E. coli BW251 13 AtyrR ApheAtyrA pJF 1 1 9 hiC pACYChbdBCD was performed in a 1 liter bioreactor. The procedure and the fermentation medium was the same as described in Example I with the exception that the fermentation medium contained 0. 1 g/1 carbon ic i llin. and 0.05 g/1 chloramphenicol instead of ampic ill in. This bioreactor fermentation yielded 2.3 mM phenol after 90 hours. No phenol could be detected in the cultures of the negative controls (see Table 2). The concentrations were determined by HPLC-UV using a gradient method of 30 minutes and UV detection at 280 nm. The phenol peaks in the samples had identical retention times and UV-spectra to a phenol standard solution (see Figure 8). HPLC-MS was used to detect the mass of phenol as a further confirmation of phenol production (see Table 2).

Example 4: Creating the recombinant strain coli BW25113 AfyrR ApheAtyrA

pJFl 19hbdBCDubiC and using it for phenol production by fermentation

In Example 1 the hbdBCD gene cluster was located downstream of the ubiC gene and had its own lac promotor. The ubiC gene had a tac promotor. In this example the hbdBCD gene cluster was moved to a position just before the ubiC gene, right after its tac promotor so that both the hbdBCD and the ubiC gene were expressed by the tac promotor. hbdBCD was cut from the pJF 119ubiChbdBCD construct described in Example 1 with the restriction enzymes Ndel and Hindlll and purified by preparative agarose gel electrophoresis. The pACYC plasmid was cut with the same restriction enzymes, dephosphoryiated and ligated with the hbdBCD gene yielding the construct pACYChbdBCD.

The construct pACYChbdBCD was then digested with EcoRI to release the hbdBCD gene with EcoRI overhangs. The construct pJF 119ubiC described in Example 1 was digested with EcoRI, dephosphoryiated and ligated with the hbdBCD gene to yield the construct pJF 1 19hbdBCDubiC. The expression of hbdBCD after induction with IPTG was

demonstrated by agarose gel electrophoresis.

The plasmid construct p J F 1 1 hhdBCD hiC was transformed in the strain BW2 1 13 AtyrR ApheAtyrA described in Example 1. A shake flask fermentation with IPTG induction performed according to the method described in Example 1 yielded 0.093 mM phenol. Example 5: Creating the recombinant strain E. coli BW25113 ApheAtyrA AtyrR fu : P, _M -aroBaroG ^ft " ^' pJ Fl \ ubiChbdBCD and using it for phenol fermentation

The genes aroB (sequence number) and aro( ^hr (sequence number) were integrated on the chromosome of the strain E. coli BW251 13 ApheAtyrA AtyrR using an insertion cassette with a Ptac promoter and a ribosome binding site upstream of the aroBaroG^ ^1" sequence, and a FRT Hanked chloramphenicol resistance downstream of the aroB roCP" ^' sequence as well as a transcription terminator (chromosomal integration). The sequence for the insertion cassette is given by SEQ ID NO: 13. The sequence for the primers used to ampl ify the insertion cassette is given by SEQ I D NO: 14 (5 '-TGC TGT GCT CAC TGT TIT TTC TIT GGG CGG TAG CCA ATA ACC TTA ACG AC A T I T TAT TA TCA AGG CGC ACT CCC GTT CTG G- 3 ') and SEQ I D NO: 15 (5 '-CAG CAT GGA GGC GAG AGT GAT AAA GTC TGC GCC AAC GTG GCC GAT GOT CAG AAC CCC CAG GGT TAT TGT CTC ATG AGC G-3 ^" ). The phage λ red recombinase method was used to integrate the cassette as described in Example 1. The cassette was integrated in the fuc locus on the chromosome between the fucP and fuel genes and disrupted these. The sequence of the cloning site on the chromosome is given by SEQ I D NO 16. This sequence includes the chromosomal DNA sequences d irect ly upstream and downstream of the cassette. The cassette is found between position 5148 and 91 12. Mutants with defects in the fuc locus are not able to grow on L-fucose as a carbon source and will form pale colonies on MacConkey medium containing 1% fucose (contrary to wild type cells which will form red colonies, as described by Albermann et al. 2010). After selection on agar plates containing chloramphenicol, the positive colonies were further tested on MacConkey medium with 1% fucose. The fucose negative colonies were further tested with by PGR using primers SEQ I D NO: 17 (GGC CT.A TIT CCC TAA AGG GTT TAT TG.A G) for the 5 ^" -test and SEQ I D NO 18 (GACGATACACTTTGGTCTCTTCAACGTTG ) for the 3 '-test. The chloramphenicol resistance was removed by expressing a flippase as described in Example 1, thus creating the strain E. coli BW25 1 1 3 ApheAtyrA AtyrR fuc:

