Field of the invention

This invention relates to a method for cultivating yeast in lignocellulosic hydroly- sate, a yeast host and a yeast culture. In addition the invention relates to method of improving yeast viability and metabolic activity when cultivated on a medium comprising lignocellulosic hydrolysate.

Background of the invention Lignocellulose is the most abundant bio-based material in the world. It is an easily available non-food carbon source. After at least partial hydrolysis to oligo-, di- and monosaccharides it can be used for microbial production of ethanol and other valuable fermentation products.

During lignocellulose hydrolysis formic acid and acetic acid are formed. Formic acid and acetic acid are also commonly used in organosolv processes for solubiliz- ing lignin and hemicellulose. Tolerance to the weak organic acids present in lignocellulosic hydrolysates has gained significant attention because of the development of second generation bioprocesses. Fungal fermentations are strongly affected by inhibitors present in these hydrolysates, including weak organic acids, furans and phenolics. Acetic and formic acid may be present at several grams per litre in lignocellulosic hydrolysates, whereas other weak organic acids, such as levulinic acid, are typically present in much lower concentrations.

Xylose is a major pentose sugar found in lignocellulose. Xylose can be fermented to ethanol using an aldose reductase and xylitol dehydrogenase, or glucose/xylose isomerase. Aldose reductases reduce xylose to xylitol which can often be found as a by-product of ethanol fermentation. Xylitol has known applications in hygiene, health care and nutrition. It can be extracted from various plant materials including birch. Industrial production of xylitol involves direct chemical hydrogenation of xylose to xylitol on a Raney Ni catalyst. Said process has certain drawbacks includ- ing low yield and involvement of several steps including extensive purification. Xylitol production using fungal or yeast aldose reductase has traditionally been carried out using a purified xylose substrate. Xylose in its pure form is expensive and suitable for use in inexpensive chemical hydrogenation using Raney Ni catalyst. Several regulatory pathways for resistance and adaptation towards weak organic acids are known in S. cerevisiae.

Pdr12 is an ATP-binding cassette (ABC) transporter localized in the plasma membrane which has been shown to be involved in resistance towards several weak organic acids. PDR12 is highly induced by moderately lipophilic acids such as sorbic and benzoic acid and also by low pH. However, it was not induced by acetic or formic acid. It has been shown that Pdr12 exported fluorescein from the cytosol by an energy-dependent mechanism, which was inhibited by sorbic or benzoic acid. It was concluded that Pdr12 transported water-soluble weak organic acids with short chain length, its activity leading to exhaustion of intracellular ATP. It has also been reported that Pdr12 is involved in export of aromatic and branch- chained amino acids produced in amino acid catabolism via the Ehrlich pathway.

The inhibition of growth by weak organic acids may result from alterations in the plasma membrane structure, challenges to the cellular energy balance which re- suit from maintaining the intracellular pH, or intracellular accumulation of anions. Exposure to acetic, benzoic or propionic acid decreases the intracellular pH of S. cerevisiae cells. Additionally, high concentrations of acetic or propionic acid reduce cell viability and metabolic activity. It has been concluded that cells adapted to growth in the presence of weak organic acids limit the diffusional entry of the acid. This response mechanism is poorly understood. Cell wall mannoproteins are known to limit the porosity of the yeast cell wall and some cell-wall related proteins such as, Ygp1 and Spi1 play a role in the resistance to organic acids by yeast. The Pdr10 transporter acts as a negative regulator of the microenvironment of Pdr12. Cells lacking Pdr10 have increased resistance to sorbic acid and increased amounts of Pdr12 located in the detergent resistant membrane fraction (lipid rafts). Additionally, exposure to sorbic acid and/or low pH dramatically increased the cellular level of Pdr12, which has been shown to be one of the most abundant proteins in cells adapted to sorbic acid stress.

Acetic and formic acids are the most common acids in most lignocellulosic hydrol- ysates, typically being present at concentrations in the range 27 to 73 mM acetic acid and 30-67 mM formic acid. They inhibit microbial growth and affect cell viability and thereby compromise the process economy. Removal of acetic acid and especially formic acid may be costly and difficult. E.g. distillation may be insufficient because of formic acid - water azeotrope formation. Thus, there is a need for processes and yeast cultures that are viable and meta- bolically active on lignocellulosic hydrolysate and have improved tolerance to acid conditions. There is also a constant need for improving the productivity of fermentation processes for more cost efficient production of products, e.g. for ethanol and xylitol production on lignocellulosic substrate. This invention meets these needs as will be discussed below.

Objects and summary of the invention

The object of the invention is to enhance viability and metabolic activity of yeast in lignocellulosic culture medium and/or culture medium containing weak organic acids, especially formic acid and acetic acid. Further object are to provide yeasts and yeast cultures suitable for use as efficient production hosts when cultivated in n lignocellulosic culture medium and/or culture medium containing weak organic acids. According to the present invention it was surprisingly found that the growth rate on conventional culture medium containing formate or acetate of yeast was improved by deletion of Pdr12 encoding gene. Also better metabolic activity, (i.e. utilization of glucose) and production of ethanol or production of xylitol was observed when the PDR12 gene was deleted from the yeast. The effect of improved growth or metabolic activity is even stronger in strains which are also Cmk1 deficient.

Surprisingly Pdr12 deficient strains showed improved properties also when cultivated in a lignocellulosic hydrolysate containing culture medium.

The first aspect of this invention is a method for cultivating yeast in lignocellulosic hydrolysate. According to the invention said method comprises that said yeast is made deficient of the plasma membrane transporter Pdr12 encoding gene or a closest homologue of said gene in other species of yeast or a fragment of said genes.

The second aspect of this invention is a yeast culture. Characteristic to said culture is that the yeast in said culture is made deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast and cultured in a culture medium comprising lignocellulosic hydrolysate.

The third aspect of this invention is a method for improving yeast viability on medium comprising lignocellulosic hydrolysate. According to the invention said method comprises that said yeast is made deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast.

The fourth aspect of this invention is a method for improving the metabolic activity of the yeast when cultivated on medium comprising lignocellulosic hydrolysate. According to the invention, said method comprises that said yeast is made deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast.

The fifth aspect of this invention is a yeast host. Characteristic to said host is that it is made deficient of a gene encoding plasma membrane transporter Pdr12 or its closest homologue in other species of yeast.

