Biological production of gangliosides in recombinant host cells

Abstract

In various embodiments, provided herein are host cells, methods, and pharmaceutical compositions, including complex sphingolipids, such as monosialic acid ganglioside 1 (GM1), monosialic acid ganglioside 2 (GM2), and monosialic acid ganglioside 3 (GM3). In some embodiments, the present disclosure provides biological production of glycosphingolipids. In still further embodiments, the present disclosure provides for the biological production of gangliosides.

Description

[0001] Related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 489,919, filed March 13, 2023. The entire teachings of the above application are incorporated herein by reference.

[0003] By referencing and incorporating XML materials

[0004] This application incorporates the sequence list contained in the following Extensible Markup Language (XML) file, which was filed concurrently with it:

[0005] a) File name: 57671008001_Sequence_Listing.xml; created on March 12, 2024, size is 100,172 bytes. Background Technology

[0006] Gangliosides are an important class of molecules composed of a lipid-based ceramide backbone and a glycohead group. Despite their numerous important roles in healthcare, the production and purification of gangliosides remain a challenge. Summary of the Invention

[0007] This article describes transgenic yeast and methods for producing complex sphingolipids, such as monosialotetrahexosylganglioside 1 (GM1), monosialotetrahexosylganglioside 2 (GM2), and monosialotetrahexosylganglioside 3 (GM3). In several respects, the production of complex sphingolipids is carried out in transgenic yeast.

[0008] On one hand, this disclosure provides a transgenic yeast comprising a transgene encoding ceramide synthase and a transgene encoding sphingolipid delta-4 desaturase, wherein the yeast does not express the endogenous yeast genes SUR2, SCS7, LAG1, and LAC1. On the other hand, this disclosure provides a transgenic yeast cell comprising ceramide synthase having the sequence SEQ ID NO:4, SEQ ID NO:46, or SEQ ID NO:48. On the other hand, this disclosure provides a transgenic yeast comprising sphingolipid delta-4 desaturase having the sequence SEQ ID NO:2, SEQ ID NO:36, SEQ ID NO:40, SEQ ID NO:42, or SEQ ID NO:44. On the other hand, this disclosure provides a transgenic yeast comprising a transgene encoding UDP-glucosylceramide glucosyltransferase. On the other hand, this disclosure provides a transgenic yeast comprising UDP-glucosylceramide glucosyltransferase having the sequence SEQ ID NO:6 or SEQ ID NO:8. On the other hand, this disclosure provides a transgenic yeast comprising β-1,4-galactosyltransferase. On the other hand, this disclosure provides a transgenic yeast comprising β-1,4-galactosyltransferase that is HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14), or HsB4GALT6 (SEQ ID NO:16). On the other hand, this disclosure provides a transgenic yeast comprising a transgene encoding bacterial Neu5Ac synthase, a transgene encoding bacterial UDP-N-acetylglucosamine 2-epomerase, a transgene encoding mammalian CMP N-acetylneuraminic acid synthase, and a transgene encoding β-galactosidase α-2,3-sialic acid transferase. On the other hand, this disclosure provides a transgenic yeast comprising Neu5Ac, said Neu5Ac having the sequence of SEQ ID NO:20, SEQ ID NO:26, or SEQ ID NO:32. On the other hand, this disclosure provides a transgenic yeast comprising UDP-N-acetylglucosamine 2-epimerase, said UDP-N-acetylglucosamine 2-epimerase having the sequence of SEQ ID NO:22, SEQ ID NO:28, or SEQ ID NO:34. On the other hand, this disclosure provides a transgenic yeast comprising CMP N-acetylneuraminic acid synthase, said CMP N-acetylneuraminic acid synthase having the sequence of SEQ ID NO:24 or SEQ ID NO:30.On the other hand, this disclosure provides a transgenic yeast comprising β-galactosyl α-2-3-sialyltransferase, said β-galactosyl α-2-3-sialyltransferase having the sequence of SEQ ID NO:18. On the other hand, this disclosure provides a transgenic yeast comprising a transgene encoding an N-acetylglucosamine epimerase and a transgene encoding β-1,4N-acetylgalactosamine transferase. On the other hand, this disclosure provides a transgenic yeast comprising a transgene encoding β-1,3-galactosyltransferase.

[0009] On the other hand, this disclosure provides a method for producing sphingolipids in yeast cells. In several aspects, the method includes culturing yeast cells in a culture medium, wherein the yeast cells express a transgene encoding GhLAG1-1 and a transgene encoding DEGS1; and wherein the yeast cells do not express endogenous yeast genes SUR2, SCS7, LAG1, and LAC1. On the other hand, the culture medium contains stearic acid. On the other hand, the yeast cells produce C18 ceramide. On the other hand, the yeast cells include a transgene encoding UDP-glucosylceramide glucosyltransferase. On the other hand, this disclosure provides a yeast cell comprising UDP-glucosylceramide glucosyltransferase having the sequence SEQ ID NO:6 or SEQ ID NO:8. On the other hand, the culture medium comprises glucose. On the other hand, the yeast cells produce glucosylceramide (GlcCer). On the other hand, the yeast cells include a transgene encoding β-1,4-galactosyltransferase. On the other hand, this disclosure provides a yeast cell comprising a β-1,4-galactosyltransferase, wherein the β-1,4-galactosyltransferase is HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14), or HsB4GALT6 (SEQ ID NO:16). On the other hand, the culture medium comprises UDP-galactose. On the other hand, the yeast cell produces lactosylceramide (LacCer). On the other hand, the yeast cell comprises a transgene encoding a bacterial Neu5Ac synthase, a transgene encoding a bacterial UDP-N-acetylglucosamine 2-epomerase, a transgene encoding a mammalian CMP N-acetylneuraminic acid synthase, and a transgene encoding a β-galactosidase α-2,3-sialic acid transferase. On the other hand, this disclosure provides a transgenic yeast comprising Neu5Ac, wherein the Neu5Ac has the sequence of SEQ ID NO:20, SEQ ID NO:26, or SEQ ID NO:32. On the other hand, this disclosure provides a transgenic yeast comprising UDP-N-acetylglucosamine 2-epimerase, said UDP-N-acetylglucosamine 2-epimerase having the sequence of SEQ ID NO:22, SEQ ID NO:28, or SEQ ID NO:34. On the other hand, this disclosure provides a transgenic yeast comprising CMP N-acetylneuraminic acid synthase, said CMP N-acetylneuraminic acid synthase having the sequence of SEQ ID NO:24 or SEQ ID NO:30.On the other hand, this disclosure provides a transgenic yeast comprising β-galactosidase α-2-3-sialyltransferase having the sequence of SEQ ID NO:18. On the other hand, the culture medium comprises CMP-N-acetylneuraminic acid ester (CMP-Neu5NAc). On the other hand, the yeast cells produce monosialotetrahexosylganglioside 3 (GM3). On the other hand, the yeast cells comprise a transgene encoding an N-acetylglucosamine epimerase and a transgene encoding β-1,4N-acetylgalactosamine transferase. On the other hand, the culture medium comprises UDP-N-acetylgalactosamine. On the other hand, the yeast cells produce monosialotetrahexosylganglioside 2 (GM2). On the other hand, the yeast cells comprise a transgene encoding β-1,3-galactosyltransferase. On the other hand, the culture medium comprises galactose. On the other hand, the yeast cells produce monosialotetrahexosylganglioside 1 (GM1). Attached Figure Description

