Optimization of signal peptide SP310 for heterologous protein production in Lactococcus lactis

Peter Ravn, José Arnau, Søren M. Madsen, Astrid Vrang and Hans Israelsen

Department of Lactic Acid Bacteria, Biotechnological Institute, Kogle Allé 2, DK-2970 Hørsholm, Denmark

Correspondence
Peter Ravn
pra{at}bioteknologisk.dk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The authors have previously reported the identification of novel signal peptides (SPs) from Lactococcus lactis using transposon insertion. Of these, SP310 caused the highest level of secretion. However, the levels were lower than those obtained using the signal peptide from Usp45 (SPUSP), the major secreted lactococcal protein. In this study, site-directed mutagenesis of signal peptide SP310 was used to investigate the effect of amino acid alterations on lactococcal secretion and to improve secretion efficiency. Several mutated SPs caused higher secretion. This increase in secretion was due to modifications in the cleavage region. In fermenter experiments, the signal peptide SP310mut2 resulted in an extracellular Staphylococcus aureus nuclease (Nuc) yield which was 45 % higher than that with the natural SP310. Surprisingly, increasing the hydrophobicity of the hydrophobic core or increasing the number of positively charged amino acids in the N-terminal region of SP310 decreased secretion. High extracellular yields of Nuc resulted from more efficient secretion, as strains with less efficient SPs accumulated more intracellular SP-Nuc precursor.


Abbreviations: c-region, cleavage region; h-region, hydrophobic region; n-region, N-terminal region; Nuc, Staphylococcus aureus nuclease; SP, signal peptide; SPUSP, signal peptide from Usp45


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Understanding the mechanisms involved in protein secretion is both interesting and important from an applied research point of view. With the development of micro-organisms for production of heterologous proteins, gene expression has been improved in a number of ways. These include development of strong and inducible promoters, high-copy-number vectors, improved codon usage, and refined production strains. When these optimized micro-organisms are used, secretion of the heterologous protein into the culture supernatant might be a limiting step for the yields of the exported proteins. Therefore, the molecular processes of protein secretion are emerging as an important subject of applied research.

Signal peptides (SPs) are the N-terminal extensions present in proteins destined for export by the general (Sec-dependent) secretion system. In bacteria, secretion includes events that occur just after translation of the mRNA, e.g. recognition of the SP in the nascent unfolded polypeptide chain by the Sec-dependent secretion apparatus, translocation through the cell membrane, and cleavage by the signal peptidase (reviewed by van Wely et al., 2001).

Although Gram-positive bacteria, which have no periplasm, are better suited for protein secretion, most published work on SP efficiency has been done on the Gram-negative bacterium Escherichia coli (Chen et al., 1996; Izard et al., 1995, 1996). However, work has also been done on Gram-positive bacteria, especially on the genera Bacillus and Streptomyces (Chen & Nagarajan, 1994; Lammertyn & Anné, 1998; Takimura et al., 1997). Recently, a study in Lactococcus lactis has been published (Le Loir et al., 2001), but it mainly focuses on the effect of amino acids downstream of the cleavage site, i.e. in the mature secreted protein.

The structure of a typical SP includes three distinct regions: (i) an N-terminal region (n-region) that contains a number of positively charged amino acids (lysines and arginines); (ii) a central hydrophobic core region (h-region); and (iii) a hydrophilic cleavage region (c-region) that contains the sequence motif recognized by the signal peptidase. Despite structural similarities, large sequence variations occur in SPs (Nielsen et al., 1997; von Heijne et al., 1983; von Heijne, 1990).

The characterization of numerous extracellular proteins has allowed development of a computer program (SignalP V1.1) that uses artificial neural networks for prediction of the presence of SPs and for the location of signal peptide cleavage sites (Nielsen et al., 1997). The use of SignalP permits the design and preliminary analysis of SP derivatives prior to their construction and test in vivo.

In the Gram-positive L. lactis, the SP of Usp45 (SPUSP), the major secreted protein, has been used for secretion of heterologous proteins (Le Loir et al., 2001; van Asseldonk et al., 1993; Steidler et al., 2000).

