Constitutive yeast LLP promotor-based expression systems

Abstract

The present invention provides a modified eukaryotic cell wherein the modified eukaryotic cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, the amount of SSN6-like protein being present in the modified eukaryotic cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or essentially no SSN6-like protein is present in the modified cell. Additionally, the present invention provides a polynucleotide sequence comprising a modified ssn6-like gene, and a vector comprising said polynucleoptide. Additionally provided is an expression vector comprising a promoter that is repressed in the presence of SSN6-like protein, and a host cell comprising said vectors. The present invention further refers to a method for determining the purity of a composition by using the modified eukaryotic cell, to a method of expressing gene(s) of interest, and eukaryotic cells comprising modified ssn6-like gene.

Claims

1. A modified Pichia pastoris cell, which is modified compared to its wildtype cell at least in that it comprises a modified ssn6-like gene, and/or a modified expression level of SSN6-like protein, or in that a ssn6-like gene is deleted, respectively to the effect that the modified Pichia pastoris cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, the amount of SSN6-like protein being present in the modified Pichia pastoris cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or essentially no SSN6-like protein is present in the modified Pichia pastoris cell wherein the following definitions apply: the SSN6-like gene is defined as comprising SEQ ID NO: 1, and the modified Pichia pastoris cell, compared to its unmodified wildtype, is defined as exhibiting reduced or no SSN6-like protein activity and/or function in regulating the expression of genes which are under the control of an LLP promoter, with this LLP promoter comprising SEQ ID NO: 133.

2. The modified Pichia pastoris cell according to claim 1, wherein said modified Pichia pastoris cell exhibits reduced SSN6-like-protein activity and/or reduced function, or no SSN6-like-protein activity and/or function at all.

3. The modified Pichia pastoris cell according to claim 1, wherein said wildtype cell contains a SSN6-like protein, which protein comprises one or both of the consensus amino acid sequences depicted in SEQ ID NO: 63 and 64.

4. The modified Pichia pastoris cell according to claim 1, comprising an expression vector comprising a promoter, wherein said promoter is characterized in that it is repressed in the presence of SSN6-like protein if said expression vector is introduced into a suitable expression system, and wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function, wherein an LLP protein, which is encoded by the nucleotide sequence depicted in SEQ ID NO: 16, is not encoded by the polynucleotide sequence of this expression vector.

5. An expression system, comprising a) the modified Pichia pastoris cell as defined in claim 1; b) an expression vector comprising a promoter, wherein said promoter is characterized in that it is repressed in the presence of SSN6-like protein if said expression vector is introduced into a suitable expression system, wherein said expression vector can also be present in linearized form and/or at least parts of the vector being integrated into the genome of the modified eukaryotic cell, and wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function, wherein an LLP protein, which is encoded by the nucleotide sequence depicted in SEQ ID NO: 16, is not encoded by the polynucleotide sequence of this expression vector.

6. The modified Pichia pastoris cell according to claim 1, further comprising a promoter that it is repressed in the presence of SSN6-like protein if an expression vector comprising the promotor is introduced into a suitable expression system, and wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function, wherein an LLP protein, which is encoded by the nucleotide sequence depicted in SEQ ID NO: 16, is not encoded by the polynucleotide sequence of this expression vector, and/or further comprising (a) gene(s) of interest being under control of the promoter, and/or wherein the gene(s) of interest is (are) selected from the group consisting of genes encoding enzymes, antibodies or fragments thereof, hormones, structural proteins, and protein-antigens being present in vaccines.

7. Method for determining the purity of a composition comprising the expression product of a gene of interest, i.e. a protein(s) of interest, comprising the following steps: (a) expressing gene(s) of interest by using the modified Pichia pastoris cell according to claim 1 and by using an expression vector comprising a the LLP promoter, wherein sad promoter is characterized in that it is repressed in the presence of SSN6-like protein if sad expression vector is introduced into a suitable expression system, and wherein LLP protein is not encoded by the polynucleotide sequence of the expression vector, wherein (a1) the modified Pichia pastoris cell comprises a gene encoding the LLP protein under control of the LLP promoter, and wherein (a2) the expression vector comprises one or more gene(s) of interest under control of the LLP promoter, wherein sad gene(s) of interest does (do) not encode the LLP protein, thereby obtaining a composition comprising the expression product of the gene(s) of interest, i.e. the protein(s) of interest, and the LLP protein, which is an expression product of the gene encoding the LLP protein which gene is comprised in the modified Pichia pastoris cell of (a1); (b) determining the amount of the expression product of the gene(s) of interest, i.e. the amount of the protein(s) of interest, and the amount of LLP protein being present in the composition obtained in step (a), wherein the amount of LLP protein compared to the amount of expression product of the gene(s) of interest, i.e. of the protein(s) of interest, is indicative for the purity of the composition obtained in step (a); and, optionally, (c) subjecting the composition of step (a) to one or more downstream purification step(s), followed by step (b) for determining the amount of the expression product of the gene(s) of interest, i.e. the protein(s) of interest, and the amount of LLP protein being present in the composition obtained after having carried out sad downstream purification step, wherein sad LLP promoter being present in the Pichia pastoris cell and being present in the expression vector is the LLP promoter respectively comprising SEQ ID NO: 133 or modified versions thereof, sad modified versions being characterized in that they still exhibit the promoter function, wherein the LLP protein, which is encoded by the nucleotide sequence depicted in SEQ ID NO: 16, is not encoded by the polynucleotide sequence of the expression vector of (a).

8. The modified Pichia pastoris cell according to claim 1, wherein A) the nucleotide sequence of the SSN6-like gene is modified by introduction of a point mutation, a partial or complete deletion, or a replacement by a different nucleotide sequence; and/or B) the expression level of the SSN6-like protein is modified by impairing the transcription or translation of the gene encoding SSN6-like protein.

9. A method for expressing a gene of interest, comprising the steps of: expressing a gene of interest by using the modified Pichia pastoris cell according to claim 1, thus obtaining a recombinant protein expressed by the gene of interest; and purifying the thus-obtained recombinant protein.

10. The method according to claim 9, wherein the modified Pichia pastoris cell exhibits reduced SSN6-like-protein activity and/or reduced function, or no SSN6-like-protein activity and/or function at all.

11. The method according to claim 9, wherein the wildtype cell, compared to which the modified Pichia pastoris cell is modified, contains a SSN6-like protein, which protein comprises one or both of the consensus amino acid sequences depicted in SEQ ID NO: 63 and 64.

12. The method according to claim 9, wherein the modified Pichia pastoris cell comprises a modified ssn6-like gene, wherein said gene has inserted a foreign nucleotide sequence, to the effect that (i) the Pichia pastoris cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, (ii) the amount of SSN6-like protein being present in the eukaryotic cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or (iii) essentially no SSN6-like protein is present in the Pichia pastoris cell, and a gene of interest that replaces a part or all of the coding region of the gene of the Pichia pastoris cell encoding the LLP protein, and that is under control of the LLP promoter, wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function.

13. The method according to claim 9, wherein the modified Pichia pastoris cell comprises a modified ssn6-like gene, wherein said gene has inserted a foreign nucleotide sequence, to the effect that (i) the Pichia pastoris cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, (ii) the amount of SSN6-like protein being present in the Pichia pastoris cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or (iii) essentially no SSN6-like protein is present in the Pichia pastoris cell, a gene of interest under control of the LLP promoter, and an llp gene, wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function.

14. The method according to claim 9, wherein the modified Pichia pastoris cell comprises a SSN6-like gene, wherein the nucleotide sequence of said SSN6-like gene is modified by introduction of a point mutation, a partial or complete deletion, or a replacement by a different nucleotide sequence; and/or wherein the modified Pichia pastoris cell expresses a SSN6-like protein, wherein the expression level of said SSN6-like protein is modified by impairing the transcription or translation of the gene encoding SSN6-like protein.

15. A Pichia pastoris cell comprising a modified ssn6-like gene, wherein said gene has inserted a foreign nucleotide sequence, to the effect that (i) the Pichia pastoris cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, (ii) the amount of SSN6-like protein being present in the eukaryotic cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or (iii) essentially no SSN6-like protein is present in the Pichia pastoris cell, and a gene of interest that replaces a part or all of the coding region of the gene of the Pichia pastoris cell encoding a LLP protein, and that is under control of a LLP promoter, wherein said LLP promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function.