[0010] The foregoing will become apparent from the following more specific description of the exemplary embodiments as shown in the accompanying drawings, in which the same reference numerals refer to the same parts in different views. The drawings are not necessarily drawn to scale, but are intended to illustrate the embodiments.

[0011] Figure 1 The structure of ganglioside GM1. GM1 consists of a ceramide (C18-ceramide) lipid backbone and pentasaccharide head groups. These sugars are sequentially added to the ceramide backbone. Different glycosylation patterns on the ceramide backbone produce different types of gangliosides.

[0012] Figure 2 Flowchart of ganglioside production in a heterologous host. To produce gangliosides, the biosynthesis of C18-ceramide is first achieved. Then, a series of five glycotransferases and two sugar biosynthetic pathways must be introduced to achieve ganglioside biosynthesis, culminating in the key ganglioside GM1. Cer: Ceramide (Structure 1). GlcCer: Glucosylceramide (Structure 2). LacCer: Lactosylceramide (Structure 3). GM3: Monosialotetrazolium ganglioside 3 (Structure 4). GM2: Monosialotetrazolium ganglioside 2 (Structure 5). GM1: Monosialotetrazolium ganglioside 1 (Structure 6). UDP-Glc: UDP-glucose. UDP-Gal: UDP-galactose. CMP-Neu5Ac: N-acetylneuraminic acid (CMP-sialic acid). UDP-GalNAc: UDP-N-acetylgalactosamine.

[0013] Figure 3Overview of sphingolipids in yeast. Based on current understanding of yeast ceramide biosynthesis, specific methods for disrupting the metabolism of Saccharomyces cerevisiae to produce and accumulate C18-ceramides can be understood.

[0014] Figure 4 This work modifies the process required to produce C18-ceramides through yeast ceramide biosynthesis. Building upon previous work, a method for producing C18-ceramides (the major lipid backbone of all gangliosides) in yeast has been realized. Four native yeast genes must be removed, and two genes must be expressed to produce and accumulate C18-ceramides.

[0015] Figure 5 LC-MS / MS chromatogram of human ceramides in yeast. After 5 days of fermentation, lipids from the *Saccharomyces cerevisiae* strain were extracted in methanol and analyzed. The extract from the engineered strain produced a peak at 23.4 min, while the extract from the wild-type strain did not. C18-ceramide standards were used for identification. This figure shows the chromatogram after MS-MS conversion using a filter specifically designed for C18-ceramides.

[0016] Figure 6 This figure depicts the LC-MS / MS chromatogram of glucosylceramide production following heterologous expression of UDP-glucosylceramide glucosyltransferase. After 5 days of fermentation, lipids from the *Saccharomyces cerevisiae* strain were extracted in methanol and analyzed. Both candidate enzymes (UGCG, SEQ ID NO:6 and GCS1, SEQ ID NO:8) produced the expected peak at 5.6 min, consistent with the retention time observed for glucosylceramide standards. This figure shows the chromatogram after MS-MS conversion using filters specifically designed for glucosylceramide.

[0017] Figure 7 This figure depicts the LC-MS / MS chromatogram of lactosylceramide production following heterologous expression of glucosylceramide β-1,4-galactosyltransferase. After 5 days of fermentation, lipids from the *Saccharomyces cerevisiae* strain were extracted in methanol and analyzed. Four candidate enzymes, HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14), and HsB4GALT6 (SEQ ID NO:16), produced the expected peak at 5.4 min. A standard for lactosylceramide was used for identification in the test strain. This figure shows the chromatogram of the MS-MS conversion of lactosylceramide, which was specifically filtered for this purpose.

[0018] Figure 8: A schematic diagram of the engineered GM3 biosynthetic pathway in yeast. Heterologous genes originate from the eukaryotic system, except for NeuC and NeuB, which are derived from bacteria. Arrows with stop symbols indicate knockout.

[0019] Figure 9 This figure depicts the LC-MS / MS chromatograms of GM3 production following heterologous expression of the hybrid CMP-Sia pathway and lactosylceramide α-2-3-sialotyltransferase. After 7 days of fermentation, lipids from a *Saccharomyces cerevisiae* strain containing two cell-free gene plasmids carrying the GM3 biosynthetic pathway were extracted in methanol and analyzed. Combinations of six different genes produced the expected peak at 4.5–4.6 min, while the lactosylceramide-producing strain used as background did not exhibit this peak. Extracts from wild-type *Saccharomyces cerevisiae* were spiked with GM3 and used for identification in test strains. This figure shows the chromatograms of the MS-MS conversion specifically filtered for GM3. Detailed Implementation

[0020] The example embodiment is described below.

[0021] structure

[0022] Structure 1: Ceramide-18 (C18 ceramide NS (N-sphingosine), human ceramide) is a compound having the following structure:

[0023]

[0024] Structure 2: Glucosylceramide is a compound having the following structure:

[0025]

[0026] Structure 3: Lactosylceramide is a compound having the following structure:

[0027] Structure 4: GM3 is a compound with the following structure:

[0028] Structure 5: GM2 is a compound having the following structure:

[0029] Structure 6: GM1 is a compound having the following structure:

[0030] sequence

[0031] Table 1 is a list of sequences used in the experiments described in this paper.