We have previously identified a number of novel lactococcal SPs using insertional mutagenesis with a Tn917 derivative containing the Staphylococcus aureus nuc gene, encoding a reporter (Nuc) suitable for analysis of secretion in Gram-positive bacteria (Ravn et al., 2000). SP310 results in higher yields of nuclease than other identified SPs, when plasmid-borne SP–nuc gene fusions are expressed. However, the yield of secreted Nuc using SP310 is only 60 % compared to that obtained using SPUSP (Ravn et al., 2000). This shows that SP310 is suboptimal, and opens up the possibility of studying whether modifications can increase secretion to levels comparable to the levels obtained with SPUSP. Our aim is to study the influence on lactococcal secretion of the amino acids constituting the n-, h- and c-regions of SPs and to optimize SP-directed secretion for production of heterologous proteins. In this study, we have shown that the c-region of SP310 is not optimal, and we have designed several SP310 derivatives which improved secretion yields. Furthermore, in contrast to previous work on E. coli, where increased hydrophobicity of the h-region leads to improved secretion (Izard et al., 1995, 1996; Rusch et al., 1994; Rusch & Kendall, 1994), we found that increasing the hydrophobicity of the h-region decreased the amount of secreted Nuc.

Studies on protein secretion are usually done during exponential growth. In contrast, industrial fermentations for production of heterologous proteins are often done as batch fermentations in which the cultures reach stationary phase. We have chosen to use the P170 promoter, which is well suited for production of heterologous proteins for two reasons. First, the growth phase is separated from protein production driven by P170, because expression from this promoter is induced in the transition from growth phase to stationary phase. Second, no addition of inducer to the culture is needed (Madsen et al., 1999). The choice of P170, however, limits measurements of secreted Nuc to the stationary phase.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and growth conditions.
E. coli K-12 strain DH10B (Grant et al., 1990), grown in LB at 37 °C, was used for cloning purposes, rescue of plasmid DNA, and propagation of plasmid DNA. The medium was supplemented with either 100 µg ampicillin ml-1 or 200 µg erythromycin ml-1, if appropriate. L. lactis subsp. cremoris strain MG1363 (Gasson, 1983) was used for analysis of SPs. L. lactis strains were grown in GM17 (Oxoid) at 30 °C and supplemented with 1 µg erythromycin ml-1, if appropriate. In fermenter experiments a concentrated version of the defined medium SA (Jensen & Hammer, 1993) was used, and pH was maintained using KOH. Bacteria were transformed by electroporation according to published procedures for E. coli (Sambrook et al., 1989) and L. lactis (Holo & Nes, 1989).

Site-directed mutagenesis of SP310 and plasmid constructions.
DNA fragments encoding SP310 and derivatives thereof were made by PCR using pPRA159 (Ravn et al., 2000) as template. PCR products were digested with SmaI and BglII, purified from agarose gels, and ligated into p{Delta}SPNuc (Ravn et al., 2000). The resulting plasmids contained the amplified SP fragments upstream of the S. aureus nuc gene from which the native SP is deleted. Expression of the gene fusion is controlled by the regulated promoter P170 (Madsen et al., 1999). Oligonucleotides used as primers in PCRs are shown in Table 1. The combinations of primers used for construction of the plasmids are shown in Table 2. Plasmids were propagated in E. coli, and the sequences of all SP derivatives were confirmed by sequencing both strands using the Thermo Sequenase fluorescent-labelled primer cycle sequencing kit (Amersham Pharmacia Biotech), Cy5-labelled primers and an ALFexpress DNA Sequencer (Amersham Pharmacia Biotech). Plasmid DNA was subsequently introduced into L. lactis. Strain AMJ627, carrying a similar plasmid encoding SPUSP (Ravn et al., 2000), was also used.