The invention provides several advantages over the prior art methods, cultures and host strains by allowing efficient cultivation and production of valuable products on lignocellulosic hydrolysate, which is readily available and cheap, instead of using starch based raw material which has nutritional value to humans and ani- mals.

Short description of the drawings

Figure 1. Measurement of biomass in Bioscreen cultures of Ctrl (solid triangles), Apdr12 (open squares) and Apdr12 Acmkl (solid squares) grown in SCD-leu me- dium containing 20 g D-glucose ¹ , in the presence of a) 100 mM acetic acid (pH 3.3) and b) 25 mM formic acid (pH 3.1 ). Dashed lines represent SEM of 15 (a) or 20 (b) replicates.

Figure 2. Percentage of viable cells in populations of cells grown in SCD-leu medium containing 20 g D-glucose ¹ , in the presence of 100 mM acetic acid (pH 3.3), expressed as the percentage of colony forming units, CFU, relative to the total cell number determined with a Cellometer Auto T4 cell counter. Cells were pre-grown in SCD-leu medium without an acid addition. Error bars show SEM for 2 biological replicates and reflect the error in both CFU determination and in the estimation of the total cell number. Ctrl (solid triangles), Apdr12 (open squares) and Apdr12 Acmkl (solid squares).

Figure 3. Measurement of biomass as optical density in Bioscreen cultures of cells pre-grown in SCD-leu medium supplemented with 0.45 mM sorbic acid or 50 mM acetic acid. Ctrl (solid triangles), Apdr12 (open squares) and Apdr12 Acmkl (solid squares) cells were pre-grown with 0.45 mM sorbic acid and were grown subse- quently in SCD-leu medium containing 20 g D-glucose I ^"1 , in the presence of, a) 100 mM acetic acid (pH 3.3), b) 120 mM acetic acid (pH 3.3) and c) 25 mM formic acid (pH 3.1 ). Alternatively, the strains were pre-grown in the presence of 50 mM acetic acid and were subsequently grown in SCD-leu medium containing 20 g D- glucose ¹ , in the presence of d) 100 mM acetic acid (pH 3.3) or e) 25 mM formic acid (pH 3.1 ). Dashed lines represent SEM of 4 - 10 replicates. Viability of cells in 100 mM acetic acid after pre-incubation in 50 mM acetic acid is shown in Fig. 3 f.

Figure 4. Growth of yeast strains after incubation for 1 or 2 days in wheat straw hydrolysate. The cells of strains with Apdr12 and Apdr12 Acmkl are able to grow after 2 days, whereas the control strain is not.

Figure 5. Glucose consumption (solid lines) and ethanol production (dashed lines) of S. cerevisiae in a) 95% steam exploded wheat straw hydrolysate or b) 80% birch hydrolysate. Cells were incubated in 50 mM acetic acid prior to inoculation into the hydrolysates. The APdr12 and Apdr12 Acmkl strains consumed more glu- cose, more rapidly than the control. The APdr12 and Apdr12 Acmkl strains produced more ethanol than the control.

Figure 6. The APdr12 strains (open and solid squares) produced more xylitol from xylose in birch hydrolysate (a) or wheat straw hydrolyzate (b), compared to the control strain (open and solid triangles), that was unable to produce xylitol in these conditions.

Definitions

"Lignocellulosic material" means any material comprising lignocellulose. Examples of such materials are hardwood and softwood chips, wood pulp, sawdust and forestry and wood industrial waste; agricultural biomass as cereal straws, sugar beet pulp, corn stover and cobs, sugar cane bagasse, stems, leaves, hulls, husks, and the like; waste products as municipal solid waste, newspaper and waste office paper, milling waste of e.g. grains; dedicated energy crops (e.g. willow, poplar, swithcgrass or reed canarygrass, and the like). Lignocellulosic hydrolysate means at least partially hydrolyzed lignocellulosic material . Preferably lignocellulose of said lignocellulosic material is hydrolyzed to disaccharides and monosaccharides but also simultaneous saccharification and fermentation applications are within scope of this invention. In this connection a phrase "make deficient of " means either a genetic modification of the host to delete or truncate a specific gene or a genetic modification of the host resulting in reduced or no expression of the gene or reduced or no activity of the gene product by any suitable method. By "inactivation" is meant a genetic modification (usually deletion) resulting in complete loss of activity of a gene product. In this invention the gene may also be silenced with a RNA silencing methodology or its expression may be lowered by using promoters enabling low transcription level/low expression.

In this connection phrase "acidic conditions" means pH conditions of pH <4.4; prefarbly <4 and more preferably <3.5 or even < 3.0.

In this connection an increased tolerance is measured by comparing the tolerance of modified and non-modified yeast. Thus tolerance of Pdr12 deficient yeast should be compared to yeast having PDR12 gene that has not been silenced by any means. Increased or enhanced growth is measured by comparing the growth of strains with and without making it deficient of a given gene. In this study growth has been measured by analysing the optical density of the growth media containing the yeast cells. Increased tolerance has been measured by monitoring the growth, and/or viability of the cells, or as metabolic activity (carbon utilization, product for- mation).

By "improved production" is here meant increased amount of product, such as protein, ethanol, xylitol, or organic acid produced when compared to a host which has not been modified in respect of the claimed genes.

In this connection metabolic activity of yeast is estimated by measuring sugar con- sumption during cultivation. Alternatively metabolic activity can be estimated by product yield, e.g. ethanol or xylitol production.

Increased tolerance means that viability or metabolic activity of a genetically modified yeast strain is higher compared to the respective non-modified strain when cultivated in the presence of formic or acetic acid or in lignocellulosic hydrolysate containing medium. Detailed description of the invention

We investigated the role of Pdr12 in resistance towards two most common weak organic acids which are present in lignocellulosic hydrolysates; acetic acid and formic acid. Also the role of Cmk1 was investigated. Pdr12 and Cmk1 encoding genes were deleted either individually or in the same strain to address their additive effect on weak acid resistance. Growth of the strains was monitored on glucose in the presence of various concentrations of the weak acids. The effect of pre-adaptation to sorbic or acetic acid on growth was also assessed.