[0032]

[0033]

[0034]

[0035]

[0036] Homologs of GhLag1-1: human CerS1; and Pichia pastoris bar1

[0037] Homologs of DEGS1: Arabidopsis thaliana Δ4-desaturase At4g04930; Candida albicans Des1p; Drosophila melanogaster DES; Schizosoma dsd1; Pichia pastoris DES1

[0038] As used herein, the term "sequence identity" refers, expressed as a percentage, to the extent that two sequences have identical residues at the same positions when aligned to achieve the highest level of identity. For sequence alignment and comparison, typically one sequence is designated as a reference sequence to be compared with the test sequence. Sequence identity between the reference and test sequences is expressed as a percentage of positions along the entire length of the reference sequence where the reference and test sequences share the same nucleotides or amino acids when aligned to achieve the highest level of identity. For example, if the test sequence has identical nucleotide residues at 70% of the same positions along the entire length of the reference sequence when aligned to achieve the highest level of identity, then the two sequences are considered to have 70% sequence identity.

[0039] Those skilled in the art can readily perform sequence alignments for comparison to achieve the highest level of identity using appropriate alignment methods or algorithms. In some cases, alignments may include introduced gaps to provide the highest level of identity. Examples include the local homology algorithm in Smith and Waterman, Advances in Applied Mathematics 2:482 (1981), the homology alignment algorithm in Needleman and Wunsch, Journal of Molecular Biology 48:443 (1970), and the similarity search method in Proceedings of the National Academy of Sciences of the United States of America 85:2444 (1988). Computerized implementations of these algorithms are described in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Avenue, Madison, Wisconsin. GAP, BESTFIT, FASTA, and TFASTA (Dr., Madison, Wis.), and visual inspection (generally see Ausubel et al., Current Protocols in Molecular Biology). In some embodiments, the codon-optimized sequences for efficient expression in different cells, tissues, and / or organisms reflect codon usage patterns in such cells, tissues, and / or organisms and contain conserved (or non-conserved) amino acid substitutions that do not adversely affect normal activity.

[0040] In some aspects, this disclosure provides a ceramide synthase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:4, SEQ ID NO:46, or SEQ ID NO:48. In some aspects, the ceramide synthase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:3, SEQ ID NO:45, or SEQ ID NO:47.

[0041] In some aspects, this disclosure provides a sphingolipid δ-4 desaturase that has at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:2, SEQ ID NO:36, SEQ ID NO:38, SEQ ID NO:40, SEQ ID NO:42, or SEQ ID NO:44. In some respects, the sphingolipid δ-4 desaturase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:1, SEQ ID NO:35, SEQ ID NO:37, SEQ ID NO:39, SEQ ID NO:41, or SEQ ID NO:43.

[0042] In some aspects, this disclosure provides a UDP-glucosylceramide glucosyltransferase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:6 or SEQ ID NO:8. In some aspects, the UDP-glucosylceramide glucosyltransferase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:5 or SEQ ID NO:7.

[0043] In some aspects, this disclosure provides a β-1,4-galactosyltransferase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:10, SEQ ID NO:12, SEQ ID NO:14, or SEQ ID NO:16. In some aspects, the β-1,4-galactosyltransferase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:9, SEQ ID NO:11, SEQ ID NO:13, or SEQ ID NO:15.

[0044] In some aspects, this disclosure provides a Neu5AC synthase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:20, SEQ ID NO:26, or SEQ ID NO:32. In some aspects, the Neu5AC synthase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:19, SEQ ID NO:25, or SEQ ID NO:31.

[0045] In some aspects, this disclosure provides a UDP-N-acetylglucosamine 2-epimerase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the amino acid sequence of SEQ ID NO:22, SEQ ID NO:28, or SEQ ID NO:34. In some aspects, the UDP-N-acetylglucosamine 2-epimerase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:21, SEQ ID NO:27, or SEQ ID NO:33.

[0046] In some aspects, this disclosure provides a CMP N-acetylneuraminic acid synthase having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the following amino acid sequences: SEQ ID NO:24 or SEQ ID NO:30. In some aspects, the CMP N-acetylneuraminic acid synthase is encoded by a nucleotide sequence having at least about 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.8%, or 99.9%) identity with the nucleotide sequence of SEQ ID NO:23 or SEQ ID NO:29.

[0047] introduction

[0048] Gangliosides are an important class of molecules found throughout the human body, especially in the brain. They consist of a lipid-based ceramide backbone (also known as human ceramide or ceramide NS) and a glycohead group (Sipione et al. 2020). These molecules play various important roles in human health. GM1 (monosialidoganglioside 1) is one of the most abundant gangliosides in brain tissue, and its depletion is associated with neurological diseases such as Huntington's disease and Parkinson's disease (Magistretti et al. 2019). Figure 1 (Structure 6). It has been shown that administration of GM1 can alleviate the symptoms and progression of these diseases (Shneider et al. 2010, Alpaugh et al. 2017). Other gangliosides, such as GM3, GD3 and GD2, are abundant on the surface of cancer cell membranes (Dobrenkov et al. 2016) and are an important component of emerging cancer treatments based on immunotherapy (Grabowska et al. 2021, Zang et al. 2022) (Structure 4).

[0049] Despite the important applications of gangliosides in many aspects of healthcare, their production and purification remain a challenge. Currently, GM1 is used as a therapeutic agent in several countries, but this GM1 is derived from animal brain tissue. Furthermore, less abundant gangliosides are difficult to accumulate and purify. Therefore, solutions for providing gangliosides at a scale suitable for commercialization and pharmaceutical use are necessary.

[0050] To address issues related to the production of GM1 and other gangliosides, this disclosure provides a synthetic biology platform for the production of gangliosides in a heterologous yeast host. Figure 2 The genes required for host cells to produce gangliosides de novo from simple carbon sources, as well as key host genes that must be eliminated to achieve high levels of biosynthesis, were identified. These engineering efforts have now yielded a strain capable of producing ganglioside GM3.