View this table:
[in this window]
[in a new window]
 
Table 1. Oligonucleotides used in this study

 

View this table:
[in this window]
[in a new window]
 
Table 2. Primer combinations used for genetic constructions

 
SDS-PAGE and Western blotting.
Total extracellular proteins were precipitated from culture supernatants with trichloroacetic acid, and proteins were redissolved in an appropriate volume of SDS sample buffer. Intracellular and cell-associated proteins were prepared by the method of Le Loir et al. (1998). Samples were run on 14 % SDS-PAGE Tris/glycine gels from Invitrogen according to the manufacturer. The gel was Coomassie stained. The Mark 12 Wide Range Standard (Invitrogen) was used to estimate molecular sizes. For Western blotting, proteins were transferred to nitrocellulose membranes using the Xcell II blot module (Invitrogen), and transfer was confirmed by Ponceau S staining. Nuc was detected using rabbit anti-Nuc antibodies and alkaline-phosphatase-coupled goat anti-rabbit antibodies from Dako. In this case, the SeeBlue Plus2 protein standard (Invitrogen) was used.

Determination of nuclease yields.
Nuclease yields were determined using a modified version of a previously described method (Cuatrecasas et al., 1967). DNA added as substrate is degraded to mono- and oligonucleotides, which are non-precipitable by 2 % (w/v) perchloric acid. The products of the enzymic degradation were quantified by measuring A260. Nuclease activity was calculated from {Delta}A260/{Delta}t, and one unit was defined as the amount of enzyme which releases 1 µmol acid-soluble nucleotides min-1.

Protein sequence analysis.
Derivatives of SP310 were analysed using the SignalP V1.1 WWW server available at http://www.cbs.dtu.dk/services/SignalP/index.html (Nielsen et al., 1997).

Experimental setup for analysis of secretion efficiency in L. lactis in batch and under fermentation conditions.
In this study, SP310 and its derivatives were analysed in plasmid p{Delta}SPNuc under transcriptional control of P170. This expression system is optimized for protein production in fermenters, but as this study would include a relatively large number of mutants, we tested whether overnight flask cultures could be used for analysis of SP310 derivatives. Strains with SP310 and SPUSP were grown overnight in rich GM17 medium in flasks and in defined SAIV medium in fermenters, and Nuc activity of supernatants was determined. Fermentation experiments were done in Applikon bench-top fermenters containing 1 litre of medium. Fermenters were set to operate at 30 °C and to maintain pH at 5·2.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Site-directed mutagenesis of the c-region of SP310
Previously, we identified several lactococcal SPs using transposon mutagenesis. The most efficient of those is SP310, which, however, yields a lower level of Nuc secretion than SPUSP (Ravn et al., 2000). To improve the secretion efficiency of SP310 and to analyse its n-, h- and c-regions, we constructed a series of mutants using site-directed mutagenesis. Initial analysis of the primary SP310 amino acid sequence was carried out using SignalP V1.1. SignalP V1.1 predicts the existence of SPs by means of neural networks trained on data from eukaryotes, Gram-negative prokaryotes or Gram-positive prokaryotes. Two individual networks assign signal sequence scores and cleavage site scores to each position in the query sequence. The scores from these two networks are used in a combined score and the presence and position of an SP is estimated from all three scores (Nielsen et al., 1997). In this study SignalP was used to identify the c-region as a probable non-optimal region in the SP, and to test in silico the potential efficiency of designed SPs. SignalP readily recognized SP310 as an SP by the means of all scores. However, while mean and maximum signal scores were far above the set cutoff values of 0·55 and 0·95, respectively, the combined score and especially the cleavage score just reached the set cutoff values of 0·34 and 0·42, respectively (Fig. 1a). This suggested that while the overall efficiency of SP310 was fairly good, the cleavage region might be poor. Furthermore, the detailed analysis of the predicted cleavage region of SP310 using SignalP indicated that alternative processing sites might be located at or adjacent to Gln-7 (Fig. 1a).



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 1. Graphical representation of two SignalP outputs and overview of SPs and nuclease yields. (a) The SignalP graphs show sequence, cleavage score, signal score and combined score for separate positions in the SPs. The cleavage and signal score are generated by neural networks trained on data from Gram-positive bacteria, and the combined score is obtained by combining the height of the cleavage score with the slope of the signal score (Nielsen et al., 1997). The presence of an SP is predicted when maximum cleavage score, maximum signal score, maximum combined score and mean signal score are above 0·42, 0·95, 0·34 and 0·55, respectively. Solid arrows indicate the predicted cleavage sites. Dashed arrows mark alternative cleavage sites. (b) The amino acid sequences of SP310 and constructed mutants are shown in grey. Sequence alterations are indicated in black. The last two residues (RS) in each sequence are encoded by the introduced BglII site. The n-, h- and c-regions in SP310 are indicated by open boxes. Nuclease yields in GM17 flask cultures are shown as a percentage of the activity obtained with SP310 (100 %). Relative activities and standard deviations for each mutant were calculated using Nuc measurements from three individual cultures.

