SE543945C2 - Synthetically evolved DNA constructs for regulating signal peptide performance as well as vectors and host cells thereof - Google Patents

Abstract

The present invention provides a simple and inexpensive system for regulating signal peptide performance by using a synthetically evolved nucleotide sequence. The invention further relates to an expression vector comprising the nucleotide sequence. Additionally, the present invention relates to host cell comprising the expression vector. Furthermore, the present invention relates to a recombinant protein expressed by the host cell as well as a method for expressing the recombinant protein.

Description

SYNTHETICALLY EVOLVED DNA CONSTRUCTS FOR REGULATING SIGNALPEPTIDE PERFORMANCE AS WELL AS VECTORS AND HOST CELLS THEREOF TECHNICAL FIELD The present invention relates to the general field of regulating signal peptide performance.More specifically, the present invention relates to regulating recombinant protein expression Via controlling signal peptide perforrnance.

BACKGROUND Bacterial cell factories are Widely used in the biotech and pharrnaceutical industries for theproduction of high-value recombinant proteins. Classic examples include industrial enzymes,horrnones and antibody fragments, Which generate billions of dollars in revenue annually [l ,2]. These recombinant proteins are typically engineered With an N-terrninal signal peptideso that they Will be secreted out of the bacterial cytoplasm [3]. For industrial enzymes, Whichare usually produced in gram-positive bacteria such as Bacillus subtilis, secretion from thecytoplasm to the culture supematant simplifies purification and downstream processing. Forhorrnones and antibody fragments, Which are usually produced in gram-negative bacteria likeEscherichia coli, secretion from the cytoplasm to the oxidizing environment of the periplasmis necessary for the formation of disulfide bonds that are essential for protein folding and activity [4,5].

Secretion out of the bacterial cytoplasm is usually mediated by the general secretion pore(Sec) [6,7]. Sec is a major hub for protein trafficking as it inserts proteins into the cytoplasmicmembrane, and secretes proteins to the envelope and beyond. Secreted proteins are typicallytargeted to Sec by an N-terrninal signal peptide. Signal peptides vary in length and amino acidsequence, but have a distinctive tripartite structure that includes a positively-charged N-terrninal region, a hydrophobic core, and a polar C-terrninal cleavage site that typicallycontains the signal peptidase recognition site (Ala-X-Ala) [8,9]. They also have a distinctivecodon usage, Which includes a biased use of the AAA (Lys) codon at the second position, anda high frequency of non-optimal codons [l0-l4]. It has been suggested that the signal peptidesloWs folding of the protein in the cytoplasm and targets it to Sec in a predominantly unfolded confirrnation [l5]. Upon arrival at Sec the signal peptide also promotes binding to the SecA chaperone, thereby allosterically activating Sec for protein secretion [16]. Given thesemultiple roles it is likely that signal peptides have co-evolved With the protein that they translocate, as Well as With the secretion machinery.

Signal peptides have unpredictable effects on the production yields of recombinant proteins.For example, a signal peptide that supports a high-level of protein synthesis and secretion forone recombinant protein often supports a low-level of protein synthesis and secretion foranother. Herein We refer to this phenomenon as signal peptide perforrnance (i.e. signal peptidestrength). Since it is not possible to predict how Well a signal peptide Will perforrn With agiven recombinant protein, it is common practice to screen large signal peptide-libraries forone that supports a high-level of protein synthesis and secretion [3]. This approach is bothtime-consuming and expensive. Hence, there is a need for a molecular understanding of signalpeptide perforrnance since as it could lead to new methods for (1) identifying suitable signalpeptides, and (2) rationally engineering signal peptides that increase production yields in bacterial cell factories.

Translation initiation is a rate-limiting step of protein synthesis in bacteria [l7-2l], Where the30S subunit of the ribosome, together With the initiation factors IFl and IF3 bind to theTranslation Initiation Region (TIR) of the mRNA. This pre-initiation complex then recruitsthe GTP bound initiation factor IF2 and the initiating forrnyl-methionine tRNAfMet. Onceassembled, GTP is hydrolyzed, the initiation factors are released and the 50S subunit isrecruited [22]. The efficiency of translation initiation is dependent on the nucleotide sequenceof the TIR, a stretch of approximately thirty nucleotides that extends from the Shine-Dalgamoregion to the fifth codon of the coding sequence (i.e. the first ribosomal footprint) [23]. TheTIR is the only variable element during translation. If all possible sequence perrnutations areconsidered, there are more than a quintillion TIRs (i.e. 430 >l x 1018). HoWever only a smallnumber of TIRs are present in bacterial cells and they contain some distinctive sequencefeatures. The most obvious is the Shine-Dalgamo (SD) sequence, a purine rich stretch of 4-9nucleotides that hydrogen bonds With the l6S rRNA of the 30S subunit 24. This sequenceguides the ribosome to the start codon, Which is typically an AUG [25]. The start codon isseparated from the SD sequence by a spacer region that is typically 9 nucleotides long in E.coli [26]. The 5” end of the coding sequence (~ 15 nucleotides) is also considered to be Withinthe TIR [27] and often harbors rare codons [28,29]. Native bacterial TIRs have co-evolved With the ribosome and are less likely to form mRNA structures compared to the rest of the coding sequence [30,3 1]. This is thought to promote accessibility of the SOS subunit duringtranslation initiation [28,29,32,33].

DNA constructs relating to signal peptides are known from US836l744. However, the 42DNA constructs disclosed in US836l744 differ significantly from the DNA constructs of thepresent invention both With respect to DNA sequence as Well as the perforrnance of the signalpeptides. Moreover, the DNA constructs of US836l744 have not been synthetically evolvedand therefore do not exhibit the technical effects of the DNA constructs disclosed in the present invention.

OBJECT OF INVENTIONThe object of the invention is to controlling signal peptide perforrnance.

A further object of the invention is to control recombinant protein expression via controlling signal peptide perforrnance.

A further object of the invention is to increase (i.e. up-regulate) or to decrease (i.e. down- regulate) signal peptide perforrnance.

A further object of the invention is to provide a simple and inexpensive system of DNAconstructs, expression vectors and host cells for increasing the production yields of single chain antibody fragments, horrnones and other recombinant proteins.