[0051] This synthetic biology strategy for producing gangliosides not only generates GM1, which has been shown to have therapeutic effects, but also produces ganglioside intermediates and alternatives. These include GM2, GM3, lactosylceramide, and glucosylceramide. To date, these gangliosides have never been produced in heterologous organisms.

[0052] Under specified fermentation conditions, the designed strains can accumulate gangliosides for final purification and use in various forms, including direct administration as a treatment and incorporation into emerging drug delivery methods to elicit specific immune responses.

[0053] This article describes a yeast strain capable of producing glucosylceramide, the first glycosylated molecule in the pathway to GM1 biosynthesis. It also describes the subsequent addition of a UGT enzyme to the strain to enable the production of the pentasaccharide head group of GM1.

[0054] General methods

[0055] Culture medium and preparations

[0056] Yeast peptone glucose (YPD) fermentation broth consists of 10 g / L yeast extract, 20 g / L peptone, and 20 g / L glucose. For YPD plates, 20 g / L agar is added.

[0057] Prepare a selective medium using 6.7 g / L of amino acid-free yeast nitrogen bases and 1.92 g / L of synthetic complete medium free of uracil, tryptophan, or histidine, or 1.6 g / L (if leucine is used for selection). For plates, add 20 g / L agar. After sterilization, add glucose to a final concentration of 20 g / L.

[0058] Nitrogen-limiting synthetic medium (NLS) consists of 1.7 g / L yeast nitrogen bases without amino acids and ammonium sulfate, 1 g / L ammonium sulfate, 1.92 g / L amino acid drops without uracil, and 76 mg / L uracil. After autoclaving, 50 ml of 40% glucose or galactose is added to obtain glucose-containing nitrogen-limiting synthetic medium (NLSD) or galactose-containing nitrogen-limiting synthetic medium (NLSGal_). The pH of the medium is then adjusted to 7.0. A solution of 25 mg / ml stearic acid (SA) in 95% ethanol is prepared, and 20 ml is slowly added while the medium is vigorously stirred.

[0059] Genome deletion program

[0060] Deletion of the *Saccharomyces cerevisiae* gene was performed according to Akada et al. (2006). Briefly, a 1kb fragment directly upstream of the desired gene to be knocked out was used to guide the URA3 expression cassette to the knockout locus. Growth of the strain in a uracil-deficient medium indicated the removal of the gene. The URA3 cassette was subsequently removed by growing the strain in a medium containing 5-fluoroorotic acid (5-FOA), which facilitated homologous recombination promoted by repetitive sequences designed into the genomic insert. Knockout was confirmed by genomic DNA extraction and PCR. For CRISPR-Cas9-based deletions, the method was modified by Lee et al. (2015). Briefly, a guide RNA was designed with 65bp flanking regions homologous to the genome directly upstream and downstream of the gene to be removed. A 20bp barcode was added for cloning confirmation. Plasmids containing the guide RNA and the Cas9 mechanism were transformed into recipient strains and plotted on selective media. Knockout was confirmed by PCR. The positive clones were re-marked to remove the Cas9 plasmid, thus successfully knocking out the target gene.

[0061] Cloning program

[0062] As previously described, the Golden Gate assembly method was used to construct integrative and free gene plasmids (Lee et al. 2015). For integrative plasmids, intermediate plasmids containing expression cassettes for testing genes at different promoter strengths and plasmids containing expression cassettes with homologous arms and auxotrophic markers (e.g., URA3, HIS3, LEU2) were constructed.

[0063] To construct cell-free gene plasmids carrying pathway modules, three types of plasmids were constructed: plasmids containing expression cassettes for testing genes under the control of relatively strong promoters, plasmids containing elements required for yeast reproduction, namely antibiotic resistance and a 2-micrometer plasmid backbone, and plasmids containing multiple expression cassettes for cell-free gene expression.

[0064] In short, 5 μL of the Jinmen reactant is transformed into 20–50 μL of *E. coli* DH-α, and after recovery, plated using an appropriate selection plate. The plates are incubated overnight at 37°C, and typically two clones are selected using restriction enzymes for correct assembly. Clones showing the expected pattern are further confirmed by DNA sequencing.

[0065] Conversion program

[0066] Cells were transformed using either chemical transformation as described by Schietz and Gietz (1989) or electroporation as described by Shao and Zhao (2014). In both methods, seed cultures grown overnight were inoculated into 50 ml of YPD medium in 250 ml baffled flasks. Cells were allowed to grow for at least two replicates (i.e., 4–5 hours). For chemical transformation, cells were centrifuged at 2500 rpm for 5 minutes. Cells were washed with water and then with 100 mM lithium acetate. The cells were then concentrated approximately 100-fold in 100 mM lithium acetate. A transformation mixture consisting of 50 mL of cell suspension, 240 mL of 50% PEG, 36 mL of 1 M lithium acetate, and 50 mL of salmon sperm (2 mg / ml) was combined with 50 mL of digestion reaction mixture, which produced a linear DNA fragment containing an expression cassette and selection marker. Heat shock was performed by incubating the cells at 30 °C for 30 minutes, followed by incubation at 42 °C for 30 minutes. Centrifuge the conversion mixture and gently wash the precipitate with 1 mL of sterile water. Finally, resuspend the cells in 150 mL of sterile water and plate them onto appropriate selection plates.

[0067] Cells collected and processed by electroporation were washed twice with water at 4000 rpm for 10 minutes each time, followed by washing with sorbitol, which concentrated the cells approximately 150-fold. The cells were resuspended in sorbitol, and 50 mL aliquots were mixed with less than 4 mL of DNA transformation mixture (i.e., a pair of free gene plasmids). Cells were electroporated at 1.5 kV using 2 mm gap cuvettes. Immediately after electroporation, 1 mL of YPD was added, and the cells were recovered at 30 °C for 1 hour before washing twice with sorbitol and plated on appropriate selective media. For integrated strains, plasmid maintenance was confirmed by PCR or by plated on selective media.

[0068] Analysis of sphingolipids (ceramides, glucosylceramides, and lactosylceramides) by LC-MS

[0069] Typically, 1-2 colonies are inoculated into liquid culture medium and cultured for 5-7 days to assess the production of the target molecule using a modified version of the previously described protocol (Bielawski et al. 2009).

