Background of the invention The term"sphingolipids"refers to a group of lipids which are derived from sphingoid bases like sphingosine. Sphingolipids occur frequently in the cellular membranes of animals, plants and microorganisms. The exact function of sphingolipids in humans remains unknown, but it is clear that this group of compounds is involved in the transmission of electrical signals in the nervous system and in the stabilization of cell membranes. It has also been suggested that glycosphingosines have a function in the immune system: specific glycosphingosines function as receptors for bacterial toxins and possibly also as receptors for bacteria and viruses.

Ceramides are a specific group of sphingolipids containing sphingosine, dihydrosphingosine or phytosphingosine as a base in amide linkage with a fatty acid. Ceramides are the main lipid component of the stratum corneum, the upper layer of the skin. The stratum corneum has an important barrier function, external compounds are generally kept outside of its barrier and the loss of moisture is limited. Topical application of compositions comprising sphingolipids such as ceramides improves for instance the barrier function and moisture-retaining properties of the skin (Curatolo, 1987, Pharm. Res. 4, 271-277; Kerscher et al., 1991, Eur. J. Dermatol. 1, 39-43).

Sphingoid bases as such are known to mediate several physiological effects by inhibiting the activity of protein kinase C, a key enzyme in the signal transduction pathway. Furthermore, sphingoid bases are included in cosmetic or dermatological compositions for their anti-inflammatory and antimicrobial activity. ~ Currently, heterogenous sphingolipid preparations for cosmetics are mainly extracted from animal sources. Obviously, this is a rather costly process on an industrial scale. Moreover, it has been found that these materials are potentially unsafe due, for example, to the possible presence of bovine spongiform encephalopathy (BSE) in bovine tissue. Thus, the cosmetic industry has demonstrated an increasing interest in new sources of pure, well-defined sphingolipids, which are obtained from sources other than animal tissues.

Microorganisms such as the yeasts Pichia ciferrii (Wickerham and Stodola, 1960, J. Bacteriol. 80,484-491) have been found to produce sphingolipids as such, as well as sphingosine, phytosphingosine and/or derivatives thereof. This discovery provides sources for sphingolipids themselves and for starting materials for the production of other commercially valuable compounds which could offer a viable alternative to the use of animal sources of these compounds.

For example, acetylated derivatives of sphingosine, dihydrosphingosine and phytosphingosine as produced by Pichia ciferrii may be deacetylated and the thus-obtained sphingosine, dihydrosphingosine or phytosphingosine may be chemically converted into related compounds such as ceramides, pseudoceramides and/or glycoceramides which in turn may be applied in cosmetic and dermatological products (International patent application WO 93/20038).

Unfortunately, none of the yeast strains studied to date, even Pichia ciferrii NRRL Y-1031 F-60-10, produce sufficient amounts of sphingolipid bases such as sphingosine, phytosphingosine or derivatives thereof to be an efficient, economically attractive source of such compounds.

International patent application WO 95/12683 discloses Pichia ciferrii strains which have an enhanced productivity of TAPS as compared to the parent strain NRRL Y-1031 F-60-10. However, the productivity of these "improved"strains is increased at the most 1.6 times. Although it would be highly desirable to obtain strains with a substantially higher productivity, it-is uncertain whether further improvements in productivity would be achieved because of the toxicity of sphingoid bases to microbial cells (Pinto et al., 1992, J. Bacteriol. 174,2565-2574 ; Bibel et al., 1992, J. Invest. Dermatol.

98, 269-273).

Description of the invention The present invention discloses Pichia strains displaying a substantially enhanced productivity of a sphingoid base or derivative thereof. In the context of the present invention, a substantially enhanced productivity of a sphingoid base or derivative thereof is meant to relate to a productivity which is at least two times higher than the productivity of Pichia ciferrii Y-1031 F-60-10, deposited as CBS 408.94 (see WO 95/12683) when measured under identical conditions. Preferably, a substantially enhanced productivity of a sphingoid base or derivative thereof is meant to relate to a productivity which is at least four, more preferably which is at least six, even more preferably which is at least eight, most preferably which is at least ten times higher than the productivity of Pichia ciferrii Y-1031 F-60-10, deposited as CBS 408.94.