 
Bacterial SPs often have an Ala at positions -3, -1 and +1 (Nielsen et al., 1997; von Heijne, 1983). The sequence of SP310 includes a Thr, a relatively infrequent amino acid, at position -3 (Thr-3) (all amino acid numbers refer to SP310). Substitution of this residue with Ala-3 in SP310mut1 resulted in an increase in Nuc secretion (Fig. 1b). Comparison with other Gram-positive SPs (Nielsen et al., 1997) showed that the c-region of SP310 is unusually long. In an attempt to shorten the c-region, Thr-3 and Asn-2 were deleted and Asn-5 was changed to Ala-3 in SP310mut2. This SP retains Ala at positions -3, -1 and +1 (Fig. 1). Nuc secretion using SP310mut2 was 22 % higher than with SP310 and also higher than when SP310mut1 was used (Fig. 1b). These results confirmed that shortening the c-region and maintaining the Ala residues at the cleavage site resulted in a considerable improvement in Nuc secretion in L. lactis. The graphical representation of SignalP analysis of SP310mut2 is shown in Fig. 1(a) as an example of SignalP analysis of mutated SPs. Compared to SP310, the cleavage and combined scores were considerably higher. The use of SP310mut3, in which the amino acid residues are identical to some of the most prevalent residues in Gram-positive c-regions, resulted in a higher level of Nuc secretion compared to SP310, but lower than using SP310mut2 (Fig. 1b).

In prokaryotes, amino acids favouring {beta}-turns (e.g. Pro, Gly, Asp and Ser) are often located in the beginning of the c-region (Nielsen et al., 1997; von Heijne, 1983). In SP310, two Ser (Ser-10 and Ser-9) and one Asp (Asp-6) residues are located in this region. We constructed two different mutants lacking these residues (SP310mut4 and SP310mut5). These alterations resulted in a considerably reduced level of Nuc as compared to SP310 (Fig. 1b).

Analysis of other changes (SP310mut7 and SP310mut8) designed to shorten the c-region of SP310mut2 resulted in significant decreases in secretion (Fig. 1b). In the efficient signal peptide, SPUSP, Asp and Thr are present downstream of the cleavage site at positions +1 and +2, respectively (van Asseldonk et al., 1993). When the amino acids in positions +1 and +2 in SP310mut2 were changed to Asp and Thr (to form SP310mut6), the amount of secreted Nuc decreased from 122 % to 108 % (Fig. 1b). This indicated that Asp and Thr in these positions are not important for the high efficiency of SPUSP.

Site-directed mutagenesis of the h-region of SP310
In both E. coli and Bacillus brevis, increasing the hydrophobicity of the h-region by adding Leu residues can improve secretion (Izard et al., 1995; Takimura et al., 1997). To study the role of Leu in the h-region in lactococcal secretion, a series of mutants were constructed and characterized. These included three, six or 14 Leu residues, in addition to Leu-19 present in SP310. In SP310mutA, SP310mutB and SP310mutC, Leu was introduced at different positions in the hydrophobic region, while the wild-type sequence was maintained in the c-region (Fig. 1b). A large reduction in Nuc yield (41 % relative to SP310) was observed using SP310mutA, and even lower yields were obtained with SP310mutB (37 %) and SP310mutC (34 %). This indicates that increasing the number of Leu residues gradually decreased the secretion level for this protein. This is in contrast to the reported results from E. coli and B. brevis (Izard et al., 1995; Takimura et al., 1997). Modified versions of SP310mutA and SP310mutB that additionally incorporated alanines at position -3 (corresponding to the c-region of 310mut1) were also studied. Using SP310mutA1, the level of secretion was somewhat higher (59 % compared to using SP310), indicating that conserved Ala positions in the c-region partially compensate for the presence of a moderate excess of Leu in the h-region in this lactococcal SP. However, when six leucines were present (SP310mutB1), the yield was similar to that obtained when using SP310mutB, indicating that the higher Leu content in the h-region cannot be compensated for by the presence of Ala-3 (Fig. 1b).