16. A Pichia pastoris cell comprising a modified ssn6-like gene, wherein said gene has inserted a foreign nucleotide sequence, to the effect that (i) the Pichia pastoris cell is not able to provide an SSN6-like protein that exerts its wildtype function and/or wildtype activity, (ii) the amount of SSN6-like protein being present in the Pichia pastoris cell differs from the amount of SSN6-like protein being present in its wildtype form, and/or (iii) essentially no SSN6-like protein is present in the Pichia pastoris cell, a gene of interest under control of the LLP promoter, and an llp gene, wherein said promoter is an LLP promoter comprising SEQ ID NO: 133 or modified versions thereof, said modified versions being characterized in that they still exhibit the promoter function.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1: Polynucleotide Sequences

(2) FIG. 1 shows various polynucleotide sequences.

(3) FIG. 1A shows the nucleotide sequence of the coding region of the ssn6-like gene of P. Pastoris (SEQ ID. NO: 1).

(4) FIG. 1B shows the nucleotide sequence of the LLP promoter of P. pastoris, 605 bp 3′ from the ATG start codon of LLP (SEQ ID NO: 2).

(5) FIG. 1C shows the nucleotide sequence of the signal sequence of the LLP protein of P. Pastoris (SEQ ID NO: 3).

(6) FIG. 1D shows the nucleotide sequence of the terminator sequence of llp gene of P. Pastoris (SEQ ID NO: 4).

(7) FIG. 1E shows the nucleotide sequence of the SSN6-like promoter of P. Pastoris, 1000 bp 3′ from the ATG start codon of SSN6-like (SEQ ID NO: 5).

(8) FIG. 1F shows the nucleotide sequence of the terminator sequence of ssn6-like gene of P. Pastoris (SEQ ID NO: 6).

(9) FIG. 1G shows the nucleotide sequence of the SSN6-like modified DNA of P. Pastoris (from the ATG start to the TAA stop codon) as present in P. Pastoris strain SSS1. This results in a ssn6-like coding region with an internal insertation of a heterologous nucleotide sequence, which disrupts the coding sequence of ssn6-like coding sequence (SEQ ID NO: 7).

(10) FIG. 1H shows the nucleotide sequence of the Kozak start sequence (SEQ ID NO: 11).

(11) FIG. 1I shows the nucleotide sequence of the LLP promoter, 1000 bp 3′ from the ATG start codon of LLP (SEQ ID NO: 12).

(12) FIG. 1J shows the amino acid sequence of the SSN6-like protein (SEQ ID NO: 13)

(13) FIG. 1K shows MFalpha pre-pro signal sequence: FIG. 1K(A) shows said signal sequence without EAEA repeat (SEQ ID NO: 14), and FIG. 1K(B) shows said signal sequence with EAEA repeat (SEQ ID NO: 21).

(14) FIG. 1L shows the codon optimized DNA sequence of encoding a single chain antibody fragment DLX521 (SEQ ID NO: 15).

(15) FIG. 1M shows the coding region of the DNA sequence of the llp-gene including the signal sequence (SEQ ID NO: 16).

(16) FIG. 1N shows the codon optimized nucleotide sequence of the human Growth Hormone (hGH) as used in the examples of this application (SEQ ID NO: 17)

(17) FIG. 1O shows the nucleotide sequence of the human serum albumin (HSA) as used in the examples of this application (SEQ ID NO: 18).

(18) FIG. 1P shows the nucleotide sequence of the penicillin V amidase (PVA) as used in the examples of this application (SEQ ID NO: 19).

(19) FIG. 1Q shows the nucleotide sequence of the vector pGAPk used for generating the SSS1 yeast cell line (plasmid without a GOI and with a Geneticin resistance marker) (SEQ ID NO: 20)

(20) FIG. 1R shows the sequence of a pLLP vector containing a Geneticin resistance marker (pLLPk) corresponding to SEQ ID NO: 22.

(21) FIG. 1S shows SEQ ID NO: 23, which is part of the nuceleotide sequence of chromosome 1 of the genomic sequence of yeast strain YJK_PVA_021 after random integration of a PVA- and a Zeocin-expression cassette into the coding region of ssn6-like (reverse-complement sequence of ssn6-like) at position 807,480 of chromosome 1 of the reference strain Pichia pastoris CBS 7435. The ssn6-like sequence is underlined, the start codon is shown in bold and double-underlined (ATG.fwdarw.reverse-complement.fwdarw.CAT). Shown is the interrupted ssn6-like coding sequence including 10 nucleotides flanking 5′ and 3′ to the ssn6-like coding sequence. The sequence shown is part of the Illumina-sequenced genome of strain YJK_PVA_021 (SEQ ID NOs: 67-115) and the sequence of SEQ ID NO: 23 is part of the genomic sequence SEQ ID NO: 103 obtained by Illumina Inc. Sequences of (reverse-complement sequence of oligo2395) and are labeled in grey.

(22) FIG. 1T shows the LLP signal sequence fused to the 3′ end of HSA coding sequence replacing the native HSA-signal sequence resulting in the sequence according to SEQ ID NO: 121.

(23) FIG. 1U shows SEQ ID NO: 129, which is the shortened LLP-promoter sequence delta29 corresponding to a 576 bases length of the shortened LLP-promoter.

(24) FIG. 1V shows SEQ ID NO: 130, which is the shortened LLP-promoter sequence delta93 corresponding to a 512 bases length of the shortened LLP-promoter.

(25) FIG. 1W shows SEQ ID NO: 131, which is the shortened LLP-promoter sequence delta133 corresponding to a 472 bases length of the shortened LLP-promoter.

(26) FIG. 1X shows SEQ ID NO: 132, which is the shortened LLP-promoter sequence delta201 corresponding to a 404 bases length of the shortened LLP-promoter.

(27) FIG. 1Y shows SEQ ID NO: 133, which is the shortened LLP-promoter sequence delta233 corresponding to a 372 bases length of the shortened LLP-promoter.

(28) FIG. 1Z shows SEQ ID NO: 134, which is the shortened LLP-promoter sequence delta300 corresponding to a 305 bases length of the shortened LLP-promoter.

(29) FIG. 2: Amino Acid Sequences

(30) FIG. 2 shows various amino acid sequences.

(31) FIG. 2A shows the amino acid sequence of the SSN6-like protein of P. Pastoris (SEQ ID NO: 8)

(32) FIG. 2B shows the amino acid sequence of the SSN6-like modified protein of P. Pastoris YJK_PVA_021 (SEQ ID NO: 9). The underlined 7 amino acids represent heterologous, non-SSN6-like amino acids originationg from the vector used to inactivate the SSN6-like protein.

(33) FIG. 2C shows the amino acid sequence of the LLP protein including LLP signal sequence (SEQ ID NO: 10).

(34) FIG. 2D shows the amino acid sequence of the LLP signal sequence of p. pastoris (SEQ ID NO: 140).

(35) FIG. 3: Mechanism of Super-Secretor Pichia Pastoris (SSS1)

(36) wt=wild-type, ssn6-like=ssn6-like gene, LLP=LLP coding sequence, Pl=LLP-promoter, LLP-prot.=LLP-protein, Pg=GAP-promoter, AR=antibiotic resistance, GOI=coding sequence of gene of interest, GOI-prot.=gene of interest protein

(37) FIG. 3A

(38) The wt-strain (wild-type strain, NRRL Y-11430) contains an intact ssn6-like gene which according to our interpretation suppresses the promoter (Pl) of the lectine like protein (LLP). The ssn6-like gene is on the reverse strand of chromosome 1 (coding sequence is from 806379-808244) and the LLP gene is on the forward strand of chromosome 1 (coding sequence is from 2492530 to 2493945).