The present invention further discloses a novel method to isolate microbial strains displaying an enhanced productivity of sphingoid bases or derivatives thereof.

The novel method of the invention comprises an initial screening step wherein the sphingoid base productivity of a series of strains is measured by determining the antimicrobial activity of the sphingoid base or derivative thereof as produced by each individual strain being present within said series of strains. This initial screening step advantageously allows the screening of a

substantial amount of individual strains for their sphingoid base productivity and the selection of a subset of strains enriched in strains displaying an enhanced productivity as compared to a reference strain.

In particular, the method of the invention is directed to the selection or isolation of a microbial cell or a strain derived therefrom displaying an enhanced productivity of a sphingoid base or derivative thereof as compared to a reference strain. Said reference strain preferably is Pichia ciferrii Y-1031 F-60-10 or a derivative thereof.

The method of the invention comprises essentially the steps of: screening the sphingoid base productivity, of a series of strains by determining the antimicrobial activity of the sphingoid base as produced by each individual strain being present within said series of strains, selecting a subset of strains for at least one further passaging of cells, further passaging of cells by transferring an amount of cells from each individual strain within said selected subset to a next medium and culturing said amount of cells in said next medium under conditions conducive to the production of a sphingoid base, selecting a strain with an enhanced productivity of a sphingoid base as compared to the productivity of a reference strain when cultured under identical conditions.

It is to be understood that the term sphingoid base as used in this invention includes sphingoid base derivatives as specified hereinbelow, although the term derivatives may not always be specifically mentioned.

It further is to be understood that the method of the invention as described hereinabove represents one cycle of screening and selection.

However, repeated cycles of screening and selection also are applicable to obtain microbial strains displaying an enhanced sphingoid base productivity.

As a first step of the initial screening of said series of strains, cells from each individual strain being present within said series of strains are inoculated in a suitable culture medium, taking care that no mixing occurs of cells originating from different individual strains. In a next step, the cells are

cultured under conditions conducive to the production of a sphingoid base compound. When sufficient growth has occurred, the sphingoid base productivity of each individual culture is measured by determining the antimicrobial activity of the sphingoid base-containing culture supernatant.

The culturing of cells to obtain a culture supernatant for determining the antimicrobial activity thereof conveniently may be performed in any suitable system. For instance, the culturing may be performed in a shake flask or in a microtiterplate.

The antimicrobial activity of a sphingoid base-containing culture supernatant is conveniently determined by measuring the extent of the inhibiton of the growth of a suitable indicator strain being produced by incubation of said indicator strain in the presence of said culture supernatant.

A suitable indicator strain may be any microbial strain which is susceptible to the antimicrobial activity of a sphingoid base or derivative thereof. Preferably, the indicator strain is a bacterial strain, more preferably a gram-positive bacterial strain.

Briefly, a suitable dilution of the indicator strain is incubated with culture supernatant which is obtained from individual strains to be screened for sphingoid base productivity. A suitable dilution may be obtained by diluting a stationary culture, for instance an overnight culture, of the indicator strain.

A suitable dilution of an overnight culture typically is about 100 to about 500 times. The extent of growth inhibition as produced by the sphingoid base-containing culture supernatant is determined by measuring the optical density of the incubated samples. The incubation conditions may be dependent on the indicator strain used and the sphingoid base productivity.

For instance, the time period for incubation may vary from 4-8 hours to an overnight incubation, depending on the extent of growth inhibition which is obtained. The incubation temperature may be dependent on the optimal conditions for growth of the indicator strain. In general, the incubation temperature is 37 ° C. For prolonged incubations, the temperature may be lowered to about 25 °C.

In a preferred embodiment of the invention, the indicator strain is a bacterial strain of the species Staphylococcus aureus or Streptococcus pyogenes.

In a preferred embodiment of the invention, the initial screening step is performed using a microtiterplate system for the incubation and analysis steps. The advantage of the use of a microtiterplate system is that a large amount of individual strains can be screened under substantially automated conditions.