SUMMARY OF THE INVENTION In the present invention, the inventors have solved the problem and anomaly in recombinantexpression plasmid typically used to produce secreted proteins. It is in the art a commonpractice to place the coding sequence of the signal peptide doWnstream of the vector encoded5 °UTR. Hence, the resulting TIR is a fusion of the 5 ”UTR and the first 5 codons of the signalpeptide in the TIR. The inventors hypothesized that such a TIR Would not function optimallyas it had not co-evolved With the ribosome. To test this hypothesis, as described in detail inthe DETAILED DESCRIPTION of the present specification, the inventors syntheticallyevolved the TIR in the presence of host cell ribosomes. The experimental results discussed inthe EXAMPLES section of the present specif1cation clearly indicate that the perforrnance of all signal peptides can be improved by synthetic evolution. The most striking example Was PelBSP, Which Was initially the Worst performing signal peptide for production of ß-lactamase,but the best perforrning following synthetic evolution of the TIR. Thus, in summary, theperforrnance of the signal peptide is largely coupled to the efficiency of translation initiation.The present invention provides a molecular understanding of this signal peptide perforrnance.More importantly, the present invention provides a simple and inexpensive systemcomprising: - DNA constructs, - expression vectors, - host cells, and - methods of production,for increasing the production yields of single chain antibody fragments, horrnones and other recombinant proteins.

The objects of the invention are attained by the subj ect-matter disclosed in the claims as Well as the subj ect-matter disclosed in the below aspects of the invention.

A first aspect of the invention relates to a DNA construct suitable for controlling signalpeptide performance, Wherein said DNA construct comprises a sequence of one of SEQ ID -28.

In a preferred embodiment, said DNA construct also comprises a signal peptide encoding SCQUCIICC.

In a preferred embodiment, said signal peptide encoding sequence comprises a sequence forexpressing a signal peptide selected from MalE (maltose-binding protein precursor), OmpA(outer membrane protein A precursor), PhoA (alkaline phosphatase precursor), DsbA (thiol:disulf1de interchange protein), and PelB (periplasmic pectate lyase).

In a preferred embodiment, said sequence of one of SEQ ID 15-28 comprises the first 24 nucleotides of said signal peptide encoding sequence.

In a preferred embodiment, said signal peptide encoding sequence is a sequence of one of SEQ 1D 34-47.

In a preferred enibodinient, said signal peptide encoding sequence expresses a signal peptide of a sequence of one of SEQ ID 29-33.

In a preferred enibodinient, said DNA construct is characterized in that: a sequence of one of SEQ ID 15, 16 and 17 coniprises the first 24 nucleotides of aMalE signal peptide encoding sequence of one of SEQ ID 34, 35 and 36, respectively;a sequence of one SEQ ID 18, 19 and 20 coniprises the first 24 nucleotides of anOn1pA signal peptide encoding sequence of one of SEQ ID 37, 38 and 39,respectively; a sequence of one SEQ ID 21 and 22 coniprises the first 24 nucleotides of a PhoAsignal peptide encoding sequence of one of SEQ ID 40 and 41, respectively; a sequence of one SEQ ID 23, 24 and 25 coniprises the first 24 nucleotides of a DsbAsignal peptide encoding sequence of one of SEQ ID 42, 43 and 44, respectively;and/or a sequence of one SEQ ID 26, 27 and 28 coniprises the first 24 nucleotides of a PelBsignal peptide encoding sequence of one of SEQ ID 45, 46 and 47, respectively.

In a preferred enibodinient, said DNA construct is characterized in that: said MalE signal peptide encoding sequence of one of SEQ ID 34, 35 and 36expresses a signal peptide of a sequence of one of SEQ ID 29; said On1pA signal peptide encoding sequence of one of SEQ ID 37, 38 and 39expresses a signal peptide of a sequence of one of SEQ ID 30; said PhoA signal peptide encoding sequence of one of SEQ ID 40 and 41 expresses asignal peptide of a sequence of one of SEQ ID 31; said DsbA signal peptide encoding sequence of one of SEQ ID 42, 43 and 44expresses a signal peptide of a sequence of one of SEQ ID 32; and/or said PelB signal peptide encoding sequence of one of SEQ ID 45, 46 and 47 expressesa signal peptide of a sequence of one of SEQ ID 33.

In a preferred enibodinient, said DNA construct coniprises a sequence of one of SEQ ID 15, 18, 21, 23 and 26.

In a preferred enibodinient, said DNA construct is a synthetically evolved DNA construct.

In a preferred enibodinient, said DNA construct further coniprises a reconibinant protein encoding sequence.

A second aspect of the invention relates to an expression Vector comprising a DNA constructaccording to the above disclosed first aspect of the invention, wherein the expression vector is preferably a plasmid, more preferably PET expression vector, and most preferably pet28A A third aspect of the invention relates to a host cell comprising the above disclosed expressionvector of the second aspect of the invention, wherein said host cell is preferably a bacterialcell, more preferably said bacterial cell is E. coli and most preferably E. coli strain BL2 l (DE 3 ) pLysS.

A fourth aspect of the invention relates to a recombinant protein expressed by the above disclosed host cell of the third aspect of the invention.

A f1fth aspect of the invention relates to a method of expressing the above disclosedrecombinant protein of the fourth aspect of the invention, said method comprising the stepsof: - introducing said DNA construct according to the above disclosed first aspect of the invention into an expression vector;- introducing the expression into a host cell;- growing the host cell; and - and recovering the recombinant protein from the host cell.

A sixth aspect of the invention relates to an RNA molecule expressed by a DNA construct according to the above disclosed first aspect of the invention.

BRIEF DESCRIPTION OF THE FIGURES Figure l. A comparison of commonly used signal peptides. (A) An overview of the expressioncassettes used in this experiment. The TIR region represented by the boxed area, from theShine-Dalgamo to the f1fth codon of the signal sequence. The coding sequences for fivecommonly used signal peptides (MalESP, OmpASP, PhoASP, DsbASP, PelBSP) were cloned intothe pET28a vector, upstream of the mature coding sequences for ß-lactamase, scFvHERz orFtYfgMáß-lw. Protein production was induced for two hours, then a volume of cellscorresponding to 0.2 OD600 units of cells were harvested, separated by a 12 % SDS-PAGE and protein levels deterrnined by immuno-blotting with antisera to ß-lactamase (B), or the poly- Histidine ta of scFvHERz C and FtYf M45470 D . To ensure that rotein loadin wasg g p g consistent between the samples, the membranes were stained with Amido black after immuno-detection. “Pre” denotes the precursor form of the protein, which contains the signal sequenceand is presumed to be in the cytoplasm. “Mat° precursor denotes the mature forrn, which is presumed to be in the periplasm as the signal peptide has been cleaved.