Transformation of the plasmid construct pJF 1 1 uhi( ^' hhdBCD described in Example 1 then yielded the strain E. coli BW25 1 1 3 ApheAtyrA AtyrR fuc: :P _ta c-aroBaroG ^fbr

pJFl \ 9ubiChbdBCD. A shake flask fermentation with IPTG induction was performed as described in Example 1. A phenol concentration of 1 .2 mM was measured by 11 PLC after 24 hours. The HPLC method described in Example 1 was used. The phenol peaks in the samples had identical retention times and UV-spectra to a phenol standard solution (see Figure 9). Example 6: Phenol production by fermentation using a raw sugar cane juice containing a high proportion of kestose

Fermentations with the strain E. coli BW25113 AtyrR ApheAtyrA pJFl \9ubiChbdBCD in shake flasks and in a bioreactor were performed using raw sugar cane juice as the sole energy and carbon source. The sugar cane juice was extracted from a transgenic sugar cane plant containing a high proportion of 1 -kestose as well as sucrose, fructose, glucose and nystose. The concentrations of these sugars in the sugar cane juice had been measured by HPLC and had the following concentrations; glucose 14 g/ ^' l, fructose 24 g/1, sucrose 130 g/1, kestose 1 19 g/1, nystose 5 g/1. Further compounds in the sugar cane juice were not analysed. The juice was then used as energy and carbon source in the Riesenberg medium described in Example 1 instead of glucose. The sugar cane juice was added to yield a total sugar concentration of 15 g/1 in the final medium. Thus the medium for the shake flask fermentation was given as follows; 13.3 g/L

KH2PO4, 4 g/L (Μ¾ ₂ Ρ0 ₄ , 1 .7 g/L citrate, 2.5 g/1 Luria Broth, 0.05 1 3/1 sugar cane juice, 0.024 g/L [ .-phenylalan ine, 0.016 g/L L-tyrosine, 10 ml/1 trace element solution, 0.6 g/L MgSC>4 * 7 H2O, 0.2 mM CaCb * 2 H >0, 0.1 g/1 ampicillin. The medium used for fermentation in the bioreactor was given as follows; 15.5 g/L KH2PO4, 4.67 g/L (NFL 2PO4, 1.98 g/L Citrate, 0.05 1 1/1 sugar cane juice, 0.5 g/L. Thiamine. 0.037 g/L [.-Phenylalanine, 0.024 g/L [.-Tyrosine, 10 ml/1 trace element solution, 0.6 g/L MgS0 ₄ * 7 H ₂ 0, 0.2 mM CaCl ₂ * 2 ILO, 0. 1 g/1 ampicillin. The trace element solution was defined in Example 1.

The fermentation in shake flasks yielded a phenol concentration of 3. 1 mM after complete depletion of all sugars while the fermentation in the bioreactor yielded a phenol concentration of

0,9 mM after depletion of ail sugars. These results demonstrate that it is possible to produce phenol from raw sugar juice containing a mixture of sugars including a high proportion of 1 - kestose. The chromatograms of the phenol I I P [.('-measurement and the UV spectra of the fermentation in shake flask arc given in Figure 10. Table 1

Strain Plasmid Gene Resistance

E. co/ BW25113 AtyrRpheAtyrA pJF1 19 aroB amp, chromosomal /can + CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroGftr amp chromosomal /can + C/IT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroL amp. chromosomal /can + CAT