Based on the outcome of the antimicrobial assay, a subset of strains is selected for at least one further passaging of cells.

Said further passaging of cells comprises the transfer of cells from each individual strain being present within said selected subset of strains to a next medium and a subsequent culturing of each individual strain under conditions conducive to the production of the sphingoid base compound. The culture medium or a suitable dilution thereof is then analyzed for sphingoid base productivity using a suitable assay.

For instance, cells from each individual strain being present within the selected subset are cultured in shake flasks or tube cultures. The sphingoid base productivity of each culture is determined by analysis, preferably a quantitative analysis, of the culture medium or a suitable dilution thereof.

Strains are then identified which fulfil a certain increase in productivity as compared to a reference strain. ~ Said series of strains to be subjected to the screening and selection method of the invention may be obtained in several ways, said ways being not critical to the invention.

For instance, said series of strains may comprise a series of unrelated strains to be analyzed for sphingoid base productivity, such as strains from different species, different genera, different families, etc.

Strains which may be subjected to the screening and selection method according to the present invention include yeast, bacterial and fungal strains capable of naturally producing sphingoid bases and/or derivatives thereof.

Preferably, strains which are selected by the method of the invention are strains which demonstrate the highest inherent production levels of the desired products.

Yeast strains for use in the present invention may be selected from species of the genera Saccharomyces, Kluveromyces, Debaromyces, Pichia, Hansenula, Lipomyces, Sporobolomyces, Cryptococcus, Torulopsis, Endomycopsis, Candida, Trichosporon, Hanseniaspora and Rhodotorula.

Preferred yeast strains belong to the genus Pichia, Hansenula, Endomycopsis, Candida, Saccharomyces and Hanseniaspora and particularly the species Pichia ciferrii, Candida utilis and Saccharomyces cerevisiae. Most preferred are yeast strains which belong to the genus Pichia and most preferably to the species Pichia ciferrii (especially Pichia ciferrii NRRL Y-1031 F-60-10).

Preferred fungi are of the genera Aspergillus and Penicillium and are more preferably of the species Aspergillus sydowi and Penicillium notatum.

Preferred bacteria are of the genera Sphingobacterium (especially S. versatilis, S. multivorum and S, mizUtae), Acetobacter (especially A. xylinum), Bacteroides (especially B. melaninogenicus, B. fragilis, B. ruminicola and B. <BR> <BR> thetaiotaomicron), Bdellovibrio (especially Bdellovibrio bacteriovus), Xanthomonas (especially Xanthomonas campestris) and Flavobacterium (especially Flavobacterium devorans).

In a preferred embodiment of the invention, said series of strains is obtained via mutagenesis of a suitable parent strain. - According to this embodiment of the invention, a parent population of cells is subjected to a mutagenic treatment with the aim to introduce genetic variation into said population. The type of mutagenic treatment which is applied is not critical to the invention. A mutagenic treatment typically may comprise a so-called classical treatment, but also may include DNA-mediated transformation.

A classical mutagenic treatment includes a treatment with a chemical mutagenic agent, such as an alkylating agent like N-methyl, N'-nitro, N-nitrosoguanidine (NTG) or ethyl methane sulfonate (EMS), or a physical

treatment, such as UV irradiation, or a combination of both treatments. The treatment preferably is performed in such a way that a population of mutagenized cells is obtained wherein a substantial amount of cells has received a single mutation and/or that a mutant strain is provided which is capable of producing enhanced levels of the desired product as compared-to its parent strain when cultured under identical conditions.

DNA-mediated transformation includes a transformation with random fragments of homologous or heterologous nucleic acid and/or a transformation with specifically selected DNA fragments, for instance DNA fragments comprising one or more genes involved in the biosynthesis of a sphingoid base or derivative thereof.