Figure 2. Improved signal peptide performance following synthetic evolution of the TIR (A) A synthetic evolution approach was used to convert a TIRUNEVOLVED to a TIRSYNJEVOLVED. TheTIR is defined as the region from the Shine-Dalgamo (half-dome) to codon 5 of the signal RUNEVOLVED Can be peptide. (B) mRNA has a high propensity to form structures, thus a TIsequestered into short- (top) or long-range structures (middle). Synthetic evolution shouldselect a TIRSYNJEVOLVED that is relaxed and more accessible to the ribosome (bottom). (C) Anoverview of the synthetic evolution process. A TIRLIBRARY was constructed by completelyrandomising the six nucleotides immediately upstream of the AUG start codon, and partiallyrandomising the six nucleotides immediately downstream of the AUG start codon (allowingsynonymous codons changes only). The TIRLIBRARY was transforrned into E. coli BL2l(DE3) RsYNgavoLvEo pLysS and plated on increasing concentrations of ampicillin. A TI was identified on the plate containing the highest concentration of ampicillin relative to the TIRUNEVOLVED variant. (D) ß-lactamase production levels from TIRUNEVOLVED / TIRSYNJEVOLVED pairs wereassessed by immuno-blotting. In this experiment, ß-lactamase production was induced for twohours, then a volume of cells corresponding to 0.2 OD600 units of cells were harvested, separatedby a 12 % SDS-PAGE and protein levels were deterrnined by immuno-blotting with antisera toß-lactamase. To ensure that protein loading was consistent between the samples, the membranewas stained with Amido black after immuno-detection. “Pre° denotes the precursor form of theprotein, which contains the signal peptide fused version of ß-lactamase, which we presume tobe in the cytoplasm as the signal peptide is still present. “Mat° precursor denotes the matureversion of ß-lactamase, which presumably is in the periplasm as the signal peptide has been cleaved. (E) ß-lactamase activity from TIRUNEVOLVED / TIRSYNJEVOLVED pairs was assessedusing the disc diffusion assay. Here a filter disc containing 2 mg of ampicillin was placed ontop of an LB-agar plate containing a lawn of bacteria expressing ß-lactamase from either a TIRUNEVOLVED or a TIRSYNJEVOLVED. The diameter of the growth-inhibition zone was measured a TIRSYNJÉVOLVED for each experiment. In all cases, conferred more resistant to ampicillin than TIRUNEVOLVED _ Figure 3. Time-course analysis of ß-lactamase production. (A) An illustration of the experimental workflow used. (B) At each time point, a volume of cells was extracted, then separated by SDS-PAGE and immuno-blotted with antisera to ß-lactamase. Band intensitieswere obtained from immuno-blots by densitometric analysis and norrnalised to the highest- Value.

Figure 4. A synthetically evolved TIR (TIRSYNJEVOLVED) is transferable. (A) Expression levelsof ScFvHERz and F tYfgM45'170 using five different signal peptides. In each instanceTIRUNEVOLVED / TIRSYNJEVOLVED pairs were assessed by immuno-blotting. The TIRSYNJEVOLVEDhad originally been selected for ß-lactamase (see Figure 2). In this experiment, proteinproduction was induced for two hours, then a volume of cells corresponding to 0.2 ODeoo unitsof cells were harvested, separated by a 12 % SDS-PAGE and protein levels were deterrnined by immunoblotting with antisera to a poly-histidine tag. “Pre° denotes the precursor forrn of the protein and “Mat” denotes the mature version.

Figure 5. Production and purif1cation of the human growth hormone (hGH) using aTIRSYNJEVOLVED. (A) Production levels of hGH using five different signal peptides. In eachinstance the difference between TIRUNEVOLVED / TIRSYNJEVOLVED pairs were assessed byimmuno-blotting. The TIRSYNJEVOLVED had originally been selected for ß-lactamase (see Figure2). In this experiment, protein production was induced for two hours, then a volume of cellscorresponding to 0.2 OD600 units of cells were harvested, separated by a 12 % SDS-PAGE andprotein levels were deterrnined by immuno-blotting with antisera to a poly-histidine tag. “Pre°denotes the precursor forrn of the protein and “Mat° denotes the mature version. (B) Anoverview of the methodology used to purify hGH. (C) Analysis of the purified hGH by Size-Exclusion Chromatography (SEC). (D) Purified hGH was analysed by SDS-PAGE underdenaturing- and non-denaturing conditions. (E) Activity of the purif1ed hGH by using the MTS cell proliferation assay.

DETAILED DESCRIPTION The present invention relates to controlling signal peptide performance With a DNA constructWherein the DNA construct comprises: a. a Shine-Dalgamo sequence, b. an ATG start codon, c. a sequence of one of SEQ ID 1-28 comprising said ATG start codon, and d. a signal peptide encoding sequence,Wherein the sequence of one of SEQ ID 1-28 comprises at least the first 9 nucleotides of thesignal peptide encoding sequence. When the DNA construct comprise a sequence of one ofSEQ ID 15-28 then such a sequence comprises the Shine-Dalgamo sequence, said sequenceof one of SEQ ID 1-14 and at least the first 24 nucleotides of the signal peptide encoding SCQUCIICC.

The signal peptide encoding sequence may comprises a sequence for expressing a signalpeptide selected from MalE, OmpA, PhoA, DsbA and PelB. A specific signal peptideencoding sequence may be a sequence of one of SEQ ID 34-47 Which may express a signal peptide of a sequence of one of SEQ ID 29-33 indicated in Table 1.

However, DNA constructs comprising a sequence of one of SEQ ID 48-57 may also be used.