Increasing the number of Phe residues in the h-region, which also increases hydrophobicity, has been shown to enhance efficiency of an SP in E. coli (Rusch & Kendall, 1992). Since the inclusion of additional Leu residues in the hydrophobic region of SP310 did not result in increased Nuc secretion in L. lactis, the effect of changing the number of Phe residues in this region was analysed. However, secretion efficiency was decreased by the introduction of both one (in SP310mut2 to give SP310mut10) and two (in SP310mutD2 to give SP310mutE11) Phe residues (Fig. 1b).

Site-directed mutagenesis of the n-region of SP310mut2
The n-region of SPs, especially from Gram-positive bacteria, is characterized by the presence of positively charged amino acids (Nielsen et al., 1997). Increasing the positive charge in this region can improve secretion efficiency in B. brevis (Takimura et al., 1997). The importance of positive charge density has also been shown for E. coli (Izard et al., 1996). To investigate the effect of increasing the positive charge density of the n-region in L. lactis, a series of mutants based on SP310mut2 were constructed. In the case of SP310mutD2, removal of Phe-32 resulted in Nuc secretion levels higher than with SP310 but lower than with SP310mut2 (Fig. 1b). Removal of Asn-31 (SP310mutE2) also resulted in a small, but significant, decrease in secretion compared to when using SP310mut2 (Fig. 1b). In SP310mutF2, the positive charge was increased by changing Asn-31 in SP310mut2 to Lys-31, resulting in the lowest level of Nuc secretion in any of the mutants analysed (Fig. 1b). Thus, the charge of the n-region in SP310 might represent the maximum allowed for efficient secretion in L. lactis for this protein.

Secretion and intracellular accumulation of Nuc using SP310 and selected mutants
Secreted proteins in supernatants from overnight cultures of 11 selected strains were analysed by SDS-PAGE (Fig. 2a). One major band corresponding in size to NucB could be seen in all strains where an SP is used. In addition, a weak band corresponding in size to NucA could be seen in some lanes. NucA arises through the action of the HtrA cell-surface protease on NucB (Poquet et al., 2000), and both NucA and NucB are enzymically active. The relative amounts of Nuc were in agreement with the activity levels measured in these strains (compare Fig. 1b and Fig. 2a).



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 2. SDS-PAGE of secreted nuclease and Western blotting of cell lysates. Strains producing the indicated SP-Nuc fusion were analysed. The positions of NucB, NucA and SP-NucB precursors are indicated. (a) Supernatants from overnight cultures were concentrated, and volumes corresponding to 0·5 ml culture supernatants were applied. The gel was Coomassie stained. (b) Western blotting of cell lysates. Volumes corresponding to 33 µl of culture were applied.

 
To determine the fate of non-secreted Nuc, its intracellular accumulation was monitored in the same subset of strains by Western blotting of cell lysates using anti-Nuc antibodies (Fig. 2b). When Nuc was produced without an SP ({Delta}SP) a band corresponding in size to NucB was seen in cell lysates. When Nuc was produced in fusion with an SP two bands could be seen, one corresponding in size to the SP-NucB precursor, and the other to NucA. In most cases where an inefficient SP was used, high intracellular accumulation of SP-NucB was seen. This was clearest for SP310mut4, SP310mutA and SP310mutC. In the case of SP310mutF2, which was the least efficient SP, a band of intermediate strength was seen. This showed that no additional intracellular accumulation took place in this case. On the other hand, the SP310mutF2-NucB band was at least as strong as the SP310mut2-NucB band, showing that the low extracellular Nuc yield obtained with SP310mutF2 was not due to low precursor production or stability. The amounts of cell-associated NucA (Fig. 2b) were proportional to the amounts of secreted NucB (Fig. 2a), and inversely proportional to the amounts of intracellular SP-NucB (Fig. 2b).