It was surprisingly found that the Apdr12 and Apdrl 2Acmk1 strains had an im- proved tolerance to acetic and formic acid and showed improved metabolic activity when grown on lignocellulosic hydrolysate. This feature makes these strains useful in conversions of sugars from lignocellulosic hydrolysates.

Surprisingly it was also found that yeast strains made deficient of Pdr12 encoding gene were capable of improved xylitol production on lignocellulosic hydrolysate when a gene coding for aldose reductase was expressed in a Pdr12 deficient strain.

This invention is directed to a method for cultivating yeast in lignocellulosic hydrolysate. In said method the yeast made deficient of the plasma membrane transporter Pdr12 encoding gene or a closest homologue of said gene in other species of yeast or a fragment of said genes is cultivated in a medium comprising lignocellulosic hydrolysate. Pdr12 deficient yeast strains have improved viability on lignocellulosic substrate. They also have improved metabolic activity compared to respective wild type strains when grown on lignocellulosic hydrolysate.

Lignocellulosic hydrolysate can be obtained by treating lignocellulosic material by any method such as alkali or acid treatment, organosolv treatment, steam explosion, enzymatic hydrolysation or any combination thereof. Preferred methods are organosolv, sulfuric acid treatment and steam explosion, steam explosion being the most preferred. Alternatively, the hydrolysate may be fractionated to cellulose, hemicellulose and lignin. Without restricing to these, fractions containing C6 sugars are preferred in e.g. ethanol production whereas C5 containing fractions may be beneficial in xylitol production. Also simultaneous saccharification and fermentation processes (SSF) are within the scope of the invention.

Often it is necessary to add nutrients such as amino acids, vitamins or salts to the hydrolysate. The method of culturing is suitable for production of any added value products from said lignocellulosic hydrolysate such as ethanol, sugar alcohols such as xylitol, organic acids or proteins. It may also be beneficial to use different fractions of lignocellulose hydrolysate for example to convert cellulose derived glucose to ethanol or hemicellulose derived xylose to xylitol. Alternatively, glucose and xylose sugars from cellulose and hemicellulose fractions may be combined and both fermented to ethanol; or cellulose and hemicellulose fractions may be combined in certain ratios to provide a carbon source for growth and maintenance while converting xylose to xylitol. Preferably the yeast used in said cultivation is capable of producing ethanol or xylitol or both of them. Ethanol produced from lignocellulosic raw material is an environmentally friendly alternative to fossil fuels. The xylose of lignocellulosic hydrolysate is an economic raw material for xylitol production.

The method can be further enhanced by adapting the yeast to acid conditions by cultivating it in acid conditions before transferring to lignocellulosic medium.

Optionally said yeast has further been made deficient of the calmodulin - dependent protein kinase Cmk1 encoding gene. Acid adaptation and Cmk1 silencing both, independently of each other, further increase the viability and metabolic activity of the yeast in said culture. Cmk1 is associated with post-transcriptional negative regulation of Pdr12 and deletion of CMK1 can improve resistance to weak organic acid stress by shortening the lag phase of cells in their presence. Cmk1 is a Ca ²⁺ -calmodulin-dependent protein kinase with broad substrate specificity. The calmodulin-Ca ²⁺ -complex regulates a large variety of cellular functions and is essential to S. cerevisiae. There are three homologues of CMK1 (Gene ID: 850568, NCBI) in the genome of S. cerevisiae, CMK2 (Gene ID: 854144, NCBI), RCK2 (alias CLK1 and CMK3, Gene ID: 850950, NCBI) and RCK1 (Gene ID: 852719, NCBI).

Within the scope of the present invention is also yeast culture comprising a yeast that has been made deficient of the closest homologue of the Pdr12 and/or Cmk1 gene in their genome or a nucleotide sequences hybridizing under stringent conditions to said genes or said homologues. Said culture is or has been cultivated in a culture medium comprising lignocellulosic hydrolysate. Stringent conditions refer here to an overnight incubation at 42 degree C in a solution comprising 50% formamide, 5x SSC (750 mM NaCI, 75 mM trisodium citrate), 50 mM sodium phosphate (pH 7.6), 5x Denhardt's solution, 10% dextran sulfate, and 20 pg/ml denatured, sheared salmon sperm DNA, followed by washing the filters in 0.1 x SSC at about 65 °C.

Also modifications (e.g. inactivation) of fragments, derivatives or other nucleotide sequences of genes mentioned here (pdr12 and cmkl and their homologues) and genes hybridizing under stringent conditions to said genes or said homologue are usable within scope of this invention. Within the scope of the present invention are also modifications of derivatives of said gene, which refer to nucleic acid sequences comprising deletions, substitutions, insertions or other modifications compared to said gene, but having the same or equivalent function as said gene.

The "closest homologue of a gene in other species of yeast" means here a gene that has the highest percentage of identical nucleotides and/or essentially same function with said gene of all the genes of the organism. The sequence identity of homologous regulatory genes in different organisms is typically very low. Typically, the sites binding either to DNA or other protein factors involved in the regulation event share homology, but the intervening sequences between these sites may not be conserved. Therefore the total % of sequence identity of homologous regulatory genes in different organisms may remain relatively low.

Examples of well known homologues of Pdr12 (Gene ID: 856049, NCBI) are the Snq2 (Gene ID: 851574, NCBI) and Pdr5 (Gene ID: 854324, NCBI) ABC drug efflux pumps (Piper 1998).

However, the percentage of sequence identity in the aligned nucleotide sequence can be used as a measure to identify the closest homologue of the gene in the other species of yeast, thus a likely functional counterpart of the gene in the other organism. Software and algorithms for homology searches as well as public databases with whole genome sequence information for a variety of species exist, such as the BLAST program ( ^{(a b)} Altschul et al and the NCBI database (http://www.ncbi.nlm.nih.gov /sutils/genom_table.cgi?organism=fungi).

A specific "gene" (encoding Pdr12 (PDR), Cmkl (CMK), xylose reductase (XR)) is here represented by a specific sequence (SEQ ID NO). Table 1. Comparison of different Pdr12-type of protein sequences BLASTP 2.2.28+ programme ( ^(a) Altschul. et al.)

Pdr12 gene is characterized by SEQ ID NO: 1 or a sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 95% or even 98% identity to said sequence.

Table 2. Comparison of different Cmk1 -type of protein sequences BLASTP 2.2.28+ programme ( ^(a) Altschul et al.)