The preferred combinations of (a) a DNA construct sequence, (b) a signal peptide peptidesequence, and/or (c) a signal peptide DNA sequence, are disclosed in Tables 1 and 2.

Table 1. Signal peptides and their corresponding peptide and DNA sequences. Several peptideDNA sequences may express the same peptide sequence due to the synonymous codon changes discussed in Example 2, Figure 2 and elsewhere in the specification.

Signal Peptide Sequence Seq ID DNA sequence Seq IDpeptideMaIESP MKIKTGARILALSALTTMMFSA 29 ATGAAAATAAAAACAGG 58SALA TGCACGCATCCTCGCATTATCCGCATTAACGACGATGATGTTTTCCGCCTCGGCTCTCGCCCAC ATGAAAATTAAAACAGGT 34GCACGCATCCTCGCATTA TCCGCATTAACGACGATGATGTTTTCCGCCTCGGCTCTCGCCCAC ATGAAGATCAAAACAGGTGCACGCATCCTCGCATTATCCGCATTAACGACGATGATGTTTTCCGCCTCGGCTCTCGCCCAC ATGAAAATAAAAACAGGTGCACGCATCCTCGCATTATCCGCATTAACGACGATGATGTTTTCCGCCTCGGCTCTCGCCCAC OmpASP MKKTADUAVALAGFATVAQA ATGAAAAAGACAGCTATCGCGATTGCAGTGGCACTGGCTGGTTTCGCTACCGTAGCGCAGGCCCAC ATGAAGAAGACAGCTATCGCGATTGCAGTGGCACTGGCTGGTTTCGCTACCGTAGCGCAGGCCCAC ATGAAGAAGACAGCTATCGCGATTGCAGTGGCACTGGCTGGTTTCGCTACCGTAGCGCAGGCCCAC ATGAAAAAGACAGCTATCGCGATTGCAGTGGCACTGGCTGGTTTCGCTACCGTAGCGCAGGCCCAC PhoAsP MKQSTIALALLPLLFTPVTKA ATGAAACAAAGCACTATTGCACTGGCACTCTTACCGTTACTGTTTACCCCTGTGACAAAAGCCCAC ATGAAGCAAAGCACTATTGCACTGGCACTCTTACCGTTACTGTTTACCCCTGTGACAAAAGCCCAC ATGAAGCAAAGCACTATTGCACTGGCACTCTTACCGTTACTGTTTACCCCTGTGACAAAAGCCCAC DsbASP TWKKDNLALAGLVLAFSASA ATGAAAAAGATTTGGCTGGCGCTGGCTGGTTTAGTT TTAGCGTTTAGCGCATCGGCGCAC ATGAAAAAGATTTGGCTG 42GCGCTGGCTGGTTTAGTTTTAGCGTTTAGCGCATCGGCGCAC ATGAAGAAAATTTGGCTG 43GCGCTGGCTGGTTTAGTTTTAGCGTTTAGCGCATCGGCGCAC ATGAAAAAGATTTGGCTG 44GCGCTGGCTGGTTTAGTTTTAGCGTTTAGCGCATCGGCGCAC PdB¶ ]MKYLLPTAAAGLLLLAAQPAhl 33 ATGAAATACCTGCTGCCG> 62A ACCGCTGCTGCTGGTCTGCTGCTCCTCGCTGCCCAG CCGGCGATGGCCCAC ATGAAGTATCTGCTGCCG' 45ACCGCTGCTGCTGGTCTGCTGCTCCTCGCTGCCCAGCCGGCGATGGCCCAC ATGAAATATCTGCTGCCG> 46ACCGCTGCTGCTGGTCTGCTGCTCCTCGCTGCCCAGCCGGCGATGGCCCAC ATGAAATATCTGCTGCCG> 47ACCGCTGCTGCTGGTCTGCTGCTCCTCGCTGCCCAGCCGGCGATGGCCCAC The invention further relates to an expression Vector comprising the above-mentioned DNAconstruct. Additionally, the present invention relates to host cell comprising said expressionvector. Furthermore, the present invention relates to a recombinant protein expressed by saidhost cell as Well as a method for expressing said recombinant protein. The DNA construct may further comprise a recombinant protein encoding sequence.

The above described DNA constructs, expression vectors, host cell and recombinant proteinshave been described in the EXAMPLES and EXPERIMENTAL PROCEDURES sections ofthis specification. Moreover, the results of the comparative tests are discussed in the EXAMPLES section to provide evidence of the increased (i.e. up-regulated) signal peptideperformance of DNA constructs comprising a sequence of one of SEQ ID l-28. HoWever, thepresent invention may alternatively be used for decreasing the signal peptide perforrnance ofDNA constructs comprising a sequence of one of SEQ ID 48-57; such an effect may be relevant in cases When the expression of recombinant protein needs to be down-regulated.

Some of the significant comparative tests discussed in the EXAMPLES are summarized in the following paragraphs before the EXAMPLES section.

As already indicated, the present invention relates to improving signal peptide perforrnanceby synthetically evolving the TIR. The present invention further provides a simple andinexpensive solution for increasing the production yields of secreted proteins in bacterial cellfactories. Moreover, the present invention Will be compatible With other published methods;such as those that use titratable promoters to tune transcription rates of secreted proteins [40].A potential problem is the need for screening of large TIRLIBRARIES. HoWever, in the presentinvention, said problem Was solved by using ß-lactamase protein, Which confers resistance toß-lactam antibiotics and can be easily screened; this embodiment of the present invention isdiscussed in detail in Examples 2 and 3. For proteins Where no simple screening assay isavailable it is possible to translationally-couple ß-lactamase to the recombinant protein andthereby solve potential problems. It is also possible to use the signal peptides in pET28a vectors from the present invention, Which possess a TIRSYNJEVOLVED and Which improvedproduction yields of a single chain antibody fragment, a hormone and another recombinant protein in Escheríchía coli; this embodiment of the invention is discussed in detail in Example A link between signal peptide perforrnance and the efficiency of translation initiation hasbeen implied previously. Punginelli and co-Workers noted that non-synonymous nucleotidechanges in the signal peptide of the Tat-dependent forrnate dehydrogenase increasedproduction levels by up to 60-fold in E. coli [38] And Ng and Sarkar noted that synonymouschanges to the Usp45sp signal peptide in Lactococcus lactís helped to increase productionlevels of a nuclease and an amylase by approximately 15% [39]. Both studies postulated thatthe nucleotide changes helped to relax mRNA structure that had sequestered the TIR.