(39) FIG. 3B

(40) An expression construct with a radomly integrated Pg/AR sequence into the coding sequence of ssn6-like gene of strain NRRL Y-11430 was obtained. The resulting strain contains a disrupted ssn6-like gene (inact. ssn6-like). According to our interpretation, disruption of the ssn6-like resulting in inact. ssn6-like removes the suppressing effect of ssn6-like on the Pl-promoter, threby activating of the Pl-promoter finally resulting in high expression of LLP-protein or of other GOIs (Gene Of Interest) under control of a LLP-promoter.

(41) FIG. 3C

(42) A gene of interest (GOI) can be introduced into the yeast strain of Fig. B by either homologous recombination thereby replacing the coding region of the LLP-gene by the coding sequence of the GOI (FIG. 3C,=Option 1), and resulting in expression of the GOI (GOI-protein), whereas no LLP-protein is produced.

(43) FIG. 3D

(44) Alternatively a gene of interest (GOI) fused to the LLP-promoter (Pl, GOI) can be introduced into the yeast strain of Fig. B by random recombination leaving the LLP intact and inserting the GOI under control of the LLP-promoter (Pl) into a random position of the yeast genome. This results in concomit expression of the GOI (GOI-prot.) and of LLP (LLP-prot.).

(45) FIG. 4: SDS-PAGE Analysis

(46) A codon optimized (done by DNA2.0 Inc., Menelo Park, USA) DNA sequence encoding the model protein DLX521 (an scFv=single chain variable Fragment of an antibody) was inserted into a plasmid with the GAP-promoter (pGAP=pJ905 from DNA2.0 Inc.) between restriction sites EcoRl and Notl, and a plasmid with the LLP-promoter (pLLP) between restriction sites EcoRV and Notl. The pGAP plasmid was linearized with Swal and transformed in NRRL Y-11430 and the pLLP plasmid was linearized with Avrll and transformed in SSS1. For the pGAP plasmid, transformants were PCR screened and 8 clones with a positive PCR signal were randomly picked and used for glycerol stock preparation. For the pLLP plasmid, transformants were PCR screened and 9 clones with a positive PCR signal were randomly picked and used for glycerol stock preparation. The glycerol stocks of the 8 pGAP and 9 pLLP strains were subjected to expression studies at microtiter plate scale (25° C., 350 rpm, 70 h). The OD (optical density) at 600 nm was comparable at harvest (OD was determined at harvest to check for biomass variability). Subsequently, 15 μL of the samples (supernatant mixed with NuPAGE® LDS sample buffer incl. 2-mercaptoethanol according to manufacturers (Invitrogen/Life Technologies) instructions) were loaded onto the gel.

(47) The SDS-polyacrylamide gel (SDS-PAGE) system used was the Novex NuPAGE® Bis-Tris 4-12% grandient gel (Invitrogen/Life Technologies) using the MES-buffer system (Invitrogen/Life Technologieis), The protein molecular weight marker (M) used was from AppliChem GmbH, Darmstadt, Germany. The used Protein Marker VI, prestained (AppliChem), shows the indicated approximate molecular weights in kilo Dalton (kDa) of the marker proteins if separated in a 10% SDS polyarcrylamide gel (SDS-PAGE) using Bis-Tris 10% MES buffer according the the manufacturer AppliChem. Loading volume was 15 μl yeast culture supernatand/lane. All pGAP- and pLLP-strains were pre-tested by PCR for scFv. Only scFv-positive strains were subsequently tested for expression of scFv protein. The table below the SDS PAGE gel lists the protein bands at the molecular weight positions of LLP and scFV, the intensity of the protein band (++ strong band, + clear band, o faint band, − no band) as well as the Option according to FIG. 1 to which this yeast clone belongs (Option 1: only GOI is expressed, no LLP; Option 2: GOI and LLP are expressed) M: Protein Marker VI prestained (AppliChem) Lane 1 scFv control strain that did not grow very well in deep well plates Lanes 2-9: strain YJK_PVA_021, expressing low amounts of PVA under control of the GAP-promoter (faint band of scFV protein can be seen in lanes 2-9) Lanes 10-18: strain PP_ESBA521_010 expressing scFv protein under control of the LLP-promoter. Lanes 10 to 14: strong protein band at the molecular weight of scFv (see←scFv) (“←” indicates the protein band corresponding to the scFv protein) no protein band at the molecular weight of LLP (see←LLP), example of Option 1 (only expresion of GOI) Lanes 15 and 16: strong protein band at molecular weight of LLP (see←LLP) (“←” indicates the protein band corresponding to the LLP protein) and clear band of scFv at molecular weight of scFv (see←scFv), example of Option 2 (parallel expression of GOI and of LLP) Lane 17: clear protein band at molecular weight of LLP (see←LLP) and strong protein band of scFv at molecular weight of scFv (see←scFv), example of Option 2 Lane 18: no protein band at molecular weight of LLP (see←LLP) and strong protein band of scFv at molecular weight of scFv (see←scFv), example of Option 1

(48) The high molecular weight protein band in the gel at the position of about the molecular weight marker 125 kDa represents a dimer of the LLP-protein. The identity of the LLP-protein was proven by the following experiments and evidences. First the LLP-band was cut out of the gel and the C-terminal amino acid sequence was determined by peptide sequencing using standard methods know in the art. Furthermore the LLP-gel band was analyzed by mass spectromety using standard methods known in the art. Both methods proved that this protein band in the gel indeed is LLP-protein. Furthermore LLP was treted with an enzyme removing glyco-strctures from proteins, namely PNGase F. Treatment of LLP with PNGase F lowered the molecular weight of the LLP-dimer from about 125 kDa to about 110 kDa. Furthermore it is known from the literatrue (IUBMB Live, 2009, 61:252-60), that lectins can form very stable multimers such as dimers. Therefore we conclude the high molecular LLP-protein band in our SDS-PAGE gels is a stable LLP-protein dimer.

(49) FIG. 5A: SDS PAGE Analysis

(50) A codon optimized DNA sequence (SEQ ID NO: 17) encoding the model protein hGH was inserted into a plasmid with the GAP-promoter (pGAP) and one plasmid with the LLP-promoter (pLLP). The pGAP plasmid was transformed in NRRL Y-11430 and the pLLP plasmid in SSS1. For pGAP, 16 transformants were randomly picked and subjected to expression studies at microtiter plate scale (25° C., 350 rpm, 70 h). For pLLP, 28 transformants were randomly picked and subjected to expression studies at microtiter plate scale (25° C., 350 rpm, 70 h). The OD at 600 nm was comparable at harvest (OD was determined at harvest to check for biomass variability). Subsequently, 15 μL of the samples (supernatant mixed with NuPAGE® LDS sample buffer incl. 2-mercaptoethanol) were loaded onto the gel. M: Protein Marker VI 10-245 from AppliChem. pGAP: a protein band corresponding to hGH was detected by visual inspection of SDS-PAGE gels in 0/16 randomly picked clones (without prior PCR screening) pLLP: a protein band corresponding to hGH was detected by visual inspection of SDS-PAGE gels in 8/28 randomly picked clones (without prior PCR screening)

(51) This result indicates that the pLLP system is superior to the pGAP system.

(52) FIG. 5B: hGH Western Blot

(53) 15, 20 or 30 μl supernatant samples of deep well plate cultures of Pichia pastoris cells transformed as described under FIG. 5A were directly loaded on SDS-PAGE gels (Novex NuPage 4-12% Bis-Tris Gels from Invitrogen with MES-running buffer). The proteins were then transferred to a PVDF membrane. After transfer and blocking, the membrane was incubated with a solution containing anti-hGH antibody (Zymed Cat.Nr. 18-0090, dilution: 1:1000) for 2 h. After washing the membrane, a secondary antibody coupled with alkaline phosphatase (anti-rabbit IgG alkaline-phosphatase conjugate, Sigma Cat. No. A3687, dilution: 1:16.000) was added for 1 h. After washing the membrane, NBT/BCIP (nitro-blue tetrazolium chloride/5-bromo-4-chloro-3-indolylphosphate toluidine salt) was added to detect hGH. M: Protein Marker VI; NRRL Y-11430 Control: wild type strain without hGH; SSS1 Control: supersecretor strain without hGH; pGAP: expression of hGH under control of GAP-promotor; pLLP: expression of hGH under control of the LLP-promoter

(54) FIG. 6: SDS PAGE Analysis

(55) Lane 6 contains a sample of SSS1-cells not tranfected with a pLLP construct (negative control).