For instance, the genes for the first biosynthetic step, LCB1 and LCB2 (encoding the subunits of the serine palmitoyl transferase) have been cloned (Nagiec et al. (1994), Proc. Natl. Acad. Sci. USA 91, 7899-7902). It may be anticipated that increasing the expression of these genes, for instance by increasing the gene dosage, could lead to a higher capturing rate of carbon from the central metabolic pathways into the sphingoid base biosynthetic sequence. Similarly, overexpression of SUR2 (encoding the hydroxylase that introduces the 4-OH group of phytosphingosine; Haak et al. (1997), J. Biol.

Chem. 272,29704-29710) and LCB3 (involved in the transport of sphingoid bases; Qie et al. (1997), J. Biol. Chem. 272,16110-16117) could lead to higher sphingoid base productivity. In addition a number of genes are known to be involved in the biosynthesis of ceramides from sphingoid bases, such as EL02 and EL03 (involved in the biosynthesis of very-long-chain-fatty acids; Oh et al. (1997), J. Biol. Chem. 272,17376-17384) and AUR1 (involved in the coupling of ceramide to the inositol moiety; Nagiec et al. (1997), J. Biol.

Chem. 272, 9809-9817). The effect of modulating the expression of these genes may be harder to predict, but it is a distinct possibility that down-tuning of this expression could lead to a larger fraction of sphingoid base being acetylated and excreted in a retrievable form.

In a preferred embodiment of the invention, a suitable parent strain to be subjected to a mutagenic treatment is a strain of the genus Pichia, more preferably a strain of the species Pichia ciferrii, most preferably a strain of the species Pichia ciferrii Y-1031 F-60-10 or a derivative thereof.

The present invention thus provides a method for the isolation of a microbial strain capable of producing enhanced levels of a sphingoid base or a derivative thereof as compared to a reference strain.

In particular, microbial strains are provided which are capable of producing enhanced levels of sphingosines and/or keto, glycosylated or acetylated derivatives thereof including 3-ketosphingosine, triacetylsphingosines, diacetylsphingosines and N-acetylsphingosine; dihydrosphingosines and/or keto, glycosylated or acetylated derivatives thereof including 3-ketodihydrosphingosine, triacetyldihydrosphingosines, <BR> <BR> diacetyldihydrosphingosines and N-acetyldihydrosphingosine ; and/or phytosphingosines and/or keto, glycosylated or acetylated derivatives thereof including tetraacetylphytosphingosine, triacetylphytosphingosines, diacetylphytosphingosines and N-acetylphytosphingosine. Preferred strains are strains capable of producing enhanced levels of phytosphingosines and partially or fully acetylated derivatives thereof such as triacetylphytosphingosines (TriAPS) and tetraacetylphytosphingosine (TAPS).

The descriptions of the growth conditions, as detailed in the Examples, are provided as a reference for the determination of the production levels of the desired products by the strains of the present invention. However, other culture conditions, which are conducive to production of the desired products, as known to the skilled artisan, may also be used without departing from the spirit of the present invention.

For instance, strains displaying a certain enhancement in sphingoid base productivity as compared to a reference strain when analyzed in a shake flask culture typically display a substantially similar enhancement when measured on a larger scale.

Quantitative analysis may be performed by various methods known in the art. For instance, quantitative analysis on the basis of TAPS production levels may be performed on culture supernatants of shake flasks cultures, said culture supernatants being extracted with a suitable solvent, such as CDCI3 and analyzed by NMR. TAPS is preferably used as a standard for NMR analysis.

Other desired products such as sphingosine, phytosphingosine and dihydrosphingosine; triacetyl, diacetyl and N-acetyl derivatives of sphingosine, phytosphingosine and dihydrosphingosine; glycosylated derivatives of sphingosine, phytosphingosine, and dihydrosphingosine; 3-keto dihydrosphingosine and/or 3-ketosphingosine may be quantitatively determined in the same manner and the TAPS standard used as the reference.

In an embodiment of the invention, Pichia strains, preferably Pichia ciferrii strains, are isolated which produce an enhanced amount of TriAPS as compared to the reference strain Y-1031 F-60-10. In a preferred embodiment of the invention, Pichia strains are isolated which produce TriAPS as the main sphingoid base, i. e. with a low or zero level of TAPS being formed.