The present invention also demonstrates that nucleotide changes in the TIR can influenceproduction of secreted proteins (although this could not be correlated to changes in mRNA structure). Significantly, the present invention goes beyond the current literature as indicated in the comparative experiments described in Example 2 and Figure 2 which demonstrate thatsignal peptides generally under-perforrn in protein production experiments because the TIR,encompassing the 5” UTR of the plasmid and the 5” terrninus of the gene coding sequence, has not co-evolved with the ribosomes of the host cell.

As further disclosed in Example 2 and Figure 2, the inventors were able to support thismolecular explanation by demonstrating that a synthetic evolution process could improve theperforrnance of all commonly used signal peptides. As indicated earlier in the specification,the most striking example was PelBSP, which was initially the worst performing signal peptidefor production of ß-lactamase, but the best performing following synthetic evolution of theTIR as illustrated in Figure 2D. Thus, the perforrnance of the signal peptide is largely coupled to the efficiency of translation initiation.

EXAMPLES The following examples are not to be interpreted as limiting the scope of the invention. Forexperimental details pertaining to the examples below, the skilled reader is directed to the separate EXPERIMENTAL PROCEDURES section below.

Example 1 - Production of periplasmic proteins with commonly used signal peptides Five signal peptides that are commonly used for the production of recombinant proteins in theperiplasm of E. coli were selected (MalESP, OmpASP, PhoASP, DsbASP and PelBSP; see Table1; see SEQ-ID 29-33). The coding sequences for these signal peptides were cloned into thecommonly used pET28a expression plasmid, upstream of the coding sequence for ß-lactamase(Figure 1A). To determine how efficiently the signal peptides supported the synthesis andsecretion of ß-lactamase the expression plasmids were transforrned into the E. coli strainBL2l (DE 3 ) pLysS and a mild induction protocol was used to initiate transcription (0.05 mMIPTG for 2 hours at 30 °C). Following the induction period, whole cells were collected, andproteins were separated by SDS-PAGE and immuno-blotted, so that the secreted (Mature) andnon-secreted (Precursor) ß-lactamase could be distinguished. The experiment indicated thatthere were large differences in production levels (Figure IB). MalESP, OmpASP, PhoASP,DsbASP supported a comparatively high-level of ß-lactamase production, whereas PelBSP did not. The experiment also indicated that there were significant differences in secretion efficiency between the different signal peptides. MalESP and PelBSP were effective insupporting the secretion of ß-lactamase to the periplasm, whereas OmpASP, PhoASP and DsbASP were deemed less effective as there was a prominent precursor band.

To evaluate the perforrnance of the signal peptides with other recombinant proteins, they werefused to a single chain variable fragment that reco gnizes the human epiderrnal growth factorreceptor protein 2 protein (scFvHERz) and a soluble fragment of the periplasmic chaperoneYfgM from Francísella tularensís (F tYfgM45'17°). Again, there were considerable differencesin production levels across the different signal peptides (Figure lC and D). Moreover, therewere considerable differences between ß-lactamase, scFvHERz and F tYfgM45'17° (Figure lB vsC vs D). Taken together, these observations demonstrate that signal peptide perforrnance isvaried and unpredictable during the synthesis and secretion of recombinant periplasmicproteins. This conclusion is supported by a large body of published work, but a molecular explanation for the phenomenon remains elusive [3].

Example 2 - Signal peptide perforrnance is coupled to translation initiation The expression plasmids used in the previous experiments had been assembled by geneticallysandwiching the nucleotide sequence encoding the signal peptide between the vector encoded5°UTR and the 5” end of the mature coding sequence for ß-lactamase, scFvHERz or F tYfgM45-170 (Figure 2A). Each expression plasmid therefore contained a different TIR (Table 2). Theinventors hypothesized that these TIRs might not be optimal for translation initiation as theyhad not co-evolved with the host cell ribosomes, possibly leading to unfavorable interactions at the mRNA level (Figure 2B). They are therefore referred to as a TIRUNEVOLVED.

Synthetic (or directed) evolution was used to select TIRs that were more compatible with thehost cell ribosomes. In the experiment, TIRLIBRARIEScontaining the MalESP, OmpASP, PhoASP, DsbASP and PelBSP fused to ß-lactamase. In the design of the TIRLIBRARIES, the six nucleotides immediately upstream from the AUG start were created from expression plasmids codon were completely randomized, and the six nucleotides immediately downstream fromthe AUG start codon were randomized with synonymous codon changes only (Figure 2C)[34,35]. Each TIRLIBRARY theoretically contained >l 8,000 expression plasmids with adifferent TIR. The TIRLIBRARIES were transforrned into BL2l(DE3) pLysS and plated onto LBagar containing 0.05 mM IPTG and increasing concentrations of ampicillin (Figure 2C). A colony that was resistant to a high concentration of ampicillin was selected, the expression plasniid Was isolated and the TIR sequenced. These TIRs as referred to as syntheticallyeVo1ved(TIRSYN~EVOLVED) (Table 2).

Table 2. Nucleotide sequences of the TIRUNEVOLVED and corresponding TIRSYNJEVOLVED used in this study. The TIR is defined as the region from the Shine-Dalgarno to codon fiVe of the signal peptide.