[0070] To detect lactosylceramide, each strain was inoculated into culture tubes containing 5 ml of NLSD + 0.05% SA and NLSGal + 0.05% SA. Liquid culture was carried out at 30°C and 250 rpm for 5 days. Samples were collected for analysis after 3 and 5 days of fermentation.

[0071] To detect GM3, each strain was inoculated into a 50 ml tube with a vent cap containing 10 ml of YPD, YPGal, or NLSGal + 0.05% SA medium. Liquid culture was carried out at 30°C and 250 rpm for 7 days. Samples were collected for analysis after 2.5, 4.5, and 7.5 days of fermentation.

[0072] In Bielawski et al. 2009 3 Lipid extraction was performed as described above. Briefly, 1 mL of cell culture was centrifuged at maximum speed for 5 minutes. The cell pellet was then resuspended in the yeast lipid extraction mixture. The sample was then sonicated for 3 minutes and briefly vortexed. The sample was centrifuged at maximum speed for 5 minutes, and the extract (approximately 1 mL) was collected. 500 mL of the yeast lipid extraction mixture was added to the remaining pellet, and the process was repeated, except that the sonication time was reduced to 1 minute. After centrifugation, the two extracts were combined and incubated overnight in a centrifuge concentrator at moderate temperature. The lipid extract was reconstituted in 0.15–1 mL of mobile phase B for analysis.

[0073] Starting with the previously described LC-MS parameters (Bielawski et al. 2009), MS / MS splitting was used to verify the identity of each ganglioside and to compare each putative product with actual standards.

[0074] Results and discussion

[0075] Engineered production of C18 ceramide

[0076] To obtain gangliosides, the production of human ceramide (C18 ceramide) (structure 1) was engineered. Yeast has produced various ceramides, and the pathways for this production have been previously characterized. Figure 3 By comparing the main yeast ceramide with ceramide that is part of GM1, key changes that must be made were identified. Figure 4First, the genes that attach hydroxyl groups to yeast ceramides (genes sur2 and scs7) must be removed, and this modification, as shown by Murakami et al., is non-toxic, producing ceramides without hydroxyl groups (Murakami et al. 2015). Second, desaturation must be introduced into the sphingosine moiety of the ceramide, as also shown by Murakami and colleagues by adding a human sphingolipid δ-4 desaturase (DEGS1; SEQ ID NO:2) and adding an endoplasmic reticulum targeting sequence to the C-terminus. See U.S. Patent No. 8,367,375 (Kodama et al.). However, Kodama et al. did not disclose modifications to the length of the lipids on the ceramide. Finally, the lipids attached to the sphingosine must be reduced from C26 (typical yeast lipid length) to C18. It has been previously shown that introducing the plant ceramide synthase GhLag1-1 (SEQ ID NO:4) and knocking out the yeast native ceramide synthases Lag1 and LAC1 can produce ceramides with a C18 chain length (Epstein et al. 2012). With this information, a strain was constructed using homologous recombination by first knocking out two hydroxylases, sur2 and scs7, that attach hydroxyl groups to yeast ceramides. Next, DEGS1 (SEQ ID NO:2) and GhLAG1-1 (SEQ ID NO:4) were integrated into the locus. Finally, Lag1 and LAC1 were removed via CRISPR-Cas9 technology to obtain a strain capable of producing C18 ceramides. When this strain was cultured in a medium containing stearic acid, the production of C18 ceramides was observed by LC-MS. This was verified by comparison with standards. Figure 5 Although ceramides with longer chain lengths have been previously produced in yeast (Murakami et al. 2015), this is the first demonstration of heterologous production of C18 ceramides, which serve as the ceramide backbone of gangliosides and are identical to those found in mammals.

[0077] Natural ceramides play many important roles in normal yeast cellular processes, such as membrane stability and the formation of more complex (e.g., modified) lipids. Most notably, inositol phosphorylceramide plays a crucial role in cell growth and division. Knocking out four endogenous genes (SUR2, SCS7, LAG1, and LAC1) and expressing two exogenous genes (GhLAG1-1 and DEGS1) prevents yeast cells from producing natural ceramides. The resulting sphingolipids, compared to natural ceramides, have double bonds, which introduce rigidity to the molecular chain and could disrupt membrane formation via sphingolipids. Furthermore, the addition of two hydroxyl groups (one on each molecular chain) is also eliminated. These hydroxyl groups are known to be important for maintaining membrane fluidity and the normal circulation of membrane components in yeast cells. Previous studies have shown that the introduction of double bonds and the removal of hydroxyl groups do not induce cell death but significantly increase sensitivity to inhibitors of enzymes that produce inositol phosphorylceramide. Observations have shown that removing two hydroxyl groups by knocking out sur2 and scs7 reduces growth rate, indicating that the substitution of ceramides causes some form of growth defect or deficiency (Murakami et al., 2015). Furthermore, in the strain, the length of the fatty acyl chain of ceramide has been reduced from 26 carbons to 18 carbons, which affects membrane fluidity and membrane structure. This has been independently shown not to be toxic to yeast. However, it has been noted that ceramide production increases significantly when the fatty acyl chain is reduced from 26 carbons to 18 carbons (Epstein et al., 2012). Therefore, it is predicted that the introduction of structural changes and shortening of the acyl chain length for ceramides will further increase the number of non-natural ceramide species, causing the growth defects reported by Murakami et al. However, no further growth defects were observed because more non-natural ceramide species accumulated in the engineered strains. Therefore, the combination of ceramide modifications described herein can produce C18 ceramides, and for the first time demonstrates that modification of the sphingosine base of ceramide and the fatty acyl chain of endogenous yeast ceramide does not cause cell death or have any other negative impact on cell viability.

[0078] Engineering of glucosyl ceramide production

[0079] Having identified the strain producing C18 ceramide, the next step in ganglioside biosynthesis is to transfer glucose to ceramide to form glucosylceramide (structure 2). This requires the action of UDP-glucosylceramide glucosyltransferase (UGCG). Two different enzymes (UGCG, SEQ ID NO:6; GCS1, SEQ ID NO:8) were screened for activity by cloning them into an input vector under a strong constitutive promoter pTDH3 and integrating them into the genome of the C18-ceramide-producing strain. After culturing the newly constructed strain, the ceramide profile was analyzed by LCMS, and the presence of glucosylceramide was detected. This activity was validated by comparison with actual glucosylceramide standards. Figure 6 This is the first description of the heterologous production of glucosylceramide in a single-celled host, which is identical to the glucosylceramide found in mammalian gangliosides.