Cmk1 gene is characterized by SEQ ID NO: "CMK" or a sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 95% or even 98% identity to said sequence.

Yeast suitable for use within this invention can be any yeast. Preferably yeast is selected from the genera Saccharomyces, Kluyveromyces, Candida, Scheffersomyces, Pachysolen, Pichia and Hansenula. Yeast species of particular interest include S. cerevisiae, S. exiguus, K. marxianus, K. lactis, K. thermotolerans, C. sonorensis, P. kudriavzeii (also known as /. orientalis and C. krusei), C. sheha- tae, Pachysolen tannophilus and Scheffersomyces stipitis. More preferably said yeast belongs to genus Saccharomyces and most preferably to species S. cerevisiae. Saccharomyces is well known production organism and e.g. an efficient ethanol producer and also capable of efficient xylitol production, when engineered to express a xylose reductase encoding gene.

This invention is also directed to a yeast culture comprising yeast made deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast wherein said yeast has been cultured in a culture medium comprising lignocellulosic hydrolysate or other formic and acetic acid containing media. Optionally the yeast in said culture is also made deficient of Cmk1 gene or its closest homologue in other species of yeast. Further, it is advantageous to adapt said culture to acid conditions. Preferably said culture comprises yeast that has been genetically modified to be capable of increased expression of gene encoding aldose reductase. The said culture can also comprise of yeast that thas been engineered for production of ethanol from C5 carbon souces such as xylose or arabinose by introducing genes coding for xylose reductase and xylitol dehydrogenase or alternatively xylose isomerase for xylose to ethanol conversion or aldose reductase, L-arabinitol dehydrogenase, L-xylulose reductase, D-xylulose reductase or L-arabinose isomerase for arabinose to ethanol conversion. In addition, the strain may have increased expression of genes coding for xylulokinase or ribulokinase, ribose-5-phosphate ketol-iso- merase, D-ribulose-5-phosphate 3-epimerase, transketolase and transaldolase. This invention is also directed to a method for improving yeast viability on lignocellulosic hydrolysate and/or improving yeasts metabolic activity when cultured on medium comprising lignocellulosic hydrolysate. In said method the yeast is made deficient of a gene encoding the plasma membrane transporter pdr12 or its closest homologue in other species of yeast and optionally also deficient of a gene encoding the calmodulin -dependent protein kinase Cmk1 or its closest homologue in other species of yeast. Still further improvement of viability and metabolic activity can be indepently achieved by acid adaptation before transferring the yeast or culture to a lignocellulosic hydrolysate. Good viability is needed e.g. when production is operated in a repeated batch mode i.e. the cells are reused.

Finally this invention is directed to a genetically modified yeast host. The yeast is made deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast as discussed above.

Preferably said yeast has further been made deficient of the calmodulin - dependent protein kinase Cmk1 encoding gene or its closest homologue in other species of yeast. It was surprisinly found that this further enhances the metabolic activity on lignocellulosic culturing medium and/or in medium containig formic acid or acetic acid or both.

Optionally the yeast is further genetically modified to be capable of increased production of a desirable product; xylitol or ethanol production are preferred. This can be achieved by stably intoducing means for expressing aldose reductase encoding gene or by genetically modifying regulatory elements of enodgeneous aldose reductase gene

By increased xylitol production is meant xylitol production which is a least 3%, preferably at least 5%, more preferably at least 10%, still more preferably at least 20%, still more preferably at least 30% or most preferably at least 50 % better than production by using the parent host strain which has not been genetically modified to be deficient of pdr12 gene but otherwise shares the features of yeast of the invention. Aldose reductase gene can be exogeneous to said yeast host. It is possible to isolate a polynucleotide fragment comprising the gene, insert it under a strong promoter in an expression vector, and transfer into yeast cells. It is also possible to express an exogeneous gene under endoegeneous promoters. Preferably any exogeneous gene is intregrated to yeast genome. Endogenous aldose reductase can also be used. Preferably expression of an endogeneous gene is enhanced by modifying the regulatory regions of said aldose reductase encoding gene.

Aldose reductases (or aldehyde reductases) are characterized by EC 1 .1 .1 .21 . Preferably the aldose reductase is xylose reductase gene having SEQ ID NO: 13 or a polynucleotide having at least 70%, preferably, at least 80%, more preferably at least 85%, 90%, 95% or 98% identity to SEQ ID NO: 13 and encdoding xylose reductase having SEQ ID NO: 14 or a polypeptide having at least 70%, preferably, at least 80%, more preferably at least 85%, 90%, 95% or 98% identity to SEQ ID NO: 14. Aldose reductase (EC 1 .1 .1 .21 ) is an NAD(P)H-dependent oxidoreductase that catalyzes the reduction of a variety of aldehydes and carbonyl compounds. Most preferred aldosereductase is xylose reductase having SEQ ID NO: 14. Xylose reductases catalyse the initial reaction in the xylose utilization pathway i.e. the NAD(P)H dependent reduction of xylose to xylitol. Xylitol as a valuable product can then be recovered.

In one embodiment the gene encoding protein characterized by SEQ ID NO: 14 or a sequence having at least 70%, 75%, 80%, 85%, 90% identity and most prefera- bly at least 95% or even 98% identity to gene encoding the polypeptide having SEQ ID NO: 14or an active fragment thereof is stably introduced to the host cell, preferably integrated into genome.

Optionally said yeast is capable of increased ethanol production. The pathways involved in ethanol production can be up-regulated in said yeast in order to enhance utilization of C6 and C5 sugars, preferably glucose or xylose, most preferably glucose and production of ethanol. The yeast can also be engineered for production of ethanol from C5 carbon souces such as xylose or arabinose by introducing genes coding for xylose reductase and xylitol dehydrogenase or alternatively xylose isomerase or aldose reductase, L-arabinitol dehydrogenase, L- xylulose reductase, D-xylulose reductase or L-arabinose isomerase into the yeast strain. In addition the strain may have increased expression of genes coding for xylulokinase or ribulokinase, ribose-5-phosphate ketol-isomerase, D-ribulose-5- phosphate 3-epimerase, transketolase and transaldolase or other beneficial genes. By increased ethanol production is meant ethanol production which is a least 3%, preferably at least 5%, more preferably at least 10%, still more preferably at least 20%, still more preferably at least 30% or most preferably at least 50 % better than production by using the parent host strain which has not been genetically modified to be deficient of PDR12 gene but otherwise shares the features of yeast of the invention.