Signal TIR Sequence Seqpeptide IDMaIESP UNEVOLVED gTTTAACcïšâiqågcišzxkëëäAGATATACCGATGAAAA 48/49SYBLEVOLVED ííTAACTgšâàgä%åcGAT 3/17OmpASP UNEVOLVED g 50/51 PhoASP UNEVOLVED liïïïëåciïêgïfïiäëêiïêçrgAGATATACCGATGAAAC 52/53DsbASP UNEVOLVED ÉTÅTFÉÉÉÉTFÉÉEÉÉÉÉÉÉAGATATACCGATGAAAA 54/55 PelBSP UNEVOLVED ïëïlšiëiêgäëëlëäëiëiêgfšAGATATACCGATGAAAT 56/57 1Underlined region Was randomised during the synthetic evolution process zNucleotides marked in bold text Were changed in TIRSYNJEVOLVED 3SEQ ID is indicated for underlined region (referred to as “short sequence” in the sequence listing) andfull nucleotide sequence (referred to as “full sequence” in the sequence listing), respectively Expression plasmids containing either a TIRUNEVOLVED or TIRSYNJEVOLVED Were re-transforrned into BL21(DE3) pLysS and the production levels of ß-lactamase compared byimmuno-blotting. After a two-hour induction period We observed that the production levels ofperiplasmic ß-lactamase Were significantly higher When using a TIRSYNJEVOLVED compared tothe TIRUNEVOLVED (Figure 2D). Note that production of ß-lactamase from each TIRUNEVOLVEDWas undetectable on these blots because the difference With the TIRSYNJEVOLVED Was too largeto capture at this time point (see below). Consistent With this observation, disc diffusion assays confirmed that the TIRSYNJEVOLVED supported a higher level of resistance to ampicillinthan the TIRUNEVOLVED (Figure 2E). Figures 2E and 2D illustrate comparative experimentsusing TIRUNEVOLVED and TIRSYNJEVOLVED Wherein (i) TIRSYNJEVOLVED comprised SEQ ID 15,18, 21, 23 and 26, and (ii) TIRUNEVOLVED comprised SEQ ID of48, 50, 52, 54 and 56, respectively.

The inventors speculate that the difference in production levels from the TIRUNEVOLVED/TIRSYNJEVOLVED pairs Was a result of mRNA relaxation, but the inventors Were unable tosupport this speculation by using mRNA fold prediction programs. The lack of a correlationcould reflect the fact that (1) mRNA relaxation is not the sole deterrninant, (2) mRNAstructure is notoriously difficult to predict, and/or (2) existing algorithms only handle shortstretches of nucleotides (not an entire mRNA). Nevertheless, the experiment doesdemonstrate that all signal peptides Were under-perforrning When a TIRUNEVOLVED Was used.And significantly, the performance of all signal peptides could be improved by synthetically evolving the TIRUNEVOLVED. This phenomenon Was most easily seen With PeIBSP, Which gave the lowest levels of ß-lactamase production When expressed from a TIRUNEVOLVED (Figure 1B) but the highest when expressed from a TIRSYNJEVOLVED (Figure 2E). Synthetic evolution of theTIR had therefore “converted° the PelBSP from a “poor-performing” signal peptide to a “top-performing” signal peptide without changing a single amino acid. The data thereforedemonstrate that signal peptide perforrnance is tightly coupled to translation initiation in bacterial cell factories.

Example 3 - Production of recombinant periplasmic proteins using a TIRSYNJEVOLVED In the previous series of experiments a mild induction protocol had been used (0.05 mM IPTGfor 2 hours at 30 °C), so that differences in protein production could be assessed in theabsence of a metabolic load on the cell. The concem about metabolic load largely relates tothe Sec translocon, which is believed to be a bottleneck in the production of periplasmicproteins [36,37]. When production levels of periplasmic proteins are too high, the transloconcould become saturated and the recombinant protein may be retained in the cytoplasm. Todetermine if expression plasmids with a TIRSYNJEVOLVED would saturate the Sec translocon,the inventors induced with either a low (0.05 mM) or a high (0.5 mM) IPTG concentrationand monitored production over a 5-hour period (Figure 3A). It was observed that, at all butone time-point, a TIRSYNJEVOLVED produced more periplasmic ß-lactamase than thecorresponding TIRUNEVOLVED (Figure 3B). This observation was made at both low and highconcentrations of IPTG. These time-course experiments therefore indicated that the Sectranslocon was able to cope with the increased production levels that were reached using aTIRSYNJEVOLVED. Figures 3B illustrates comparative experiment using TIRUNEVOLVED andTIRSYNJEVOLVED wherein (i) TIRSYNJEVOLVED comprised SEQ ID 15, 18, 21, 23 and 26, and (ii)TIRUNEVOLVED comprised SEQ ID of 48, 50, 52, 54 and 56, respectively.

RsYNgavoLvEo Example 4 - Using TI as a generic solution M45-170 In this set of experiments the coding sequences of scFvHER2 and F tYfg were RUNEVOLVED expressed as fusions to the original five signal peptides, using both the TI and TIRSYNJEVOLVED pairs. The expression plasmids were again transforrned into BL21(DE3)pLysS and production was monitored using a mild induction protocol (0.05 mM IPTG for 2hours at 30 °C). As we had observed for ß-lactamase, the TIRSYNJEVOLVED always produced more protein than the corresponding TIRUNEVOLVED (Figure 4). It was noted that signal peptide performance was varied; the most effective signal peptide for production of scFvHERz was PhoASP, whilst the most effective for F tYfgMáß-UO was MalESP. Thus, signal peptideperformance might partly be explained by compatibility of the signal peptide. Figure 4 illustrates comparative experiments using TIRUNEVOLVED and TIRSYNJEVOLVED wherein (i)TIRSYNJEVOLVED comprised SEQ ID 15, 18, 21, 23 and 26, and (ii) TIRUNEVOLVED comprised SEQ ID of 48, 50, 52, 54 and 56, respectively.

A similar approach was taken to produce the human growth hormone (hGH). Here weobserved that the most effective TIRSYNJEVOLVED for production of hGH was the one coupledto the PelBSP (Figure 5A). To assess how much more protein was produced the N-terrninallyHis-tagged hGH was purif1ed by Immobilized Metal Aff1nity Chromatography (IMAC), theHis-tag removed by proteolytic processing, and the sample polished by Size ExclusionChromatography (SEC) (Figure 5B). The yield of purif1ed hGH was more than 3-fold higherusing the TIRSYNJEVOLVED compared to the TIRUNEVOLVED (2.56 mg/L vs 0.79 mg/L).Importantly, we could not detect any difference in the quality of the purif1ed hGH, as judgedby monodispersity of the sample following SEC (Figure 5C), the proportion of protein thathad formed disulphide bonds (Figure 5D), or the activity of the protein when tested by theMTS cell proliferation assay (Figure 5E). Figures 5A, 5C, 5D and 5E illustrate comparativeexperiments using TIRUNEVOLVED and TIRSYNJEVOLVED wherein (i) TIRSYNJEVOLVED SEQ ID 15, 18, 21, 23 and 26, and (ii) TIRUNEVOLVED comprised SEQ ID of 48, 50, 52, 54 and56, respectively. comprised Taken together, this series of experiments indicate that the pET28a-based vectors containing signal peptides with a TIRSYNJEVOLVED can be used as a generic solution to increase productionof single chain antibody fragments, horrnones and other recombinant proteins in the periplasm of E. coli without compromising protein quality.