(56) SDS-PAGE analysis. A codon optimized DNA sequence encoding the model protein HSA was inserted into on plasmid with the GAP-promoter (pGAP) and one plasmid with the LLP-promoter (pLLP). The pGAP plasmid was transformed in NRRL Y-11430 and the pLLP plasmid in SSS1. Colony PCR was applied to screen for transformants which were then subjected to expression studies at microtiter plate scale (25° C., 350 rpm, 70 h). The OD at 600 nm was comparable at harvest (OD was determined at harvest to check for biomass variability). Subsequently, 15 μL of the samples (supernatant mixed with NuPAGE® LDS sample buffer incl. 2-mercaptoethanol) were loaded onto the gel. M: Protein Marker VI 10-245 from AppliChem. pGAP: a faint protein band corresponding to HSA was detected by visual inspection of SDS-PAGE gels in 2/5 selected clones (lanes 3 and 5) (with prior PCR screening) pLLP: a protein band corresponding to HSA was detected by visual inspection of SDS-PAGE gels in 4/4 selected clones (with prior PCR screening)

(57) This result indicates that the pLLP system is superior to the pGAP system.

(58) FIG. 7 Comparison of pGAP Expression System with pLLP Expression System

(59) This result indicates that the pLLP system is superior to the pGAP system.

(60) 7A:

(61) Pichia pastoris strains transformed with HSA either under control of the GAP promoter (pGAP =P. pastoris glyceraldehyde-3-phosphate dehydrogenase promotor) or under control of the LLP promoter (pLLP=P. pastoris Lectin Like Protein promotor) were cultured in a 5 liter fermenter for 144 h at 25° C., pH 6.0 and 30% oxygen, using glycerol in batch growth phase and using constant glucose feed in main growth stage. Culture supernatant was analyzed for HSA employing a Human Serum Albumin ELISA (Enzyme Linked Immuno Sorbent Assay) kit from Cygnus Technologies (Cat. Nr. F055) according to the manufacturer's instructions. Mixtures of transformed yeast strains (pGAP) and mixtures of transformed yeast strains (pLLP) were tested in order to account for expression differences in individual strains, e.g. to determine the average expression rate of pGAP- and of pLLP-strains.

(62) 7B:

(63) PVA enzyme assay. A codon optimized DNA sequence encoding PVA (Pleurotus ostreatus-Penicillin V Amidase) was inserted into a plasmid with the GAP-promoter (pGAP) and into a plasmid with the LLP-promoter (pLLP). The pGAP plasmid was transformed in NRRL Y-11430 and the pLLP plasmid in SSS1 yeast cells. For each plasmid, 16 transformants were randomly picked and subjected to expression studies at microtiter plate scale (25° C., 350 rpm, 70 h). The PVA titer (g/kg phenoxy actetic acid) and OD at 600 nm were determined at harvest and used for normalization (g/kg phenoxy normalized). The diagram shows the mean value±SEM (standard error of the mean) of PVA activity as measured by PVA-mediated-conversion of phenoxymethylpenicillin into phenoxy acetic acid g/kg. The highest individual value in each group is labeled with an asterix (*). For expression purpose one would usually choose those strains which do express the highest amount of PVA-activity (labeled with * in the diagram). In this case the best strain expressing PVA under control of the pLLP expressed about 2.9-fold more PVA-activity as compared to the best pGAP strain (pGAP: n=13, max. value was 0.17 g/kg, pLLP: n=15, max. value was 0.49 g/kg), The pGAP vector from DNA2.0 Inc. and the pGAP vector from Sandoz are almost identical. All Vector elements such as promoters, terminators, resistance-marker, pUC ori, etc. are the same in both vectors. Only minor differences regarding the used restriction enzyme sites and short nucleotide sequences which connect some of these elements within the vectors and are slightly different between both vectors.

(64) FIG. 8: Expression of Gene of Interest (GOI) in Fermentor Yeast Cultures

(65) 8A

(66) A codon optimized DNA sequence encoding the enzyme PVA (Penicillin V Amidase) was inserted into a plasmid under the control of the GAP-promoter (pGAP). The pGAP plasmid was transformed in NRRL Y-11430 yeast cells (Pichia pastoris), and cells were grown in a fermentor under standard conditions using standard media and culturing conditions as know in the art. The pGAP vector integrated randomly into the SSN6-gene, thereby activating the LLP-promoter. The superantant of the fermentation culture was subsequently treated with the enzyme PNGase F (Peptide N-Glycosidase F; cleaves asparagine-linked high mannose as well as hybrid complex oligosaccharides from glycoproteins; New England Biolabs, Catalogue Number P0704S). Samples of the untreated cell culture supernatand (lane 1) and of PNGase F-treated cell culture supernatand (lane 2) were separated in SDS-PAGE. Lane 1 shows the glycosylated LLP-dimer as a very strong band at about 125 kDa, whereas the PVA-band is somewhat diffuse at around 72 kDa, probably because individual molecules of PVA are glycosylated to a different extend. Lane 3 shows at about 36 kDa the protein band of the added PNGase F, at about 110 kDa the very strong band of the deglycosylated LLP-dimer (←LLP de-glyc.) and at about 52 kDa a now very clear, distinct band representing declycosylated PVA (←PVA de-glyc.).

(67) 8B

(68) A codon optimized DNA sequence encoding a single chain antibody fragment (scFv), in this case the scFv named DLX521, was inserted into a plasmid under control of the LLP-promoter (pLLP). The pLLP plasmid containing a Geneticin resistance was randomly transformed into YJK_PVA_021(the already PVA-transformed yeast cells of FIG. 8A), and cells were grown in a fermentor under standard conditions using standard media and culturing conditions as know in the art. Supernatant of the fermentation was diluted 20-fold and 15 μl were loaded onto SDS-PAGE in the presence of 2-mercaptoethanol. A very strong protein band at the molecular weight position of the scFv is visible in the gel in lane 1. In addition a minor band at the position of the LLP-dimer is seen at about 125 kDa and a strong protein band of PVA is seen at about 72 kDa. No PNGase was added so all proteins are glycosylated. 20-fold dilution of the cell culture supernatand was done in order to depict the very high expression rate of the scFv relative to the LLP. This yeast strain is an example of the “Option 2” as depicted in FIG. 3D, except that 2 GOI are expressed (PVA and DLX521). PVA is under control of the GAP promoter and randomly integrated by chance into the ssn6-like gene.

(69) This is an example showing, that besides the LLP promoter also other promotors, such as the GAP promotor, can be used in conjuction with the expression system of the invention.

(70) 8C

(71) This scheme shows the expression construct used in FIG. 8A. A codon optimized DNA sequence encoding the enzyme PVA under control of the GAP-promoter was inserted into a plasmid (pGAP), This pGAP plasmid was transformed in NRRL Y-11430 yeast cells (Pichia pastoris), and cells were grown in a fermentor under standard conditions using standard media and culturing. OD600 measurements were used to adjust cell densities. The pGAP vector integrated randomly into the ssn6 gene, thereby activating the LLP-promoter and resulting in an expression of LLP protein.

(72) This expression construct is a proof-of-principle, again showing that interrupting the ssn6 gene results in an improved expression of the gene being under control of the LLP promoter, in this case in an improved expression of the LLP protein, in addition to the expression of the PVA protein.

(73) FIG. 9:

(74) A:

(75) Partial aligment of 39 different protein sequences showing sequence homology to P. pastoris SSN6-like protein (CCA36593.1) (SEQ ID NO: 8)

(76) B:

(77) P. pastoris SSN6-like consensus sequence 1 (SEQ ID NO: 63)

(78) C:

(79) P. pastoris SSN6-like consensus sequence 2 (SEQ ID NO: 64)

(80) FIG. 10: Expression Vector pLLP

(81) This figure shows the expression vector pLLP without inserted GOI (FIG. 10A), and its full length sequence (FIG. 10B) (SEQ ID NO: 65).