In one embodiment said yeast is genetically modified to be able to produce both ethanol and xylitol on lignocellulosic culture medium.

Methods for production of products by recombinant technology in different host systems are well known in the art (e.g. Glick et al). Preferably the products are secreted into the culture medium, from which they can easily be recovered and isolated. The spent culture medium of the production host can be used as such, or the host cells may be removed therefrom, and/or it may be concentrated, filtered or fractionated. Optionally said yeast (e.g. yeast host, yeast comprised in yeast culture) is adapted to acid conditions by cultivating it in acid conditions before cultivation on lignocellulosic hydrolysate containing medium. Suitable acidic conditions are e.g. 0.45 mM sorbic acid or 50 mM acetic acid. Adaptation to acidic conditions further increases metabolic activity during cultivation on lignocellulosic hydrolysate containg medium.

Preferably the yeast suitable for use within this invention has an increased tolerance towards formic acid and acetic acid compared to a yeast where pdrl 2 gene has not been modified. Preferably said yeast tolerates at least 25mM formic acid, more preferably even at least 30 mM formic acid. Acetic acid and formic acid are present in lignocellulosic hydrolysate and prevent growth of the yeast or at least decrease metabolic activity. The pH of SC medium with 25 mM formic acid is 3.1 at the start of the cultivations, used as examples for this invention. Acetic acid and formic acid are present in lignocellulosic hydrolysate and prevent growth of the yeast or at least decrease metabolic activity.

Preferably the yeast suitable for use within this invention is adapted to acidic conditions before cultivation on medium comprising lignocellulosic hydrolysate. Lignocellulosic hydrolysate is typically acidic. Precultivation in acid conditions increases metabolic activity of the culture during culturing on lignocellulosic hydrolysate. In addition acidic growth conditions decrease the contamination risk.

Preferably said yeast is further made deficient of a gene encoding the calmodulin -dependent protein kinase Cmk1 or its closest homologue in other species of yeast.

Preferably said yeast has increased tolerance towards formic acid and acetic acid compared to a non-modified yeast. Those are the most common acids in a lignocellulosic culturing medium and thus improved tolerance is necessary. According to the invention glucose consumption with the Pdr12 deficient strain improved to -17 g/l in 140 h and with acid adapted cells to -10 g/l in 70 h in 95% straw hydrolysate compared to control which did not consume any glucose. Similar results were obtained with birch hydrolysate.

The most preferred host belongs to the genus Saccharomyces, most preferably to species S. cerevisiae. A yeast cell capable of producing xylitol when cultivated in medium comprising lignocellulosic hydrolysate can be constructed as follows:

(a) providing a yeast cell; and

(b) genetically modifying said cell to be deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast and

(c) genetically modifying said cell to enhance expession of aldose reductase encoding gene; or

(d) introducing means for expressing aldose reductase encoding gene into said cell; or

(e) both (c) and (d); and

(f) optionally adapting said cells to adicic conditions by cultivating in acidic conditions and

(g) cultivating said cells in conditions allowing expression of aldose reductase encoding gene and production of xylitol

(h) recovering xylitol.

Optionally the cell is further genetically modified to be deficient of a gene encoding the calmodulin-dependent protein kinase Cmk1 encoding gene or its closest homologue in other species of yeast.

A yeast cell capable of producing ethanol when cultivated in medium comprising lignocellulosic hydrolysate can be constructed as follows:

(a) providing a yeast cell; and

(b) genetically modifying said cell to be deficient of a gene encoding the plasma membrane transporter Pdr12 or its closest homologue in other species of yeast and

(c) genetically modifying said cell to express xylose reductase and xylitol dehydrogenase or alternatively xylose isomerase for xylose to ethanol conversion or aldose reductase, L-arabinitol dehydrogenase, L-xylulose reductase, D-xylulose reductase or L-arabinose isomerase for arabinose to ethanol conversion. Optionally the strain also has increased expression of genes coding for xylulokinase or ribulokinase, ribose-5-phosphate ketol- isomerase, D-ribulose-5-phosphate 3-epimerase, transketolase and transaldolase; or

(d) introducing into said cell means for expressing genes encoding activities for C5 sugars-to-ethanol conversion xylose reductase and xylitol dehydrogenase or alternatively xylose isomerase for xylose to ethanol conversion or aldose reductase, L-arabinitol dehydrogenase, L-xylulose reductase, D-xylulose reductase or L-arabinose isomerase for arabinose to ethanol conversion. Optionally the strain also has increased expression of genes coding for xylulokinase or ribulokinase, ribose-5-phosphate ketol- isomerase, D-ribulose-5-phosphate 3-epimerase, transketolase and transaldolase; or

(e) both (c) and (d); and

(f) optionally adapting said cells to adicic conditions by cultivating in acidic conditions; and

(g) cultivating said cells in conditions allowing expression of genes encoding activities for C5 sugars-to-ethanol conversion such as xylose reductase, xylitol dehydrogenase, xylose isomerase, aldose reductase, L- arabinitol dehydrogenase, L-xylulose reductase, D-xylulose reductase, L- arabinose isomerase, xylulokinase, ribulokinase, ribose-5-phosphate ketol-isomerase, D-ribulose-5-phosphate 3-epimerase, transketolase, transaldolase, xylose dehydrogenase, arabinose dehydrogenase, pentonate dehydratase, aldolase, pyruvate decarboxylase or alcohol dehydrogenase, and production of ethanol; and

(h) recovering ethanol. Optionally the cell is further genetically modified to be deficient of a gene encoding the calmodulin-dependent protein kinase Cmk1 encoding gene or its closest homologue in other species of yeast.

Expression of a gene can be enhanced by modifying regulatory regions of said gene using commonly known methods. Means for expression include the DNA encoding the polypeptide to be expressed operably linked to suitable regulatory regions.