[0080] Engineering of lactosylceramide production

[0081] The next glycotransfer reaction in ganglioside production involves the addition of a galactose unit to glucosylceramide, which is accomplished by β-1,4-galactosyltransferase (B4GalT) to form lactosylceramide (structure 3). The B4GalT enzyme was cloned into an input vector under the strong constitutive promoter pTDH3 and integrated into the genome of a strain producing glucosylceramide. To assess activity, the resulting strain was grown in a medium supplemented with stearic acid and galactose. Sample extraction and measurements by LCMS revealed that four enzymes (HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14), and HsB4GAL6 (SEQ ID NO:16)) were capable of converting glucosylceramide to lactosylceramide (…). Figure 7 This is the first description of the heterologous production of lactosylceramide in a single-celled host, which is identical to the lactosylceramide found in mammalian gangliosides.

[0082] GM3 production engineering

[0083] To achieve the biosynthesis of GM3, two different pathways were engineered. GM3 contains the Neu5Ac (also known as sialic acid or N-acetylneuraminic acid ester) sugar moiety, which is not a naturally occurring sugar in the Saccharomyces cerevisiae host (structure 4). Therefore, the biosynthesis of the sugar donor CMP-Neu5Ac was first engineered into the yeast host. US2008 / 0085540 A1 (Hamilton) discloses the engineering of CMP-Neu5Ac biosynthesis into the yeast host Pichia pastoris to enable sialylation of the protein, and a similar approach was adopted in this paper. To achieve this biosynthesis, a combination of bacterial and mammalian CMP-Neu5Ac biosynthetic genes was heterologously expressed on plasmids in yeast strains. This included bacterial Neu5Ac synthase (NeuB), bacterial UDP-N-acetylglucosamine 2-epomerase (NeuC), and mammalian CMPN-acetylneuraminic acid synthase (CMAS). Simultaneously, β-galactosidase α-2,3-sialyltransferase was expressed on a separate plasmid, which transfers Neu5Ac sugar from CMP-Neu5Ac to lactosylceramide to produce GM3. The lactosylceramide-producing strain was transformed using the CMP-Neu5Ac biosynthetic pathway and β-galactosidase α-2,3-sialyltransferase, and then cultured to evaluate GM3 production. Figure 8 After growth and sample preparation, six gene combinations were found to induce GM3 production. For each combination, the production of GM3 was determined by using a strain derived from Neisseria meningitidis. Figure 9 T1 and T4 in the middle), agalactococcus ( Figure 9 T2 and T5 in the middle) or Escherichia coli ( Figure 9 The biosynthesis of Neu5Ac is achieved through bacteria NeuB and NeuC (T3 and T6 in T6). This is accomplished via mammalian CMAS (derived from mice or humans). Figure 9 The T1-T3 and T4-T6 groups respectively promote the activation of Neu5Ac to CMP-Neu5Ac, and in all combinations, human β-galactosidase α-2,3-sialyltransferase is used for GM3 biosynthesis. GM3 production was confirmed by comparison with purchased actual standards. Figure 9 This is the first description of heterologous production of GM3 in a single-celled host.

[0084] Table 2 is a list of the six gene combinations used.

[0085]

[0086] Engineered GM2 biosynthesis (a predictive example)

[0087] To achieve the biosynthesis of GM2, two genes need to be engineered into the strain that produces GM3 (Structure 5). GM2 biosynthesis requires the carbohydrate precursor UDP-N-acetylgalactosamine (UDP-GalNAc), which is not naturally present in *Saccharomyces cerevisiae*. Therefore, UDP-galactosyl / UDP-N-acetylglucosamine epimerases (GALE), such as BsGALE (SEQ ID NO:50), MmGALE (SEQ ID NO:52), and HsGALE (SEQ ID NO:54), will be introduced into the host. Secondly, β-1,4N-acetylgalactosamine transferases, such as HsB4GalNT1 (SEQ ID NO:56) or CjCgtA (SEQ ID NO:62), will subsequently be engineered to facilitate the transfer of GalNAc to GM3. Together, these two genes will enable the biosynthesis of GM2 (the penultimate precursor of GM1).

[0088] Engineered GM1 biosynthesis (a predictive example)

[0089] In order to engineer a strain capable of producing the therapeutic molecule GM1, a final gene addition must be performed. Figure 1 (Structure 6). β-1,3-galactosyltransferases, such as HsB3GalT4 (SEQ ID NO:60) or CjGgtB (SEQ ID NO:62), will be engineered into strains that produce GM2, and GM2 will be converted to GM1 by utilizing UDP-galactose. This result will be the first description of GM1 biosynthesis in a heterologous host for drug applications targeting the treatment of neurological diseases and symptoms.

[0090] Enhanced biosynthesis of UDP-galactose and engineered UDP-N-acetylgalactosamine (predictive example)

[0091] In order to provide the sugar donor molecules necessary for ganglioside biosynthesis, UDP-galactose and UDP-N-acetylgalactosamine must be present in the host cell. Figure 2 Yeast produces UDP-galactose, but it does not accumulate in large quantities. However, UDP-N-acetylgalactosamine is not produced at all in *Saccharomyces cerevisiae*. Therefore, three dual-function UDP-glucose / UDP-N-acetylgalactosamine epimerases (GALEs), namely BsGALE (SEQ ID NO:50), MmGALE (SEQ ID NO:52), and HsGALE (SEQ ID NO:54), were identified that can convert UDP-glucose and UDP-N-acetylgalactosamine into UDP-galactose and UDP-N-acetylgalactosamine, respectively. This would be an important gene addition in the study of the biosynthesis of gangliosides GM2 and GM1.

[0092] References

[0093] Zang, H.; Siddiqui, M.; Gummuluru, S.; Wong, WW; Reinhard, BM. Ganglioside-Functionalized Nanoparticles for Chimeric Antigen Receptor T-Cell Activation at the Immunological Synapse. ACS Nano, 2022, 16(11), 18408–18420. doi.org / 10.1021 / acsnano.2c06516.