Nuc secretion in fermenter cultures
Strains with SP310 and SPUSP secreted approximately 25-fold more Nuc in fermenter cultures than in overnight cultures. Therefore, three mutants representing high (SP310mut2), medium (SP310mut6) and low (SP310mutB) secretion efficiency were compared to SPUSP and SP310. SP310mut2 yielded 20·8 units Nuc l-1 in fermenter culture, representing a 45 % improvement relative to SP310 (Table 3). SP310mut6 yielded 25 % more than SP310, while SP310mutB yielded 50 % less. The results obtained in fermenter cultures resembled the results of the initial screening in overnight cultures, showing that this screening was successful, and confirming that we have succeeded in obtaining a mutant of SP310, SP310mut2, which was superior to the natural SP310.


View this table:
[in this window]
[in a new window]
 
Table 3. Comparison of nuclease secretion in fermenter and in overnight cultures

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study, we have examined the signal peptide SP310 from L. lactis (Ravn et al., 2000) in detail. This was done to learn more about the effect on secretion of amino acid alterations in the three domains of the SP, and to improve the secretion efficiency. Site-directed mutagenesis was carried out to introduce mutations in one or more of the three domains. As secretion reporter we used the S. aureus Nuc protein previously used for secretion studies in L. lactis (Le Loir et al., 1998, 2001; Poquet et al., 1998). We found that the natural signal peptide, SP310, could be improved further. Several changes in the cleavage region improved Nuc secretion efficiency, while changes in other regions all decreased secretion efficiency. Immunoblotting experiments demonstrated that improved extracellular yields originated from more efficient secretion, because intracellular precursors accumulated to a lesser extent with more efficient SP variants.

The sequence diversity of SPs, within Gram-positive as well as Gram-negative bacteria, impedes the direct design of optimal sequences for secretion of heterologous proteins. Furthermore, in E. coli the natural SPs are not optimized with regard to secretion efficiency (Rusch et al., 1994), and therefore SP designs based solely on natural SPs are not likely to be optimal. Initial studies of SPs can be carried out effectively using SignalP. Therefore, a detailed investigation by site-directed mutagenesis was carried out to optimize secretion efficiency. Additionally, this approach may provide insight into the role of individual amino acids in signal peptides.

SignalP analysis suggested that the c-region of SP310 was not optimal. This is seen by the relatively low cleavage score and combined score (Fig. 1a). All mutants of SP310 constructed for this study had higher cleavage and combined scores. An example of this is SP310mut2 (Fig. 1a). Furthermore, the c-region of SP310 was unusually long compared to other known lactococcal SPs (Table 4). The increased efficiency of SP310mut1, SP310mut2 and SP310mut3 (Fig. 1) showed that better c-regions could indeed be designed. The importance of having an Ala residue in position -3 was demonstrated by SP310mut1. Ala is also preferred in this position in natural SPs (Nielsen et al., 1997). The secondary structure of SPs is probably important for secretion, and turn-promoting residues like Pro and Gly are often located at the beginning of the c-region (Nielsen et al., 1997; von Heijne et al., 1983). SP310, however, does not contain Pro or Gly. Instead, it contains one Asp and two Ser residues, and these amino acids are also known to be associated with {beta}-turns. The decreased efficiency of SP310mut4 and SP310mut5 (Fig. 1b) shows the importance of these amino acids for secretion efficiency. Other lactococcal SPs do also contain turn-promoting residues at this position (Table 4). Asp followed by Thr or Ser are often found at position +1 and +2 in lactococcal SPs (Table 4). We demonstrated that the change of Ala+1 and Glu+2 in SP310mut2 to Asp+1 and Thr+2 in SP310mut6 resulted in a decrease in Nuc secretion efficiency. Therefore, Asp and Thr in positions +1 and +2 do not seem to be important for lactococcal secretion of this protein.