The invention is illustrated by the following non-limiting examples. It should be understood, however, that the embodiments given in the description above and in the examples are for illustrative purposes only, and that various changes and modifica- tions are possible within the scope of the invention. Examples

Example 1. Production of wheat straw and birch sawdust hydrolysate

Wheat straw (0.7 kg d.w.) and birch sawdust (1 .0 kg d.w.) were impregnated in 0.5% H ₂ SO ₄ for 2 h and steam exploded at 200 ^° C with treatment time of 5 min. During the treatment the sample was kept in a steel mesh container from which the steam exploded fibre fraction was collected after treatment. The liquid condensate produced during the treatment containing some C5 sugars and soluble inhibitors was collected separately after treatment. Approximately 1 1 % (w/w) of straw and 18% of sawdust dry matter solubilized during steam explosion.

The fibre fraction was hydrolysed in a 5 L horizontal high consistency reactor for 24 h with a cocktail of hydrolytic enzymes containing 15 FPU/g Celluclast 1 .5L, 500 nkat/g Novozyme 188 (β-glucosidase activity) and 1500 nkat/g Depol 740 (xy- lanase activity). Dry matter content was adjusted to 12% for straw (3 L total vol- ume) and 21 % for sawdust (4 L total volume) using the liquid fractions produced in steam explosion. pH was adjusted to 4.8 with 10 M NaOH. The slurry was centri- fuged (4000 rpm, 20 min) and supernatant was decanted and boiled for 5 min to inactivate enzymes. Finally the solution was filtered (90 μιτι wire cloth) to remove any precipitation and stored frozen in bottles. Approximately 50% of straw dry mat- ter and 40% of sawdust dry matter subjected to enzymatic hydrolysis was solubilized during the treatment, the majority of this being sugars.

All together ca. 1 .8 L straw hydrolysate with 99 mg/ml dry matter content and ca. 2.0 L birch sawdust hydrolysate with 144 mg/ml dry matter content were produced. Birch sawdust hydrolysate appears to contain some fine insoluble solids even after centrifugation and filtration. Total solubilisation over both of the steps was approximately 56% for straw and 51 % for sawdust, but 16% of solubilized straw and 30% of solubilized sawdust dry matter were discarded between the treatments and did not end up in the final hydrolysate. Both hydrolysates contained formic and acetic acid.

Example 2. Strains and strain construction and culture conditions

Saccharomyces cerevisiae CEN.PK2-1 D (VW-1 B) (MATa, /eu2-3/112 ura 3-52 trpl- 289 his3A^ MAL2-8 ^C SUC2) (Boles et al. 1996J was used as a parental strain. Strains used in the study leading to the present invention are listed in Table 3. Strains carry a genomic LEU2 gene, or express the LEU2 gene in the modified pYX242 vector (R&D Systems, UK; Toivari et al. 2010). All yeast transformations were carried out using the method described by Gietz et al. (1992).

Table 3. Strains

Strain number Genotype Plasmid

Ctrl H4131 CEN.PK2-1 D, TRP1 + mpYX242

Acmkl H4132 Acmkl Y. TRP1

Apdr12 H4133 Apdr12:±EU2, TRP1

Apdr12Acmk1 H4134 Apdr12: :LEU2; Acmkl: : TRP1

Ctrl + XR H4257 CEN.PK2-1 D, TRP1, LEU2 pMLV122

Ctrl H4259 CEN.PK2-1 D, TRP1, LEU2 B547

Apdr12 + XR H4260 Apdr12:±EU2, TRP1 pMLV122

Apdr12 H4262 Apdr12::LEU2, TRP1 B547

Apdr12Acmk1 + pMLV122

H4263 Apdr12: :LEU2; Acmkl: : TRP1

XR

Apdr12Acmk1 H4265 Apdr12: :LEU2; Acmkl: : TRP1 B547 Table 4. Sequences and SEQ ID NO's

SEQ ID

Primer/Gene NO:

Pdr12_Leu2_Fwd 3

Pdr12_Leu2_Rev 4

Cmk1_His3_Fwd 5

Cmk1_His3_Rev 6

Cmk1_Trp1_Fwd 7

Cmk1_Trp1_Rev 8

Trp1_Fwd 9

Trp1_Rev 10

Pdr12 SEQ ID NO:1

Cmk1 SEQ ID NO: 2

The deletion of the PDR12 gene was achieved by replacement of the ORF by the LEU2 gene. The LEU2 gene, with its endogenous promoter and terminator sequences, was amplified from plasmid p425MET25 (Mumberg et al. 1994) using the Pdr12_Leu2_Fwd and Pdr12_Leu2_Rev primers, having flanks overlapping the sequences in the 5'- and 3'- regions of the PDR12 gene (Table 3). The PCR product was then introduced in to the parental strain, generating the Apdrl 2::LEU2 strain, here named Apdrl 2 (Table 3).

The deletion of the CMK1 gene was achieved by replacement of the ORF by the TRP1 gene. The TRP1 gene was amplified with endogenous promoter and terminator sequences using genomic DNA as template. The Cmk1_Trp1_Fwd and Cmk1_Trp1_Rev primers were used for amplification of the TRP1 gene (Table 4). The primers were designed with flanks overlapping the sequences in the 5'- and 3'- regions of the CMK1 gene. The PCR products were then introduced to the pa- rental strain or Apdr12 strain, generating strains Acmk1 TRP1 (Table 3). All deletions were verified using PCR.

The parental strain was cured to tryptophan prototrophy by introducing the functional endogenous TRP1 gene. This was achieved by replacement of the complete ORF for TRP1, amplified from genomic DNA template using the Trp1_Fwd and Trp1_rev primers (Table 4). Transformants were selected by growth in medium lacking tryptophan.

Finally, the empty modified pYX242 plasmid was introduced into all strains, except the strains with deleted PDR12. In this way, all strains generated were leucine pro- totrophs and they could be grown in the same medium. The parental strain con- taining the empty modified pYX242 plasmid is in this invention referred to as the control strain Ctrl.

Media and culture conditions

Modified synthetic complete medium (YSC), lacking leucine, with 20 g D-glucose ¹ (SCD-leu medium) was used for cultivations. 20 or 50 ml overnight pre-cultures were grown in 100 or 250 ml Erienmeyer flasks, at 250 rpm and 30°C. When the effect of the prior induction of the endogenous PDR12 gene was tested, 0.45 mM sorbic acid was added to the growth medium. The effect of adaptation to growth in the presence of acetic acid was studied by using cells from pre-cultures supple- mented with 50 mM acetic acid for 16 hours.