(82) Expression vector (without inserted GOI): The gene of interest (GOI) can be inserted into this vector using the restriction endonuclease cleavage sites Notl and/or EcoRV, resulting in the GOI being under control of the LLP promoter and the ADH terminator. In order to insert the expression vector into the yeast genome, the vector can be linearized by cleaving with the restriction endonuclease Avrll, which restriction site is located between the LLP promoter and the LLP terminator sequence. The resulting linearized vector contains at its 5′-end the LLP promoter and at its 3′-end the LLP terminator. If inserted into the yeast genome by homologous recombination via the LLP promoter- and the LLP terminator-sequences, this homologous recombination removes the native LLP-coding sequence from the yeast genome. At the same time the GOI under control of the LLP promoter and the ADH terminator as well as a Zeocin expression cassette and all other parts of the pLLP vector are inserted into the yeast genome.

(83) FIG. 11: Expression Vector pGAP with DLX521 as GOI

(84) This figure shows the expression vector pGAP with DLX521 as GOI (FIG. 11A), and its full length sequence (FIG. 11B) (SEQ ID NO: 66).

(85) Linearization of the vector has been carried out with Swal.

(86) FIG. 12: Expression of HSA Using LLP-Signal Sequence

(87) This figure shows the expression of human serum albumin (HSA) utilizing the secreation signal sequence of LLP. The LLP-signal sequence was fused N-terminally to the coding sequence of HSA (llps-HSA), transformed into yeast cells. 10 randomly picked clones were grown in deep well plates and the supernatant checked for HSA expression (FIG. 12A). One clone was expressed also in a 1 liter fermenter for 48, 72 and 96 hours (FIG. 12B). Expression was shown by SDS-PAGE followed by Coomassie blue staining.

(88) FIG. 13A-D: Analysis of the Functionality of LLP-Promoter Fragments (Truncation Analysis)

(89) The LLP-promoter was successively shortened by cloning of PCR-generated LLP-promoter fragments into YJK_PVA_021 yeast strain. 11 randomly picked clones for each LLP-promoter length were picked and grown in deep well plates and the supernatant analyzed by SDS-PAGE followed by Coomassie blue staining. The lengths of the tested LLP-promoter fragments were as follows: Fig. A: 576 bp, Fig. B: 512 bp, Fig. C: 472 bp, and Fig. D: 372 bp. Each figure shows two representative lines, with “Option 1” denoting that only scFv is expressed, and with “Option 2” denoting that scFv and LLP is expressed.

(90) FIG. 14: Sequences S. Cerevisae ssn6 and TUP1 (Complete Coding Region)

(91) This figure shows the respective complete coding region of the following:

(92) A. Ssn6 nucleic acid, of S. cerevisiae (SEQ ID NO: 135);

(93) B: SSN6 Protein, of S. cerevisiae (SEQ ID NO: 137);

(94) C: Tup-1 nucleic acid of S. cerevisiae (SEQ ID NO: 136);

(95) D: Tup-1 protein of S. cerevisiae (SEQ ID NO: 138).

METHODS

(96) 1. Assessing the amount of a candidate ssn6-like related gene

(97) The level of expression of a candidate ssn6-like related gene can be measured for example by measuring the level of mRNA of said ssn6-like related gene by northern blotting or by quantitative Polymerase Chain Reaction (qPCR) or reverse transcriptase qPCR, or measuring the activity of the promoter of said ssn6-like related gene for example by using luciferase reporter gene assays or by using ssn6-like related promoter-green fluorescent protein (GFP) constructs, etc. All these methods are well known to a person skilled in the art and represent routine work. A textbook comprising protocols for routine methods is for instance Sambrook et al., “Molecular Cloning: A Laboratory Manual”, 4.sup.th Edition, Cold Spring Harbor Laboratory Press, (2012), referred to herein as Sambrook et al.

(98) 2. Assessing whether a candidate ssn6-like related gene resembles the ssn6-like gene as defined herein with respect to function, activity and sequence

(99) 2.1 Measuring the amount of LLP protein expressed in cell culture:

(100) The amount of a protein (or its expression level) can be determined according to any suitable method that is known to a person skilled in the art, for instance by measuring the amount of LLP protein in the supernatant of a cell culture by ELISA, by western blotting, or SDS PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) analysis of LLP protein.

(101) In the present invention, the amount of LLP protein in the supernatant has been measured by carrying out a qualitative SDS PAGE analysis (Novex NuPage 4-12% Bis-Tris Gels with MES running buffer, from Invitrogen).

(102) Alternatively the level of LLP protein can be determined indirectly by measuring LLP-mRNA using the same methods as described above in paragraph 1 “Assessing the amount of a candidate ssn6-like related gene”.

(103) 2.2 Measuring the Amount of a Gene of Interest (G01) Expressed Under Control of the LLP-Promoter in Cell Culture

(104) The same methods for measuring the amount of LLP-protein, as described above under 2.1, can also be used for measuring GOI-protein expressed under control of a LLP-promoter according to the invention.

(105) 2.3 Measuring Activity and Function of a Protein

(106) In order to measure the function and activity of SSN6-like protein or SSN6-like related protein, any suitable protocol that is known to a skilled person can be used. Especially any method suitable to measure the functioning of the LLP-promotor can be use to measure the function and activity of SSN6-like protein or SSN6-like related protein, such as the methods mentioned under 2.1 and 2.2 above.

(107) Once an SSN6-like protein or a SSN6-like related protein has been identified, its LLP-promoter-suppressing activity can be inhibited by inactivating the corresponding ssn6-like gene or ssn6-like related gene by methods described elsewhere in this application. Once this blockage has been performed the functioning (meaning the reduced amount of, or the complete lack of) the corresponding SSN6-like protein or mRNA or SSN6-like related protein or mRNA can be measured. Suitable methods are described elsewhere in this application.

(108) PVA enzyme assay: an aliquot of the supernatant was mixed with the substrate Phenoxymethylpenicillin-Kalium and incubated at 22° C. The presence of active PVA enzyme will result in cleavage of the substrate in a titer-dependent manner. The reaction was stopped after 60 min by addition of ice-cold methanol. The amount of the cleavage product phenoxy acetic acid was determined by quantitatve HPLC.

(109) 3. Comparison of Different Expression Systems

(110) In order to compare the protein expression of different expression systems, any suitable method that is known to a person skilled in the art can be carried out. In the present invention, the following protocol has been applied for this purpose:

(111) The supernatants obtained from the pLLP expression system and from the pGAP system (same volume) were directly loaded on SDS-PAGE gels for relative comparisons of the pLLP vs. pGAP system.

(112) 4. Identification of Regulatory DNA Sequences

(113) There are various tools available to predict regulatory DNA sequences, reviewed for example by Wassermann et al., Nature Reviews Genetics 5, 276-287. There are also tools for the prediction the total length of a promotor sequence, as well as tools predicting distinct positions within such a promoter, which supposedly bind to certain transcription factors, etc. One of these online tools is available from the University of Copenhagen, Bioinformatics Center. We used the JASPAR development server, Version 5.0_ALPHA. Settings of this online tool were: JASPAR CORE fungi, all JASPAR matrix models were choosen (we searched for all transcription recognition sites listed for fungi in the JASPER CORE fungi database), profile score threshold was set to 95%. Our input sequence was the LLP-promoter sequence SEQ ID NO: 12. This anlaysis resulted in 63 predicted binding sites for transcription factors within the tested sequence as shown in the table below.

(114) More details on transcripton factor binding site prediction can be found in Nat Rev Genet. 2004:4, 276-87.