EXPERIMENTAL PROCEDURESMolecular cloning The sequences encoding MalESP, OmpASP, PhoASP, DsbASP, PelBSP, ß-lactamase, hGH andF tIQfgA/[ÜUO were chemically synthesised (Genscript, USA). The sequence encoding scFvHERzwas obtained from the pHP2-15 plasmid [44]. To generate expression clones, the coding sequences and the pE T 28a vector were amplified by PCR using the Q5 polymerase (New England Biolabs, UK). The coding sequences were then cloned between the Nc0I and NdeIrestriction enzyme sites using the Gibson cloning method. Enzymes used for Gibson cloning were obtained from New England Biolabs, UK.

Synthetic evolution of the TIR TIRLIBRARIES were generated by amplifying the expression plasmids by PCR, using overlapping primers as previously described [34,35]. The forward primer was approximately45 nucleotides in length and was partly degenerate. The design enabled completerandomization of the six nucleotides upstream of the AUG start codon, and partialrandomization of the six nucleotides downstream stream of the AUG start codon(synonymous codons only). The reverse primer was always the same sequence (5 ”-CTCCTTCTTAAAGTTAAACAAAATTATTTCTAGAGGGGAATTGTTATC-3 ”). Itoverlapped with the forward primer by l3 nucleotides thus allowing circularization of thePCR product by homologous recombination in E. coli MCl06l. The PCR was carried outusing the Q5 polymerase (New England Biolabs, UK) in a program that consisted of 94 °C for5 min and then 30 cycles of 95 °C for 45 s, 48-68 °C for 45 s (using a gradient therrnocycler),72 °C for 6 min and a final elongation step of 72 °C for 5 min. Specific PCR products thatwere amplif1ed at the lowest annealing temperature were treated with Dpnl, then transforrnedinto chemically competent E. coli MCl06l. The transformation was seeded into 100 mL ofLuria-Bertani containing 50 ug/mL kanamycin and incubated ovemight at 37 °C. Isolation ofthe TIRLIBRARIES was carried out using ten E.N.Z.A DNA mini kit purification columns (Omega Biotek, USA) and pooling of the eluates.

TIRLIBRARIES were screened by transforrning chemically competent BL2l (DE3) pLysS and identifying clones that survived on the highest concentration of ampicillin. Here 0.5 ug of theTIRLIBRARY was transforrned into 50 uL of chemically competent BL2l (DE 3 ) pLysS usingstandard protocols. The entire transformation was then seeded into 3 mL of LB containing 50ug/mL kanamycin and 34 ug/mL chloramphenicol. Cultures were grown at 37 °C withshaking for 16 h. Cultures were then back-diluted (l 350) into 5 mL of LB containing 50ug/mL kanamycin and 34 ug/mL chloramphenicol and incubated as before until an ODsoo of~0.3 was reached. Expression of the coding sequence was induced by streaking a volume of cells corresponding to 0.002 OD600 units on LB agar containing 0.05 mM isopropyl-ß-Dthiogalactopyranoside (IPTG) and increasing concentrations of ampicillin (l00-5000 ug/mL).

Note that kanamycin and chloramphenicol Were omitted from the plates. The plates Were thenincubated for 16 h at 37 °C. Colonies formed at higher ampicillin concentrations Were selected for further analysis and sequencing (Eurofins MWG operon, Gerrnany).

Immuno-blotting Cultures Were groWn at 37 °C With shaking for 16 h, then back-diluted (1 :50) into 5 mL of LBcontaining 50 ug/mL kanamycin and 34 ug/mL chloramphenicol and incubated as before untilan OD600 of ~0.3-0.5 Was reached. Expression of the coding sequence Was induced With 0.05mM IPTG for 2 h at 30 °C. A Volume of cells corresponding to an OD600 of either 0.02 or 0.2Was harVested by centrifugation then resuspended in 2x Laemlli loading buffer [125 mM Tris-HCl pH 6.8, 4% SDS, 3% Glycerol, 0.02% bromophenol blue, 20% ß-mercaptoethanol].Proteins Were separated by 12% SDS-PAGE then transferred to a nitrocellulose membraneusing a semi-dry transfer apparatus (Bio-Rad, USA). The nitrocellulose membranes Wereprobed With an antibody against either ß-lactamase (Therrno Scientific, USA) or the poly-histidine tag (His-Probe, TherrnoFisher Scientific, USA). Binding Was detected using anti-mouse IgG linked to horseradish peroxidase (GE healthcare, USA) and a SuperSignal Westfemto luminol/enhancer solution (ThermoFisher Scientific, USA). Luminescence emitting from the nitrocellulose membrane Was detected using an Azure Biosystems c600 device.

Disc diffusion assays Cells Were groWn in LB containing 50 ug/mL kanamycin and 34 ug/mL chloramphenicoluntil an ODóoo of ~0.3. A volume of cells corresponding to an ODeoo of 0.002 Was then platedonto LB agar (lacking all antibiotics). A sterile filter disc containing 2 mg ampicillin Was thenplaced on top of the cells and the plates Were incubated at 37 °C for 16 h. Zones of growth inhibition Were measured using a standard ruler.