SCD-leu medium with various acids added was used for growth assays in a Bi- oscreen apparatus (Bioscreen C MBR automated turbidometric analyser, Growth Curves Ltd, Finland). Formic and acetic acid were prepared as 200 mM stocks, with no pH adjustment. Medium (270 μΙ) containing an appropriate final concentra- tion of the acid was added to the wells of a Bioscreen microtitre plate (100-Well Honeycomb plate). Cells from pre-cultures were diluted into deionised water to an OD of 0.5. 30 μΙ of cell suspension was inoculated into the wells containing the medium. Growth at 30°C with continuous, extra intensive shaking was monitored by measuring the optical density (OD) at 600 nm. The pH of the medium without added acid was 6.0, whereas the pH of acid-containing media was generally <4, depending on the acid and the amount of acid added. The effect of inorganic acid on growth was assessed in a medium where the pH was adjusted to pH 4.5 by addition of HCI. Cells were defined as growing from the time when the OD of the cells exceeded 0.2. Each condition assessed in the Bioscreen was performed in at least four replicates.

Example 3. Modulated expression of PDR12 or CMK1 affects growth on pure medium Deletion of PDR12 from the control strain or from the Acmkl strain increased the specific growth rate on glucose, when examined in defined standard culture medium (described in Example 2). At pH 6, when cells were pre-grown without acid, the specific growth rate of the Apdr12 strain was increased by 7% compared to the control strain, whereas the Apdr12Acmk1 strain grew 30% faster than the Acmkl strain (p < 0.05, Table 5). Deletion of CMK1 decreased the specific growth rate on glucose by 21 % (p < 0.05), compared to the control strain (Table 5). When the in- ocula were grown in medium containing 0.45 mM sorbic or 50 mM acetic acid to allow adaptation to weak organic acid stress, the specific growth rate on glucose was not affected, except that the growth rate of the Acmkl strain now remained comparable to the control strain (p < 0.05, Table 5). Likewise, when the pH of the medium was adjusted to 4.5, the specific growth rate of the Acmkl strain was increased, compared to growth in the medium with initial pH of 6.0 (p< 0.05, Table 5), but the specific growth rate of the other strains was not affected. Table 5. Specific growth rate on glucose, in SCD-leu medium with 20 g D-glucose 1, 300 μΙ cultures in Bioscreen microtitre plates. Strains were pre-grown in SCD- leu medium or in SCD-leu medium with 0.45 mM sorbic acid or 50 mM acetic acid. Data shown is the mean ± SEM for 4 to 31 replicate cultures. The pH refers to the pH of the medium at the time of inoculation.

Strain Initial pH =6 Initial pH = 4.5 Pre-growth Pre-growth with sorbic acwith acetic acid, initial pH = id, initial pH = 6 6

Ctrl 0.32 ± 0.004 0.34 ± 0.004 0.31 ± 0.002 0.32 ± 0.001 Acmkl 0.26 ^* ± 0.004 0.29 ^* ± 0.002 0.30 ± 0.008 0.32 ^* ± 0.003 Apdr12 0.34 ^* ± 0.003 0.36 ^* ± 0.004 0.35 ^* ± 0.003 0.35 ^* ± 0.001 Acmkl Acmkl 0.34 ^* ± 0.004 0.36 ± 0.004 0.34 ^* ± 0.003 0.36 ^* ± 0.001

^* significantly different (p<0.05) from the control in the given condition

Example 4. Effect of PDR12 deletion on resistance to acetic and formic acid on defined medium

The Apdr12 and Apdrl 2Acmk1 strains were more tolerant to acetic and formic acid than the control strain, with shorter lag phases in the presence of 100 mM acetic or 25 mM formic acid (Fig. 1 ), and reaching higher final biomass concentrations within the 70h studied. Cells growing in 100 mM acetic acid did not consume the acetic acid (data not shown). Example 5. Apdrl 2 strains recover more rapidly from exposure to inhibitory concentrations of acetic acid than strains with PDR12

The viability of cells exposed to acids was determined by comparing the number of colony forming units (CFU) on agar-solidified SCD-leu medium to the total cell number determined microscopically, using a Cellometer Auto T4 cell counter (Nexcelom Bioscience LLC, USA). Cells were cultured in Bioscreen and the device was stopped to remove samples to assess viability.

When exposed to 100 mM acetic acid, viability decreased in all strains studied; less than 1 % of the cells were viable after 16 hours of incubation (Fig. 2). The percentage of viable cells of the control and Apdr12 strains had increased after 40 hours in the presence of 100 mM acetic acid (Fig. 2), corresponding with the end of the lag phases of these strains (Fig. 1a). By 72 hours, when Apdrl 2 and Apdr12Acmk1 had grown to OD's of 0.7 to 0.9 (Fig. 1a), the viability of these strains was 71 %, whereas the control strain was only 31 % viable.

Example 6. Adaptation to growth in the presence of sorbic or acetic acid en- hances the benefits of deleting PDR12 (with or without deletion of CMK1)

The lag phase of all strains in medium supplemented with 100 mM acetic acid was shorter when they were pre-grown in the presence of 0.45 mM sorbic acid (Fig. 3a), than when they were not (Fig. 1 a), but the control did not grow as well as the Apdr12 strains. Only the Apdr12 and the Apdrl 2Acmk1 cells were able to grow in the presence of 120 mM acetic acid when cells had been pre-grown in 0.45 mM sorbic acid, the Apdrl 2Acmk1 strain having a shorter lag phase than the Apdr12 strain (Fig. 3b). None of the strains were able to grow in 1 10 mM acetic acid, if they had not been pre-grown in 0.45 mM sorbic acid. When the sorbic acid pre- grown cells were exposed to 25 mM formic acid, the Apdrl 2Acmk1 strain had a slightly shorter lag phase compared to cells pre-grown without sorbic acid. The lag phase of both the Apdr12 and the control strain was increased and biomass production decreased after pre-growth in sorbic acid, compared to pre-growth without sorbic acid (Fig. 3c and 1 b), but the Apdrl 2 strain grew better than the control.