(115) TABLE-US-00002 Model predicted site  name Score Start End Strand sequence SKO1 14.890 274 281 1 ACGTAATG ABF1 13.893 204 219 −1 CCGTAAAAAGCGATAC (SEQ ID NO: 116) CST6 13.595 271 279 1 ATGACGTAA SUM1 11.962 346 354 1 ATAATTTTT SPT23 11.708 750 757 −1 AAAATCAA SPT23 11.708 516 523 1 AAAATCAA YOX1 11.696 638 645 −1 TTAATTAT RFX1 11.668 39 46 −1 CGTTGCTA STE12 11.624 917 923 −1 TGAAACG YHP1 11.563 499 504 1 TAATTG AFT2 11.464 69 76 −1 CACACCCT TEA1 11.362 265 272 1 GCGGACAT YML081W 11.251 72 80 −1 ACCCCACAC ARR1 11.154 684 691 −1 ATTTGAAT TOS8 11.143 380 387 1 GTGTCAAA MBP1::SWI6 10.880 717 723 1 TCGCGTT PHD1 10.721 104 113 −1 ACCTGCATCA  (SEQ ID NO: 117) YPR022C 10.671 73 79 −1 CCCCACA YGR067C 10.573 67 80 −1 ACCCCACACCCTAC (SEQ ID NO: 118) ACE2 10.293 333 339 1 CCCAGCA ADR1 10.260 74 80 −1 ACCCCAC MIG3 10.179 73 79 −1 CCCCACA STB5 10.040 835 842 −1 CGGTATTA MIG2 10.037 73 79 −1 CCCCACA MIG1 9.822 74 80 −1 ACCCCAC YAP5 9.428 704 709 −1 AAGCAT YAPS 9.428 698 703 1 AAGCAT YAP5 9.428 456 461 1 AAGCAT SIG1 9.387 222 226 −1 ATATA ARR1 9.314 106 113 −1 ACCTGCAT MOT3 9.052 296 301 1 AAGGTA YAP5 8.787 259 264 1 AAACAT HAP2 8.684 424 428 1 TTGGT HAP2 8.684 159 163 1 TTGGT SKN7 8.580 401 406 −1 GGCCAT HAP2 8.495 665 669 −1 TTGGC HAP2 8.495 622 626 1 TTGGC HAP2 8.495 404 408 −1 TTGGC PHO2 8.463 436 441 −1 ATAATA GLN3 8.378 570 574 1 GATAA GLN3 8.378 6 10 1 GATAA MBP1 8.257 716 722 −1 ACGCGAT PHO2 8.131 119 124 1 ATATTA PHO2 8.131 90 95 −1 ATATTA SKN7 8.026 403 408 1 GGCCAA ARG80 7.974 956 961 −1 TGACAC ARG80 7.974 380 385 −1 TGACAC YAP5 7.933 185 190 1 AGACAT FZF1 7.932 551 556 −1 CTATCA PHO2 7.806 488 493 1 TTATTA PHO2 7.806 435 440 1 TTATTA PHO2 7.733 638 643 1 ATAATT PHO2 7.733 346 351 1 ATAATT PHO2 7.733 51 56 1 ATAATT PHO2 7.408 640 645 −1 TTAATT PHO2 7.402 587 592 −1 ATATTT GLN3 7.272 552 556 1 GATAG GLN3 7.272 28 32 −1 GATAG GLN3 7.272 2 6 1 GATAG HAP2 7.252 592 596 −1 TTGGA HAP2 7.252 529 533 1 TTGGA HAP2 7.252 453 457 −1 TTGGA HAP2 7.252 196 200 −1 TTGGA

(116) 5. Transformation and Cultivation of Strains

(117) Expression constructs were linearized by digestion with suitable restriction enzyme and transformed in Pichia strains by electroporation. The transformants were subseuquently plated on agar plates containing Zeocin (final concentration: 100 mg/L) and/or Geneticin (final concentration: 300 mg/L).

(118) Single colonies or glycerol stocks were subjected to expression studies in 48-well plates using a starch/amylase based cell culture medium. OD (Optical Densitiy) at 600 nm at harvest was around 10,

(119) 6. Testing of Promoter Activity

(120) Promoter activity can be measured by any suitable method that is known to a person skilled in the art. An example of such a method is the use of qPCR or reporter gene assays (e.g. Luciferase, Green Fluorescent Protein (GFP) etc.), both of which are standard methods that are known to a person skilled in the art. For example the most suitable part of the LLP-promoter for high level expression of a GOI can be determined by successively shortened versions of the LLP-promoter sequence according to SEQ ID NO. 12, by inserting such shortened LLP-promoter sequences together with a Kozak sequence and a model protein sequence such as DLX521 (scFv), hGH, HSA, PVA, etc. and together with a signal sequence such as the MF-alpha pre-pro signal sequence with or without EAEA repeat, the natural signal sequence of said model protein, etc. into a pLLP vector carrying a resistance-marker such as Geneticin, Zeocin, etc. and transfecting such pLLP vetor into a suitable yeast cell such as for example YJK_PVA_021-cells, SSS1-cells, NRRL Y-11430-cells, etc. Individual clones or pooled clones of such transformed yeast cells then can be grown under standard growth conditions in deep well plated, shaker flasks, fermetors, etc. and the amount of expression of said model protein being measured using methods such as SDS-PAGE, ELISA, or protein-activity assays such as the PVA-assay described elsewhere in this application, etc. Shortened versions of the promoter could represent parts of the promoter sequence disclosed in SEQ ID NO: 12, for example a LLP-promoter having a length of 1000, 775, 675, 605, 576, 512, 472, 415, 404, 372, 305, 285, 235, 165, 100 nucleotides, etc. counted in each case from the 3′-end of SEQ ID NO: 12.

(121) 7. Assessing Degree of Identity of Nucleotide Sequences or Amino Acid Sequences

(122) “Sequence identity” or “% identity” refers to the percentage of residue matches between at least two polypeptide or polynucleotide sequences aligned using a standardized algorithm. Such an algorithm may insert, in a standardized and reproducible way, gaps in the sequences being compared in order to optimize alignment between two or more sequences, and therefore achieve a more meaningful comparison of the sequences. For purposes of the present invention, the sequence identity between two amino acid or nucleotide sequences is determined using the NCBI BLAST program version 2.2.29 (Jan. 06, 2014) (Altschul et al., Nucleic Acids Res. (1997) 25:3389-3402). Sequence identity of two amino acid sequences can be determined with blastp set at the following parameters: Matrix: BLOSUM62, Word Size: 3; Expect value: 10; Gap cost: Existence=11, Extension=1; Filter=low complexity activated; Filter String: L; Compositional adjustments: Conditional compositional score matrix adjustment. For purposes of the present invention, the sequence identity between two nucleotide sequences is determined using the NCBI BLAST program version 2.2.29 (Jan. 06, 2014) with blastn set at the following exemplary parameters: Word Size: 11; Expect value: 10; Gap costs: Existence=5, Extension=2; Filter=low complexity activated; Match/Mismatch Scores: 2,-3; Filter String: L; m.

(123) 8. Generation of the Super-Secretor Strain (SSS1)

(124) An expression cassette encoding the enzyme PVA (Penicillin V Amidase) under the control of the GAP promoter and another expression cassette coding for Zeocin was transformed into the P. pastoris strain NRRL Y-11430 (see table below). Clones secreting high levels of active PVA were screened by the following enzyme assay:

(125) The cleavage of the PVA substrate penicillin V to the products 6-APA (6-Aminopenicillanic acid) and phenoxyacetic acid was determined by quantifying the phenoxyacetic acid amount by HPLC,

(126) One clone showing high PVA titers was identified and named YJK_PVA_021. Surprisingly, this clone not only secreted high amounts of PVA but also the LLP protein at even higher levels. Thus, YJK_PVA_021 was further characterized by whole genome sequencing done by the company Illumina Inc., San Diego, Calif., USA, After de-novo assembly of the genome done by Illumina Inc. (see SEQ ID NOs: 67-115, showing the results of genomic sequencing), the localization of the sequence of the PVA expression construct was identified to be in SEQ ID NO: 103 and the adjacent genomic sequences were compared to the reference strain (P. Pastoris CBS 7435, gi|328351301|emb|FR839629.1|) for identification of the insertion site. A single copy of the expression cassette encoding PVA was found to be integrated randomly at position 807,480 of chromosome 1 of the reference strain Pichia pastoris CBS 7435. This position lies within the ssn6-like gene sequence resulting in disruption of the ssn6-like gene leading to a C-terminal truncated protein (see FIG. 2B, see SEQ ID NO: 9) and to a 3′-truncated coding region of the ssn6-like gene, respectively (for details see FIG. 1S, see SEQ ID NO: 23. No further obvious deviations from the reference sequence were found. Thus, this single random integration event resulted in high level secretion of LLP. The PVA expression cassette was removed from YJK_PVA_021 resulting in clone SSS1 (see also FIG. 3).