Purification of hGH Expression plasmids harboring pET28a pelB-h GH Were transforrned into the expression hostBL21(DE3) pLysS and groWn on LB agar plates containing 50 ug/mL kanamycin and 34 ug/mL chloramphenicol. Single colonies Were used to inoculate 100 mL of LB plus antibiotics medium which was grown ovemight at 37 °C with shaking at 180 RPM. Overnightpre-cultures were used to inoculate 2 L flasks containing 1 L of LB media plus antibiotics, toa starting ODóoo of 0.05. Cultures were grown to an OD600 of 0.7, at which point, flasks wereincubated on ice for 10 minutes. Induction proceeded with the addition of 0.01 mM IPTG andincubation for 16 hours at 18 °C with shaking at 180 RPM. Cells were harvested for 20minutes at 4,000 x g. Cell pellets were resuspended in 50 mL suspension buffer (50 mM TrispH 8.0, 500 mM NaCl, 20 mM imidazole pH 8.0 and lx protease inhibitor cocktail(cOmplete, Roche, USA)). Cell suspensions were homogenized with a glass Douncehomogenizer followed by cell disruption using an Avestin emulsiflex C3 high-pressurehomogenizer (Avestin, Canada). Cell debris was removed by centrifugation at 20,000 x g for30 minutes. Samples were applied to 2.5 mL Ni-sepharose (GE Healthcare) and batchincubated at 4 °C for one hour on a benchtop roller. The column was washed with 20 columnvolumes (50 mL) of wash buffer (50 mM Tris pH 8.0, 500 mM NaCl and 50 mM imidazolepH 8.0), followed by elution with 30 mL of elution buffer (50 mM Tris pH 8.0, 500 mM NaCland 500 mM imidazole pH 8.0). The elution fraction was concentrated and buffer exchanged(50 mM Tris pH 8.0, 150 mM NaCl and 20 mM imidazole) using a centrifugal filter with anominal MWCO of 10 kDa (Amicon, Merck Millipore). The N-terrninal his-tag wasproteolytically removed with TEV protease (purified in-house) at a 1:10 weight ratio andallowed to incubate ovemight at 4 °C. Samples were reverse Ni purified, concentrated andapplied to size exclusion chromatography using a Superdex 200 10/300 GL column (GEHealthcare, Sweden) in 50 mM Tris pH 8.0 and 100 mM NaCl. Relevant fractions werepooled, and concentrated. Sample concentration was measured by the BCA protein assay kit(Pierce, TherrnoFisher Scientific, USA) and protein quality assessed by SDS-PAGE.Calculation of final yield per liter was deterrnined by accounting of final volume, final OD at the conclusion of expression, and final concentration of purified hGH.

MTS cell proliferation assay The breast cancer MCF7 cell line (ATCC) was maintained in RPMI-1640 medium containing10% FBS, 2mM glutamine and 1% penicillin streptomycin (Gibco/Therrno Fisher Scientific)at 37 °C in a humidified atmosphere at 5% C02. Cell proliferation following titration ofpurified hGH was deterrnined according to the CellTiter 96 AQueous Non-Radioactive CellProliferation assay (MTS) protocol (Promega). Briefly, 1x104 MCF7 cells were seeded in triplicate, in 100 uL aliquots into 96 Well plates, followed by serum starvation for 24 hours,prior to commencing the proliferation assay. Serially diluted hGH Was added to the mediumat a final concentration ranging from 0 to 400 ng/mL. Cell proliferation Was assessed after 48hours of incubation, by addition of MTS and the electron coupling reagent PMS. The conversion of MTS to forrnazan Was measured by absorbance at 490 nm using a SpectraMax plate reader. Background absorbance Was corrected by subtraction of Wells containing RPMI. hGH EC50 Was calculated using GraphPad Prism 8.1.0.

Claims (14)

1. DNA construct suitable for regulating signal peptide performance, Wherein said DNA construct comprises a sequence of one of SEQ ID 15-28.

2. DNA construct according to claim 1, Wherein said DNA construct also comprises a signal peptide encoding sequence.

3. DNA construct according to claim 2, Wherein said signal peptide encoding sequencecomprises a sequence for eXpressing a signal peptide selected from MalE, OmpA, PhoA, DsbA and Pelb.

4. DNA construct according to any one of the previous claims 2-3, Wherein said signal peptide encoding sequence is a sequence of one of SEQ ID 34-47.

5. DNA construct according to any one of the previous claims 2-4, Wherein said signalpeptide encoding sequence eXpresses a signal peptide of a sequence of one of SEQ ID29-33.

6. DNA construct according to claim 4, Wherein: said MalE signal peptide encoding sequence of one of SEQ ID 34, 35 and 36 eXpresses a signal peptide of a sequence of one of SEQ ID 29; - said OmpA signal peptide encoding sequence of one of SEQ ID 37, 38 and39 eXpresses a signal peptide of a sequence of one of SEQ ID 30; - said PhoA signal peptide encoding sequence of one of SEQ ID 40 and 41eXpresses a signal peptide of a sequence of one of SEQ ID 31; - said DsbA signal peptide encoding sequence of one of SEQ ID 42, 43 and44 eXpresses a signal peptide of a sequence of one of SEQ ID 32; and/or - said PelB signal peptide encoding sequence of one of SEQ ID 45, 46 and 47 eXpresses a signal peptide of a sequence of one of SEQ ID 33.

7. DNA construct according to any one of the previous claims 1-6, Wherein said DNA construct comprises a sequence of one of SEQ ID 15, 18, 21, 23 and 26.

8. DNA construct according to any one of the previous c1aims, Wherein said DNA construct further comprises a recombinant protein encoding sequence.

9. EXpression vector comprising the DNA construct according to any one of c1aims 1-8,Wherein the eXpression Vector is preferab1y a p1asmid, more preferab1y PET eXpression vector, and most preferab1y pet28A Host ce11 comprising the vector according to c1aim 9, Wherein said host ce11 ispreferab1y a bacteria1 ce11, more preferab1y said bacteria1 ce11 is E. co1i and most preferab1y E. co1i strain BL21(DE3) pLysS.

10. Method of eXpressing a recombinant protein, comprising the steps of:a. introducing said DNA construct according to c1aims 8 into an eXpressionvector;b. introducing the eXpression vector into a host ce11;c. growing the host ce11; and d. recovering the recombinant protein from the host ce11.

11. RNA expressed by a DNA construct according to any one of the c1aims 1-8. Use of a DNA construct according to any one of the c1aims 1-8 for regu1ating signa1 peptide performance.

12. Use of a DNA construct according to c1aim 13 for regu1ating signa1 peptideperformance, Wherein said regu1ating signa1 peptide performance is up-regu1ating signa1 peptide performance.