View this table:
[in this window]
[in a new window]
 
Table 4. Signal peptide structure in L. lactis

 
The hydrophobicity of the h-region is an essential feature of an SP. But contrary to results from E. coli and B. brevis, increasing the degree of hydrophobicity of SP310 by adding Phe or Leu residues resulted in a decrease in Nuc secretion (Fig. 1b). In E. coli, increasing the number of Leu or Phe residues in the h-region can lead to increased PhoA secretion and can even suppress defects in other regions of the SP, or promote secretion of an otherwise non-secretable PhoA variant (Izard et al., 1995, 1996; Rusch et al., 1994; Rusch & Kendall, 1992). In B. brevis, additional Leu residues in the h-region also had a positive effect on IL-2 secretion (Takimura et al., 1997). This inconsistency might be due to differences that are SP- or species-specific or related to the secreted protein.

Four SP mutants with altered n-regions were made and tested. All had increased positive charge density of the n-region, and use of these SPs yielded less secreted Nuc than use of SP310mut2 (Fig. 1b). SP310mutF2, which contains an extra lysine, was least efficient of all tested SPs. For SP310mutD2 and SP310mutE2, the effect was less pronounced, and in these two cases it remains unclear whether the change in length, alteration in charge density, or polarity in the n-region was responsible for the decrease in the yield of secreted Nuc. Because the n-regions of SP310mutD2 and SP310mutE11 are identical, the very low efficiency of the latter can be ascribed to the presence of extra Phe residues in the h-region. In B. brevis, addition of positively charged amino acids to the n-region of an SP can increase secretion (Takimura et al., 1997). In the case of SP310, however, the n-region contains one Arg and three Lys residues. These four residues contribute more positive charge than usually found in SPs from L. lactis (Table 4), and might therefore represent the maximum charge allowed.

The extracellular yield of Nuc could also be affected by factors such as efficiency of transcription and translation, or stability of mRNA and intracellular SP-Nuc precursors. Western blotting, carried out on 11 selected strains, showed intracellular accumulation of SP-Nuc precursor in most of the analysed strains having less efficient SPs, demonstrating that low extracellular yields were not a result of low precursor production or stability. In the case of SP310mutF2, high intracellular SP-Nuc accumulation was not observed. The intracellular level was as least as high as with SP310mut2, showing that also in this case the low extracellular yield was not a result of low precursor production or stability. While NucB was mainly observed in the culture supernatants (Fig. 2a), NucA was associated with cells (Fig. 2b). It is possible that NucA either had a stronger binding to the cell surface compared to NucB, or alternatively, HtrA mainly acted on NucB molecules that were already associated with the membrane surface or cell wall.

Representative SP310 derivatives were tested in fermenter cultures. The results corroborate the initial finding, obtained with SP310 and SPUSP, that expression levels in fermenter cultures were higher, but that the ratio between SP efficiencies was conserved. Furthermore, this indicates that the factors hampering SP efficiency influence the fraction of secreted precursor rather than creating a fixed limit on the amount of secreted protein. Higher yields in fermenter cultures were a result of higher promoter activity because intracellular protein production driven by P170 was also improved in fermenter cultures (data not shown).

Secretion of proteins depends on the efficiency of the SP, on intrinsic properties of the secreted protein and on host factors. In this study we have focused on optimization of the SP, but improvements in lactococal secretion have also been obtained by modification of the N-terminus of the secreted protein (Le Loir et al., 1998) or by inactivation of the surface protease HtrA (Miyoshi et al., 2002).

Based on our results, we conclude that naturally occurring lactococcal SPs can be optimized, as we were able to construct more efficient derivatives of SP310, and that optimization can successfully be carried out using bioinformatics tools such as SignalP. However, it is clear that theoretically optimal SPs should be tested experimentally.

We have succeeded in designing several SPs that caused improved efficiency in both simple flask cultures and in fermenter cultures with high production levels. We are currently using SP310mut2 for production of other heterologous proteins in L. lactis. For some of these proteins, the extracellular yield has been identical to or even higher than that obtained when SPUSP was used (data not shown).