(127) The vector pGAPk was PCR-amplified (linearized by PCR) using oligo2395 (TCCTCGTCCAATCAGGTAG; SEQ ID NO: 119) and oligo2398 (AGTGGTACCTGCAGCTAAG; SEQ ID NO: 120) and the PCR-product was transformed into yeast strain YJK_PVA_021. Homologeous recombination replaced the pGAP-PVA expression vector containing the Zeocin-resistance marker with the empty vector sequence of pGAPk (empty expression cassette and Geneticin-resistance marker), Clones with Geneticin-resistance and without PVA activity were screened. PVA was measured as indicated above. One strain which does not express PVA but expresses high amounts of LLP was denoted SSS1 and used for subsequent expression studies of different GOIs.

(128) 9. Characterization of the LLP Signal Sequence

(129) The secreted LLP protein was N-terminally sequenced by Edman degradation to identify the LLP signal sequence cleavage site (Seq. ID NO. 3, FIG. 1C). In order to proof the general functionality of the LLP signal sequence, resulting in the secretion of heterologous proteins fused to the LLP signal sequence, the following experiments were performed:

(130) The LLP signal sequence was fused to the 3′ end of HSA coding sequence replacing the native HSA-signal sequence resulting in the sequence according to SEQ. ID NO: 121, (FIG. 1T) and cloned into the pLLP plasmid via EcoRV and Notl restriction enzyme sites. The HSA coding sequence was codon optimized (done by DNA2.0 Inc., Menelo Park, USA) and a silent point mutation was inserted into the LLP-signal sequence at position 45 exchanging G for a C, which does not change the amino acid sequence of the LLP-signal peptide, but which deletes the restriction enzyme site Pstl from the LLP-signal sequence. The resulting plasmid was transformed into SSS1 yest strain. 10 clones were randomly picked and subjected to cultivation in deep well plates. 15-20 μl/lane supernatant of deep well plate cultures were directly loaded on SDS-PAGE gels (Novex NuPage 4-12% Bis-Tris Gels from Invitrogen with MES-running buffer). FIG. 12A shows the results of this experiment: Lanes 3 and 7 represent clones bearing no HSA insert, lanes 2, 4, 8 and 10 represent clones according to option 1 (only HSA expressed), and lanes 1, 5, 6 and 9 represent clones according to option 2 (HSA expressed in parallel to LLP). FIG. 12B shows the result of a 1 liter fermenter expression using the same clone used in lane 2 of the gel in FIG. 12A. A sample from the fermenter supernatant was collected after 48, 72 and 96 hours and 15 μl of each sample subjected to SDS-PAGE, followed by Coomassie blue staining using the “SimplyBlue SafeStain” system from Life Technologies according to the manufacturer's instruction.

(131) 10. Characterization of the ssn6-like/LLP-Promoter Expression System

(132) The table below shows the basic characteristics of the expression constructs/vectors and the yeast cells used in order to evaluate the functioning of the expresson system for various classes of proteins/genes of interest (GOIs), namely hormones, antibodies, enzymes and structural proteins. The expression results of GOI depicted in the figures were generated with the expression contructs/vetors/host strain combinations listed in the table below. Comparison of the expression of a GOI using a standard GAP-promotor or the LLP-promoter was done using SDS-PAGE, ELISA and/or enzymatic activity assays, as described in the figures.

(133) TABLE-US-00003 GOI Codon inserted Plasmid Protein Protein usage Signal in vector (linearized Resistance Host name class for GOI sequence between with) marker strain Comments hGH Hormone codon S. cerevisiae EcoRV pLLP Zeocin SSS1 pLLP from optimized MF-alpha and Notl (AvrII) Sandoz for Pichia (without EcoRI pGAP Zeocin NRRL pGAP from pastoris EAEA and Notl (SwaI) Y-11430 DNA2.0 repeat) (pJ905) DLX521 Antibody codon S. cerevisiae EcoRV pLLP Zeocin SSS1 pLLP from (fragment optimized MF-alpha and Notl (AvrII) Sandoz scFv) for Pichia (without EcoRI pGAP Zeocin NRRL pGAP from pastoris EAEA and Notl (SwaI) Y-11430 DNA2.0 repeat) (pJ905) EcoRV pLLP Geneticin YJK_ pLLP from and Notl (AvrII) PVA_021 Sandoz PVA Enzyme codon S. cerevisiae EcoRV pLLP Zeocin SSS1 pLLP from optimized MF-alpha and Notl (AvrII) Sandoz for Pichia (with EAEA EcoRI pGAP Zeocin NRRL pGAP from pastoris repeat) and Notl (BgIII* Y-11430 Sandoz HSA Structural codon Human EcoRV pLLP Zeocin SSS1 pLLP from protein optimized HSA pre- and Notl (AvrII) Sandoz for Pichia pro signal EcoRI pGAP Zeocin NRRL pGAP from pastoris sequence and (SwaI) Y-11430 DNA2.0 Notl I (pJ905) Pichia EcoRV pLLP Zeocin SSS1 pLLP from pastoris and Notl (AvrII) Sandoz LLP-signal sequence**, *** *This plasmid was also linearized by PCR as described elsewhere in this application (see paragraph 8. “Generation of the super secretor strain (SSS1)”). **HSA coding sequence was independently codon optimized resulting in identical amino acid sequences but in slightly different nucleotide sequences as compared to the HSA-sequence one row above ***P. pastoris signal sequence contains a silent point mutation at position 45 changing G to C in order to delete a restriction enzyme site

(134) 11. Identification of ssn6-Like Consensus Sequences

(135) The ssn6-like sequence of Pichia pastoris strain NRRL Y-11430 was used for a BLAST-search of similar sequences. BLAST parameters were “automatically adjust paramters for short input sequences”, Expect threshold: 10, Word size: 3, Max matches in a querry range: 0, Matrix: BLOSUM62, Gap Costs: Existence: Extension:1, Compositional adjustments: “Conditional compositional score matrix adjustment. The top 39 sequences originate from the following organisms (also see FIG. 9A):

(136) Komagataella pastoris CBS 7435 (Synonym/other names: Pichia pastoris, Pichia pastoris CBS 7435), Komagataella pastoris GS115 (Synonym/other names: Pichia pastoris, Pichia pastoris GS115), Scheffersomyces stipitis CBS 6054 (Synonym/other names: Pichia stipitis, Pichia stipitis CBS 6054), Millerozyma farinosa CBS 7064 (other name: Pichia farinosa CBS 7064), Candida parapsilosis, Candida orthopsilosis Co 90-125, Debaryomyces hansenii CBS767, Spathaspora passalidarum NRRL Y-27907, Candida albicans, Candida albicans SC5314, Candida maltosa Xu316, Candida tropicalis MYA-3404 (other name: Candida tropicalis T1), Lodderomyces elongisporus NRRL YB-4239 (other name: Saccharomyces elongisporus), Clavispora lusitaniae ATCC 4272 (genebank anamorph: Candida lusitaniae ATCC 42720), Meyerozyma guilliermondii ATCC 6260 (genebank anamorph: Pichia guilliermondii ATCC 6260), Wickerhamomyces ciferrii, Ogataea parapolymorpha DL-1 (synonym and other names: Hansenula polymorpha, Hansenula polymorpha DL-1, Ogataea angusta DL-1, Ogataea parapolymorpha ATCC 26012, Ogataea parapolymorpha DL-1, Pichia angusta DL-1), Cyberlindnera fabianii (synonyms and other names: Hansenula fabianii, Pichia fabianii, . . . ), Kuraishia capsulata CBS 1993, Dictyostelium discoideum AX4 (belongs to social amoebae), Tetrapisispora phaffii CBS 4417 (synonym: Fabospora phaffii, Dictyostelium purpureum (belongs to social amoebae), Pseudozyma flocculosa PF-1, Malassezia globosa CBS 7966, Botryobasidium botryosum FD-172 SS1 (basidiomycete), Naumovozyma dairenensis CBS 421 (synonyme; Saccharomyces dairenensis), Tetrapisispora blattae CBS 6284, Mucor circinelloides f. circinelloides 1006PhL (Early diverging fungal lineage), Malassezia sympodialis ATCC 42132, Kazachstania naganishii CBS 8797 (Saccharomyces naganishii), Saccharomyces cerevisiae YJM789, Saccharomyces cerevisiae FostersB, Saccharomyces cerevisiae, Saccharomyces cerevisiae S288c, Ustilago hordei (Corn smut fungus, basidiomycete), Meyerozyma guilliermondii ATCC 6260 (synonym/other names: Candida guilliermondii, Pichia guilliermondii ATCC 6260), Ustilago maydis 521, (Corn smut fungus, basidiomycete).