A further aspect of the present invention relates to a method for the production of a sphingoid base or a derivative thereof comprising the fermentation of a strain obtainable by the method of the invention under conditions conducive to the production of said sphingoid base or derivative thereof. The sphingoid base or derivative thereof thus-produced may be recovered from the fermentation broth, preferably from the fermentation supernatant, using common extraction and recovery technology.

Once the desired product is obtained it may be used directly or may be further processed depending on the desired use. For example, acetylated derivatives of sphingosine, dihydrosphingosine and/or phytosphingosine may be deacetylated either enzymatically or chemically.

Deacetylated sphingoid base derivatives, e. g. sphingosine, dihydrosphingosine and/or phytosphingosine, may be used directly or may subsequently be used as starting materials in the synthesis of commercially

valuable sphingolipid products such as ceramides (see W093/20038), glycoceramides and pseudoceramides.

Direct use of the sphingoid base compounds obtained according to the invention includes the use of said compounds in cosmetic and dermatological compositions.

The following examples are provided so as to give those of ordinary skill in the art a complete disclosure and description of the present invention and are not intended to limit the scope of what the inventors regard as their invention.

Example 1 Test of indicator strains for bioassay of TAPS production The following indicator strains were tested on agar plates for their sensitivity towards a TAPS standard: Staphylococcus aureus ATCC 14458 Streptococcus pyogenes ATCC 12344 Micrococcus luteus ATCC 1 0204 Serratia marcescens ATCC 14041 The four bacterial strains were grown in 100 ml Brain Heart Infusion (BHI) broth in 500 ml shake flasks at 37°C, until the stationary phase was reached. Subsequently, 0.5 ml of the stationary phase culture was mixed with 100 ml molten BHI agar (45°C), containing 1 mM CaCl2. This mixture was poured into Petri dishes, and allowed to solidify. Then sterile metal rings (6 mm internal diameter) were placed on the agar surface, and 50 NI of a 50 mg/ml TAPS stock solution in 50% ethanol was introduced in the ring.

Subsequently, the agar plates were incubated at 37°C, and the zone of growth inhibition extending from the metal rings was assessed. Clear inhibition of growth was found for S. aureus and S. pyogenes, and to a lesser degree for M. luteus. In contrast, growth of S. marcescens was not inhibited by the TAPS solution under these conditions.

Example 2 Bioassav of TAPS production in microtiter plates: primary selection The indicator strain, Staphylococcus aureus ATCC 14458, was stored in deep frozen condition. One freeze tube was used to inoculate a 100 ml shake flask containing 25 ml BHI medium. The culture was incubated overnight (16-17 h) at 37 ° C. Subsequently, the overnight culture was diluted 250-500 times in fresh BHI medium, containing 1 mM CaClz.

The production strains were inoculated from an agar surface into 100 ml shake flasks containing 25 ml of a medium containing 33 g/l dextrose, 1 g/l yeast extract, 4.83 g/I NH4C1, 1.0 g/I KHzP04, 0.88 g/I MgS04. 7aq, 0.06<BR> g/I NaCI, 0.2 g/l CaC12. 2aq, 20 g/l KH-phthalate, 0.3 ml/l of a concentrated trace element solution, 1.5 ml/l of a concentrated vitamin solution and 0.16 ml/I Structol antifoaming agent. After 72h of growth on a rotary shaker (25°C) the flasks were sampled, and the samples were centrifuged for 10 minutes at 4000 rpm.

10 NI of the supernatant fraction was transferred to a well of a 96-well flat-bottom microtiter plate. This procedure was repeated 7 times, in order to fill 1 array of the microtiter plate with the 8 replicates of 1 sample. On every microtiter plate, 8 replicates of a TAPS standard solution were included, as well as 2*8 medium blanks.

Subsequently, 150 to 200, ul of the diluted suspension of the indicator strain was added to the wells. The microtiter plates were incubated 5-6 h at 37 ° C on a Flow Laboratories"Titertek"shaking incubator (speed 4). After incubation, the optical density of the wells was determined at 620 nm, using an Anthos Htlll microtiter plate reader, and the mean value of the 8 replicates of one strain was taken as a measure of its TAPS production.