E. co// BW25113 pJF1 19 ub/C amp, chromosomal /can + GAT

E. co// BW25113 pUC19 bodBCD amp. chromosomal kan+CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroBaroL amp, chromosomal /can + CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLaroB amp, chromosomal /can + CAT

E. coll BW25113 AtyrRpheAtyrA pJF1 19 aroLubiC amp. chromosomal /can + GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 ubiCaroB amp. chromosomal /can + GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 ubiCaroL amp. chromosomal /can + GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroGtbraroL amp, chromosomal /can + GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroGfbrubiC amp, chromosomal /can + GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiCaroB amp. chromosomal /can+GAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroGfbr aroLaroB amp. chromosomal kan+CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroGfbraroLubiC amp, chromosomal /can + CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiCaroBa ro G/&- amp, chromosomal kan+CAT

E. co// BW25113 AtyrRpheAtyrA pJF1 19 ubiChbdBCD amp, chromosomal kan + CA T or kan only

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiChbdBCD amp, chromosomal kan + CA T or kan only

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiChbdBCD amp, chromosomal kan+CATor kan only

E. co// BW25113 AtyrRpheAtyrA pJF1 19 ubiCaroLhbdBCD amp. chromosomal kan + CA Tor kan only

E. co// BW25113 AtyrRpheAtyrA pJF1 19 ubiCaroLhbdBCD amp, chromosomal kan+CA Tor kan only

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiCaroBhbdBCD amp, chromosomal kan + CA T or kan only

E. co// BW25113 AtyrRpheAtyrA pACYC hbdBCD CAT

E. co// BW25113 AtyrRpheAtyrA pACYC /pJF1 19 pj F1 19uD/C, pACYChbdBCD CA T

E. co// BW251 13 AtyrRpheAtyrA pACYC /pJF1 19 pJF1 19aroLubiC, pACYChbdBCD CA T

E. co// BW25113 AtyrRpheAtyrA pACYC /pJF1 19pJF1 ubiCaroL, pACYChbdBCD CA T

E. co// BW25113 AtyrRpheAtyrA pJF1 19 aroLubiCaroBaroGfb, hbdBCD amp. chromosomal kan+CA T

Table 2

Strain Fermentation Cone. Phenol Detection of Experiment device [mMj phenol by

measured by HPLC_MS

HPLC-UV

E. coli BW25113 Shake flask below no Example 1, 2, ApheAtyrAAtyrR detection limit 3

pJFl 1 Δ

E. coli BW25113 Shake flask below Not measured Example I, 2,

ApheAtyrAAtyrR detection limit 3 pUCl 9 hbdBCD

E. CO//BW25113 Shake flask 1.44 yes Example 1

ApheAtyrAAtyrR

p J F 1 \9 biChbdBCD

E. coft BW25113 Fermenter 4.9 yes Example 1

ApheAtyrAAtyrR

pJFl \9ubi hbdB D

E. coli BW25113 Shake flask 0.59 yes Example 2 ApheAtyrAAtyrR

pJFl \ aroLuhi hhdB D

E. cofiBW25113 Fermenter 2.3 yes Example 3 ApheAtyrAAtyrR

pACYChbdBCD

pJF 9ubiC

E. co/z BW25113 Shake flask 0.093 Not measured Example 4 ApheAtyrAAtyrR

pJFl \9hhdB Dubi E. coli BW25113 Shake flask 1.2 Not measured Exam le 5 ApheAtyrA AtyrR

fuc::Ptac-aroBaroG ^fbr

pJFl \ uhi hhdB D

E. co/z BW25113 Shake flask 3.1 Not measured Example 6 ApheA tyrA AtyrR with sugar cane

pACYChbdBCD juice

pJFU9ubiC

E. coli BW25113 Fermenter with 0.9 Not measured Example 6

ApheAtyrAAtyrR sugar cane juice

pACYChbdBCD

pJFl \ biC

"No" signifies that the concentration is below 1 mg/i. "Yes" signifies that the concentration is above 1 mg/1