(137) The top 39 identified amino acid sequences were aligned to the ssn6-like-amino acid sequence (=identical to CCA36593.1) revealing significant similarities (see FIGS. 9B and 9C). The alignment was done using the online tool Clustal Omega provided by the EMBL-EBI which is described by Sievers et al., Molecular Systems Biology 7, article number: 539, Valentin et al., Nucleic acids research, 2010, 38, Suppl, W695-9, and by McWilliam et al., Nucleic acids research, 2013, July; 41 (Web Server issue): W597-600. Clustal Omega uses the HHalign algorithm and its default settings as its core alignment engine. The algorithm is described in Söding, J. (2005) ‘Protein homology detection by HMM-HMM comparison’. Bioinformatics 21, 951-960. The default transition matrix is Gonnet, gap opening penalty is 6 bits, gap extension is 1 bit. The symbols used for the consensus sequence at the bottom of the alignment are as follows. An “*” (asterisk) indicates amino acid positions which have a single, fully conserved residue across all 40 aligned sequences. A “:” (colon) indicates conservation between groups of amino acids of strongly similar properties—scoring>0.5 in the Gonnet PAM 250 matrix. A “.” (period) indicates conservation between groups of amino acids of weakly similar properties—scoring≥0.5 in the Gonnet PAM 250 matrix. Sequence CCA36593.1 corresponds to the Pichia pastoris SSN6 sequence identified in this application. FIG. 9A shows only that part of the alignment with the highest similarity between all 40 sequences. FIG. 9B shows the consensus sequence corresponding to Pichia pastoris SSN6-like (CCA36593.1) amino acids 352 to 372. FIG. 9C shows the consensus sequence corresponding to Pichia pastoris SSN6-like (CCA36593.1) amino acids 394 to 417.

(138) Amino acids which are identical in all 40 sequences are written in white with black background-labelling. The consensus sequence shows either individual amino acids which are identical in all 40 sequences or shows a group of amino acids in brackets “( )” and each of the amino acids in such a set of brackets can be chosen alternatively for that position in the consensus sequence. A “-” (dash) indicates that the amino acid at this position may also be omitted from the consensus sequence, which is for example one possibility for positions 406 and 407 of the consensus sequence depicted in FIG. 9C. For example the starting amino acid at position 352 of the consensus sequence in FIG. 9B is “W”, meaning that the consensus sequence at this position contains a Tryptophan, position 353 of the consensus sequence in FIG. 9B is written as “(CGL)” meaning that at this position of the consensus sequence can either be located Cysteine, Glycine or Leucine, position 354 is labelled “(SLTA)” meaning that at this position there can either be located Serine, Leucine, Threonine or Alanine, etc. The same nomenclature is used for the second consensus sequence depicted in FIG. 9C. Consensus sequences can be shorter or longer as the two exemplary consensus sequences shown in FIGS. 9B and 9C. Consensus sequences can be deduced from the sequence alignment of FIG. 9A or can be deduced from other parts of the sequence alignment prepared from the 40 above mentioned sequences using the sequence alignment method as described above. Preferably the consensus sequence contains at least 24 amino acids, preferably 23, 22, 21, 20, 19, 18, 17 amino acids, more preferably at least 16, 15, 14, 13, 12, 11, or 10 amino acids, most preferably at least 9, 8, 7, 6, 5 or 4 amino acids. Preferably a SSN6-like protein contains at least one or two consensus sequences, more preferably both consensus sequences shown in FIG. 9B and FIG. 9C, more preferably at least one consensus sequence, most preferably a consensus sequence selected from the sequences shown in FIGS. 9B and 9C.

(139) 12. Characterizatin of the Functionality of the LLP-Promoter

(140) The plasmid pLLPk containing DLX521 with MFalpha signal sequence was used as PCR-template in combination with the following PCR-primers to generate shortened/truncated versions of the LLP-promoter (Δ-fragment).

(141) Used reverse primer:

(142) TABLE-US-00004 SEQ Primer Sequence  ID name (5′ to 3′ end) Δ-fragment NO.: 2892 TGTCGAACCACCAC used for all  122 TAC fragments see FIG.  13 A to FIG. 13 D

(143) Used forward primer:

(144) TABLE-US-00005 Length of SEQ Primer Sequence (5′ to 3′ Δ- Promoter ID name end) fragment fragment NO.: Yo_218 TATACCTAGGTGGTGGAACT Δ29 576 123 TTATTATTCTTTC Yo_219 TATACCTAGGTATTAGCTGG Δ93 512 124 TAATTGAGCG Yo_220 TATACCTAGGTTGGAGGGT Δ133 472 125 ATGGTCAGAG Yo_221 TATACCTAGGTTTCATTCCA Δ201 404 126 TCTTGCCATC Yo_222 TATACCTAGGCTTACATCAA Δ233 372 127 TAATTAAAAC Yo_223 TATACCTAGGGCAAGCATAT Δ300 305 128 GCTTAAAAGG

(145) The resulting PCR products were ligated in via Spel and Avrll restriction enzyme sites into the vector pLLPk_containing DLX521. The correct sequences of the resulting plasmids were confirmed by DNA sequencing. The plasmids were linearized with Avrll and transformed in strain YJK_PVA_021. 11 clones per plasmid were randomly picked and subjected to cultivations in deep well plates using a synthetic medium. At harvest, 15 -30 μl supernatant samples were direclty loaded on SDS-PAGE gels (Novex NuPage 4-12% Bis-Tris Gels from Invitrogen with MES-running buffer) to analyze expression of scFv by staining the gels (Simply Blue Safe Stain). The OD600, meaning the yeast cell numbers per ml culture medium at harvest, was comparable for all clones analyzed.

(146) M=Protein Marker VI.

(147) Results are shown in FIG. 13A to FIG. 13D.

(148) 13. Examples of ssn6-Like Mutants

(149) The yeast strain contains the modified ssn6-like gene coding for amino acids 1 to 367 and in addition seven amino acids (EWYLQLR; SEQ ID NO: 139) originating from the vector inserted into the ssn6-like gene in order to disrupt the ssn6-like gene. Alternatively, modified versions of ssn6-like might contain a modified ssn6-like gene coding for the following regions of amino acids of SSN6-like protein, to the effect that the modified SSN6-like protein is not able to exert its wildtype function and/or wildtype activity:

(150) Modified versions of SSN6-like protein comprising amino acids 1 to 44, 1 to 77, 1 to 100, 1 to 122, 1 to 155, 1 to 189, 1 to 235, 1 to 275, 1 to 315, 1 to 348, 1 to 400, 1 to 450, 1 to 500, 1 to 550, 1 to 600, 1 to 650, 1 to 367, 1 to 400, 1 to 450, 1 to 500, 1 to 550, 1 to 600, 1 to 650, and/or 1 to 700 of SSN6-like protein according to SEQ ID NO. 9.

(151) In another alternative version the modified versions of ssn6-like might contain the region of the ssn6-like gene coding for blocks of SSN6-like amino acids according to SEQ ID NO. 9, namely amino acids 1 to 44, 45 to 77, 78 to100, 101 to 122, 123 to 155, 156 to189, 190 to 235, 236 to 275, 276 to 315, 316 to 348, 348 to 367, 368 to 400, 401 to 450, 451 to 500, 501 to 550, 551 to 600, 601 to 650, 651 to 700, and/or 701 to 736. Each of these blocks of amino acids might be combined with one or more other blocks of amino acids, preferably in the same order as they occur in SEQ ID NO. 9, wherein none, one or more amino acid block(s) is/are lacking in between two amino acid blocks.