[0094] Galleguillos, D.; Wang, Q.; Steinberg, N.; Zaidi, A.; Shrivastava, G.; Dhami, K.; Daskhan, GC; Schmidt, EN; Dworsky-Fried, Z.; Giuliani, F.; Churchward, M.; Power, C.; Todd, K.; Taylor, A.; Macauley, MS; Sipione, S. Anti-Inflammatory Role of GM1 and Other Gangliosides on Microglia. Journal of Neuroinflammation, 2022, 19(1), 9. doi.org / 10.1186 / s12974-021-02374-x.

[0095] Grabowska, J.; Stolk, DA; Nijen Twilhaar, MK; Ambrosini, M.; Storm, G.; van der Vliet, HJ; de Gruijl, TD; van Kooyk, Y.; den Haan, JMM. A liposomal nanovaccine containing α-galactosylceramide and ganglioside GM3 stimulates robust CD8+ T cell responses via CD169+ macrophages and CDC1. Vaccines, 2021, 9(1), 56. doi.org / 10.3390 / vaccines9010056.

[0096] Sipione, S.; Monyror, J.; Galleguillos, D.; Steinberg, N.; Kadam, V. Gangliosides in the Brain: Physiology, Pathophysiology and Therapeutic Applications. Frontiers in Neuroscience, 2020, 14, 572965. doi.org / 10.3389 / fnins.2020.572965.

[0097] Patry, RT; Stahl, M.; Perez-Munoz, ME; Nothaft, H.; Wenzel, CQ; Sacher, JC; Coros, C.; Walter, J.; Vallance, BA; Szymanski, CM. Bacterial AB5 Toxins Inhibit the Growth of Gut Bacteria by Targeting Ganglioside-like Glycoconjugates. Nature Communications, 2019, 10(1), 1390. doi.org / 10.1038 / s41467-019-09362-z.

[0098] Magistretti, PJ; Geisler, FH; Schneider, JS; Li, PA; Fiumelli, H.; Sipione, S. Gangliosides: Treatment Avenues in Neurodegenerative Disease. Frontiers in Neurology, 2019, 10.

[0099] Alpaugh, M.; Galleguillos, D.; Forero, J.; Morales, LC; Lackey, SW; Kar, P.; Di Pardo, A.; Holt, A.; Kerr, BJ; Todd, KG; Baker, GB; Fouad, K.; Sipione, S. Disease-Modifying Effects of Ganglioside GM1 in Huntington's Disease Models. EMBO Molecular Medicine, 2017, 9(11), 1537–1557. doi.org / 10.15252 / emmm.201707763.

[0100] Dobrenkov, K.; Ostrovnaya, I.; Gu, J.; Cheung, IY; Cheung, N.-KV. Oncotargets GD2 and GD3 Are Highly Expressed in Sarcomas of Children, Adolescents, and Young Adults. Pediatric Blood & Cancer, 2016, 63(10), 1780–1785. doi.org / 10.1002 / pbc.26097.

[0101] Murakami, S.; Shimamoto, T.; Nagano, H.; Tsuruno, M.; Okuhara, H.; Hatanaka, H.; Tojo, H.; Kodama, Y.; Funato, K. Producing Human Ceramide-NS by Metabolic Engineering Using Yeast Saccharomyces Cerevisiae. Scientific Reports, 2015, 5(1), 16319. doi.org / 10.1038 / srep16319.

[0102] Lee, ME; DeLoache, WC; Cervantes, B.; Dueber, JE. A Highly Characterized Yeast Toolkit for Modular, Multipart Assembly. ACS Synthetic Biology, 2015, 4(9), 975–986. doi.org / 10.1021 / sb500366v.

[0103] Shao, Z.; Zhao, H. Manipulating Natural Product Biosynthetic Pathways via DNA Assembler. Current Protocols in Chemical Biology, 2014, 6(2), 65–100. doi.org / 10.1002 / 9780470559277.ch130191.

[0104] Kodama, Y.; Okuhara, H.; Funato, K. Methods for Producing Human Ceramide Using Yeast Transformants. US8367375B2, February 5, 2013. patents.google.com / patent / US8367375 / en

[0105] Epstein, S.; Castillon, GA; Qin, Y.; Riezman, H. An Essential Function of Sphingolipids in Yeast Cell Division. Molecular Microbiology, 2012, 84(6), 1018–1032.

[0106] doi.org / 10.1111 / j.1365-2958.2012.08087.x.

[0107] Schneider, JS; Sendek, S.; Daskalakis, C.; Cambi, F. GM1 Ganglioside in Parkinson's Disease: Results of a Five Year Open Study. Journal of the Neurological Sciences, 2010, 292(1), 45–51. doi.org / 10.1016 / j.jns.2010.02.009.

[0108] Bielawski, J.; Pierce, JS.; Snider, J.; Rembiesa, B.; Szulc, ZM.; Bielawska, A. Comprehensive quantitative analysis of bioactive sphingolipids by high-performance liquid chromatography–tandem mass spectrometry. Lipidomics; edited by Armstrong, D.; Methods in Molecular Biology; Humana Press: Totowa, NJ, 2009; Vol. 579, pp. 443–467 doi.org / 10.1007 / 978-1-60761-322-0_22.

[0109] Hamilton, S. Method of Engineering a Cytidine Monophosphate-Sialic Acid Synthetic Pathway in Fungi and Yeast. US20080085540A1, April 10, 2008.

[0110] patents.google.com / patent / US20080085540A1 / en

[0111] Akada, R.; Kitagawa, T.; Kaneko, S.; Toyonaga, D.; Ito, S.; Kakihara, Y.; Hoshida, H.; Morimura, S.; Kondo, A.; Kita, K. PCR-Mediated Seamless Gene Deletion and Marker Recycling in Saccharomyces Cerevisiae. Yeast, 2006, 23(5), 399–405. doi.org / 10.1002 / yea.1365.

[0112] Schiestl, RH; Gietz, RD. High Efficiency Transformation of Intact Yeast Cells Using Single Stranded Nucleic Acids as a Carrier. *Curr Genet*, 1989, 16(5–6), 339–346. doi.org / 10.1007 / BF00340712.

[0113] By reference; equivalent

[0114] All teachings of the patents, published applications and references cited in this article are incorporated herein by reference in their entirety.