When the strains were pre-grown in the presence of 50 mM acetic acid, only the Pdr12 deficient strains (Apdrl 2 and Apdrl 2Acmk1) showed improved growth (-25 hours shorter lag phase) when subsequently exposed to 100 mM acetic acid (Fig. 3d and 1a). The Apdrl 2Acmk1 strain also showed higher tolerance to 25 mM formic acid (10 hours shorter lag phase), when the cells were pre-grown in the presence of 50 mM acetic acid, compared to cells pre-grown in a medium without any added acetic acid (Fig. 3e and 1b). After the pre-growth in the presence of 50 mM acetic acid the control strain did not grow when exposed to 100 mM acetic acid or 25 mM formic acid (Fig. 3d-e). Viability tests demonstrated that only the Apdr12 strains were able to regain viability in 100 mM acetic acid after pre-incubation in 50 mM acetic acid (Fig. 3f).

Example 7. Effect of PDR12 deletion or deletion of both PDR12 and CMK1 on viability, glucose consumption and ethanol production on wheat straw hy- drolysate

The wheat straw hydrolysate contained 3.2 g/l acetic acid (53 mM) and 0.5 g/l (1 1 mM) formic acid in addition to sugars (mainly glucose and xylose). An initial test of the viability of the deletion strains in the hydrolysate with added SC-leu components was performed (Fig. 4). Apdr12 strains survived more than 2 days in the hydrolysate, whereas the control strain was killed.

When grown on 95% straw hydrolysate the Pdr12 deficient strain consumed -17 g/l glucose in 140 h (-10 g/l in 70 h with acid adapted cells) compared to the control which did not consume any glucose (Figure 5a). In addition, ethanol production with the Pdr12 deficient strain was 8 g/l with non-adapted cells and 12 g/l with adapted cells. Moreover, if the strain was deficient both in Pdr12 and Cmk1 the glucose consumption in 70 h was 17 g/l and ethanol formation 15 g/l with the acid adapted cells (Figure 5a). Similar results were observed in birch hydrolysate (Figure 5b). As a conclusion, it was surprisingly found that Pdr12 was undesirable for viability and metabolic activity in biomass hydrolysate.

Example 8. Construction of strains for xylitol production The xylose reductase encoding gene from Scheffersomyces stipitis between PGK1 promoter and terminator (pPGK-XYL1 -tPGK) was moved from vector pUA103 (B383) as Hind Ill-fragment to Hindlll site of Yeplac195, resulting in plasmid pMLV122. The plasmid pMLV122 containing the XYL1 gene under PGK1 promoter was transformed into yeast strain H4133 (deltaPDR12 (TRP)), resulting in strain H4260 and transformation of the empty vector (B547) resulted in strain H4262. Transformation of plasmid pMLV122 into strain H4134 (deltaPDR12 del- tacmkl (TRP)) resulted in strain H4263 and transformation of empty vector (B547) into strain H4265. CEN.PK2-1 D strain with functional TRP1 and LEU 2 genes was transformed with plasmid pMLV122 resulting in strain H4257 and transformation of empty vector (B547) resulted in strain H4259. Strains H4257 and H4259 were used as control strains.

Example 9. Effect of PDR12 deletion or deletion of both PDR12 and CMK1 on xylitol production in strains expressing in addition the XYL1 of Schefferso- myces stipitis in medium containing formate or acetate or in lignocellulosic hydrolysates

Xylitol production with strains described in example 8 was studied on pure growth medium SC-ura with 20 g/L xylose and 20 g/l glucose supplemented with 15 mM formic acid. Strains with the APdr12 +/- ACmkl +XR produced more / produced a higher xylitol concentration (20 g/l) compared to the control+XR (15 g/l). When the medium was supplemeted with 80 mM acetic acid instead of formic acid the APdr12 and APdr12 ACmkl strains had higher growth rate or shorter lag phase compared to the control strains. In addition, APdr12+XR and APdr12 ACmk1+XR had higher specific xylitol production rate than the control. Xylitol production in 60 and 70% birch or 80% straw hydrolysates were tested with non acid adapted cells. In birch hydrolysate Apdr12+XR consumed both glucose and xylose faster compared to the control (data not shown). The Apdr12+XR (red) strain produced xylitol, whereas the strain with intact/ wild type PDR12 did not (Figure 6 a). On 80% straw hydrolysate glucose was consumed only by the Pdr12 deficient strain (data not shown). In addition, on 80% straw hydrolysate xylitol was only formed with the Apdr12+XR strain (Figure 6 b).

In conclusion, deletion of PDR12 leads to an improved production of xylitol from xylose in pure medium containing formic or acetic acid and from lignocellulosic hydrolysates.

References

Boles E, Gohlmann HW, Zimmernnann FK (1996) Cloning of a second gene encoding 5-phosphofructo-2-kinase in yeast, and characterization of mutant strains without fructose-2,6- bisphosphate. Mol Microbiol 20: 65-76

Gietz D, St Jean A, Woods RA, SchiestI RH. Improved method for high efficiency transformation of intact yeast cells. Nucleic Acids Res. 1992 Mar 25;20(6):1425 Glick, Pasternak and Patten 2009, Molecular Biotechnology- principles and applications of recombinant DNA.Fermentation Microbiology and Biotechnology, Third Edition. E. M. T. El-Mansi (Editor), C. F. A. Bryce (Editor), Arnold L. Demain (Editor), A.R. Allman (Editor). 201 1 . Piper P, Mahe Y, Thompson S, Pandjaitan R, Holyoak C, Egner R, Muhlbauer M, Coote P, Kuchler K. The pdr12 ABC transporter is required for the development of weak organic acid resistance in yeast. EMBO J. 1998 Aug 3;17(15):4257-65

^(a) Altschul SF., Madden TL, Schaffer AA., Zhang J., Zhang Z., Miller W., and Lipman DJ. (1997), "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs", Nucleic Acids Res. 25:3389-3402

^(b) Altschul SF.,. Wootton J. C, Gertz EM., Agarwala R., Morgulis A., Schaffer A A. and Yu Y. (2005) "Protein database searches using compositionally adjusted sub- stitution matrices", FEBS J. 272:5101 -5109)

Toivari MH, Ruohonen L, Richard P, Penttila M, Wiebe MG. Saccharomyces cere- visiae engineered to produce D-xylonate. Appl Microbiol Biotechnol. 2010

Oct;88(3):751 -60