If the growth inhibition was too severe, the incubation was extended overnight, at 26 ° C.

Example 3 Use of the bioassay in secondary selection In order to check whether the strains selected by means of the bioassay in the primary selection were indeed higher producers, they were subjected to a secondary selection in triplicate. Each selected strain was inoculated from a deep-frozen glycerol suspension into 500 ml bafifled shake flasks, containing 100 ml of the medium described in Example 2. After 40h of growth on a rotary shaker at 25°C, this seed culture was used to inoculate three 500 ml baffled shake flasks containing 100 ml of the same medium, at a level of 5% by volume. Subsequently these production cultures were incubated for another 48h on the shaker, and sampled. The samples were treated as described in Example 2.

The supernatant fractions were subjected to the bioassay as described in Example 2, but this time the TAPS concentration was also measured directly, using NMR spectroscopy. Strains were selected on the basis of the NMR results, which were also used to assess whether the bioassay was still able to disciminate between the higher production levels. It was found that this was always the case, provided that the sample supernatant was diluted in proportion to the TAPS concentration. Dilutions up to 5-fold were used to tune the bioassay to the higher TAPS-concentrations found in cultures of the more advanced strains.

As an example of the ability of the bioassay to discriminate between strain of different production levels, the results of a number of strains with known production levels are compiled in the following table: Strain Bioassay OD (620) Relative TAPS Production blank 649 - COS 1 A 567 100% COS 6 465 152% COS 10 361 182% COS 17 344 242%

Example 4 Direct comparison of strains COS 1A and COS 10 Two strains were compared: COS1 A, which is a working cell bank of the haploid wild-type strain Pichia ciferri F-60-10, and COS 10, which is a working cell bank of COS 106, one of the strains obtained by chemical mutagenesis and selected by the primary selection described in Example 2.

COS 10 has been deposited at the"Centraal Bureau voor Schimmelcultures" at July 23,1998, as CBS 101070.

The comparison was performed as described in Example 3. The results are presented in the following table: Strain Bioassay OD (620) TAPS by NMR (g/I) COS 1A0.2420~35 (Working Cell Bank of 0.369 0.31 F-60-10)0.1600.41 COS 10"0.1180. 75 (Working Cell Bank of 0.129 0.74 COS 106) 0. 119 0. 79 It is clear that there was a good correlation between the growth inhition in the bioassay and the TAPS production measured by NMR. As a result, a strain has been selected which produces twice as much TAPS as the wild-type strain under the same conditions.

Example 5 Comparison of COS17 and COS144 In subsequent cycles of strain improvement (mutagenesis and selection) the best strain obtained thusfar, COS 17 was used as a control. In this experiment the results of a selected strain, COS 144, are compared with this control, using the procedure described in Example 3, but with 4-fold dilution of the sample supernatant.

Bioassay OD (620) 1 : 4 TAPS by NMR 204 0. 9 COS17 159 0.9 327 0. 6 110 1. 1 COS144 103 1. 2 94 1. 1 ~ It is clear that strain COS 144 showed an increased TAPS production over COS 17, by about 25%, which gives a COS 144 a production level of 302% compared to COS1A. Therefore a working cell bank was prepared of this strain, designated COS19A, which was shown to have the same TAPS production as did COS 144.

Example 6 Ability of the bioassay to select TriAPS producing strains The bioassay can also be used to select TriAPS producers, since the growth inhibition caused by this compound seems to be of the same order of magnitude as for TAPS. By using the procedure described in Example 2, it was found that both strains producing primarily TAPS, producing both TAPS and TriAPS, and producing solely TriAPS were selected, as shown in the following table: Total acetylated TAPS as % of TriAPS as % of Strain PS by NMR as % Total APS Total APS of COS17 COS17 100 90 10 COS141 60 0 100 COS19A 133 80 20 COS 154 213 53 47 It is clear that the productivity of total acetylated phytosphingosine was the highest in COS154, which gives this strain a production level of 515 % compared to the wild-type strain COS1 A.