[0115] Although exemplary embodiments have been specifically shown and described, those skilled in the art will understand that various changes in form and detail may be made thereto without departing from the scope of the embodiments covered by the appended claims.

Claims

1. A transgenic yeast cell comprising a transgene encoding a ceramide synthase and a transgene encoding a sphingolipid δ-4 desaturase; wherein the yeast cell does not express endogenous yeast genes SUR2, SCS7, LAG1, and LAC1.

2. The transgenic yeast cell according to claim 1, wherein the ceramide synthase is SEQ ID NO:4, SEQ ID NO:46 or SEQ ID NO:

48.

3. The transgenic yeast cell according to claim 1 or 2, wherein the sphingolipid δ-4 desaturase is SEQ ID NO:2, SEQ ID NO:36, SEQ ID NO:38, SEQ ID NO:40, SEQ ID NO:42 or SEQ ID NO:

44.

4. The transgenic yeast cell according to any one of claims 1 to 3, further comprising a transgene encoding UDP-glucosylceramide glucosyltransferase.

5. The transgenic yeast cell according to claim 4, wherein the UDP-glucosylceramide glucosyltransferase is SEQ ID NO:6 or SEQ ID NO:

8.

6. The transgenic yeast cell according to claim 4, further comprising a transgene encoding β-1,4-galactosyltransferase.

7. The transgenic yeast cell according to claim 6, wherein the β-1,4-galactosyltransferase is HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14) or HsB4GALT6 (SEQ ID NO:16).

8. The transgenic yeast cell according to claim 6, further comprising a transgene encoding Neu5Ac synthase, a transgene encoding UDP-N-acetylglucosamine 2-epomerase, a transgene encoding CMP N-acetylneuraminic acid synthase, and a transgene encoding β-galactosidase α-2,3-sialic acid transferase.

9. The transgenic yeast cell according to claim 8, wherein the Neu5Ac synthase is SEQ ID NO:20, SEQ ID NO:26 or SEQ ID NO:

32.

10. The transgenic yeast cell according to claim 8, wherein the bacterial UDP-N-acetylglucosamine 2-epimerase is SEQ ID NO:22, SEQ ID NO:28 or SEQ ID NO:

34.

11. The transgenic yeast cell according to claim 8, wherein the mammalian CMP N-acetylneuraminic acid synthase is SEQ ID NO:24 or SEQ ID NO:

30.

12. The transgenic yeast cell according to claim 8, wherein the β-galactosidase α-2-3-sialylate is SEQ ID NO:

18.

13. The transgenic yeast cell according to claim 8, further comprising a transgene encoding N-acetylglucosamine epimerase and a transgene encoding β-1,4N-acetylgalactosamine transferase.

14. The transgenic yeast cell of claim 13, further comprising a transgene encoding β-1,3-galactosyltransferase.

15. A method for producing sphingolipids in yeast cells, the method comprising culturing the yeast cells in a culture medium, The yeast cells expressed transgenes encoding GhLAG1-1 and DEGS1; and The yeast cells in question do not express the endogenous yeast genes SUR2, SCS7, LAG1, and LAC1.

16. The method of claim 15, wherein the culture medium comprises stearic acid.

17. The method of claim 16, wherein the yeast cells produce C18 ceramide.

18. The method according to any one of claims 15 to 17, wherein the yeast cell further comprises a transgene encoding UDP-glucosylceramide glucosyltransferase.

19. The method according to claim 18, wherein the UDP-glucosylceramide glucosyltransferase is SEQ ID NO:6 or SEQ ID NO:

8.

20. The method of claim 18, wherein the culture medium comprises glucose.

21. The method of claim 20, wherein the yeast cells produce glucosylceramide (GlcCer).

22. The method of claim 18, wherein the yeast cell further comprises a transgene encoding β-1,4-galactosyltransferase.

23. The method according to claim 22, wherein the β-1,4-galactosyltransferase is HsB4GALT5 (SEQ ID NO:10), MmB4GALT5 (SEQ ID NO:12), PrB4GALT5 (SEQ ID NO:14) or HsB4GALT6 (SEQ ID NO:16).

24. The method of claim 22, wherein the culture medium comprises galactose (Gal).

25. The method of claim 24, wherein the yeast cells produce lactosylceramide (LacCer).

26. The method of claim 22, wherein the yeast cell further comprises a transgene encoding a bacterial Neu5Ac synthase, a transgene encoding a bacterial UDP-N-acetylglucosamine 2-epomerase, a transgene encoding a mammalian CMP N-acetylneuraminic acid synthase, and a transgene encoding a β-galactosidase α-2,3-sialic acid transferase.

27. The method according to claim 26, wherein the Neu5Ac synthase is SEQ ID NO:20, SEQ ID NO:26 or SEQ ID NO:

32.

28. The method according to claim 26, wherein the bacterial UDP-N-acetylglucosamine 2-epimerase is SEQ ID NO:22, SEQ ID NO:28 or SEQ ID NO:

34.

29. The method of claim 26, wherein the mammalian CMP N-acetylneuraminic acid synthase is SEQ ID NO:24 or SEQ ID NO:

30.

30. The method according to claim 26, wherein the β-galactosidase α-2-3-sialylate is SEQ ID NO:

18.

31. The method of claim 26, wherein the culture medium comprises CMP-N-acetylneuraminic acid ester (CMP-Neu5NAc).

32. The method of claim 31, wherein the yeast cells produce monosialotetrahexosylganglioside 3 (GM3).

33. The method of claim 26, wherein the yeast cell further comprises a transgene encoding an N-acetylglucosamine epimerase and a transgene encoding a β-1,4N-acetylgalactosamine transferase.

34. The method of claim 33, wherein the culture medium comprises N-acetylgalactosamine (GalNAc).

35. The method of claim 34, wherein the yeast cells produce monosialotetrahexosylganglioside 2 (GM2).

36. The method of claim 33, wherein the yeast cell further comprises a transgene encoding β-1,3-galactosyltransferase.

37. The method of claim 36, wherein the culture medium comprises galactose.

38. The method of claim 37, wherein the yeast cells produce monosialotetrahexosylganglioside 1 (GM1).

39. A glycosphingolipid produced by a transgenic yeast cell according to any one of claims 1 to 14 or by a method according to any one of claims 15 to 38.

Images