   ACKNOWLEDGEMENTS
 
This work was supported by the FØTEK-2 program in a collaboration between the Biotechnological Institute and MFF (the Danish Dairy Board). We thank Anne Cathrine Steenbjerg, Annemette Jørgensen and Eva Hjerl-Hansen for outstanding technical assistance. We also thank Alexandra Gruss and Philippe Langella (INRA, France) for the gift of anti-Nuc antibodies and for useful discussions, and Bjarne Albrechtsen for helpful comments on this manuscript.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Chen, M. & Nagarajan, V. (1994). Effect of alteration of charged residues at the N termini of signal peptides on protein export in Bacillus subtilis. J Bacteriol 176, 5796–5801.[Abstract]

Chen, H., Kim, J. & Kendall, D. A. (1996). Competition between functional signal peptides demonstrates variation in affinity for the secretion pathway. J Bacteriol 178, 6658–6664.[Abstract]

Cuatrecasas, P., Fuchs, S. & Anfinsen, C. B. (1967). Catalytic properties and specificity of the extracellular nuclease of Staphylococcus aureus. J Biol Chem 242, 1541–1547.[Abstract/Free Full Text]

Gasson, M. (1983). Plasmid complements of Streptococcus lactis NCDO712 and other lactic streptococci after protoplast-induced curing. J Bacteriol 154, 1–9.[Medline]

Grant, S. G., Jesse, J., Bloom, F. R. & Hanahan, D. (1990). Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants. Proc Natl Acad Sci U S A 87, 4645–4649.[Abstract]

Holo, H. & Nes, I. F. (1989). High-frequency transformation, by electroporation of Lactococcus lactis subsp. cremoris grown with glycine in osmotically stabilized media. Appl Environ Microbiol 55, 3119–3123.

Izard, J. W., Doughty, M. B. & Kendall, D. A. (1995). Physical and conformational properties of synthetic idealized signal sequences parallel their biological function. Biochemistry 34, 9904–9912.[Medline]

Izard, J. W., Rusch, S. L. & Kendall, D. A. (1996). The amino-terminal charge and core region hydrophobicity interdependently contribute to the function of signal sequences. J Biol Chem 271, 21579–21582.[Abstract/Free Full Text]

Jensen, P. R. & Hammer, K. (1993). Minimal requirements for exponential growth of Lactococcus lactis. Appl Environ Microbiol 59, 4363–4366.[Abstract]

Lammertyn, E., & Anné, J. (1998). Modifications of Streptomyces signal peptides and their effects on protein production and secretion. FEMS Microbiol Lett 160, 1–10.[CrossRef][Medline]

Le Loir, Y., Gruss, A., Ehrlich, S. D. & Langella, P. (1998). A nine-residue synthetic propeptide enhances secretion efficiency of heterologous proteins in Lactococcus lactis. J Bacteriol 180, 1895–1903.[Abstract/Free Full Text]

Le Loir, Y., Nouaille, S., Commissaire, J., Bretigny, L., Gruss, A. & Langella, P. (2001). Signal peptide and propeptide optimization for heterologous protein secretion in Lactococcus lactis. Appl Environ Microbiol 67, 4119–4127.[Abstract/Free Full Text]

Madsen, S. M., Arnau, J., Vrang, A., Givskov, M. & Israelsen, H. (1999). Molecular characterization of the pH-inducible and growth phase-dependent promoter P170 of Lactococcus lactis. Mol Microbiol 32, 75–87.[CrossRef][Medline]

Miyoshi, A., Poquet, I., Azevedo, V. & 7 other authors (2002). Controlled production of stable heterologous proteins in Lactococcus lactis. Appl Environ Microbiol 68, 3141–3146.[Abstract/Free Full Text]

Nielsen, H., Engelbrecht, J., Brunak, S. & von Heijne, G. (1997). A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Int J Neural Syst 8, 581–599.[Medline]

Poquet, I., Ehrlich, S. D. & Gruss, A. (1998). An export-specific reporter designed for gram-positive bacteria: application to Lactococcus lactis. J Bacteriol 180, 1904–1912.[Abstract/Free Full Text]

Poquet, I., Saint, V., Seznec, E., Simoes, N., Bolotin, A. & Gruss, A. (2000). HtrA is the unique surface housekeeping protease in Lactococcus lactis and is required for natural protein processing. Mol Microbiol 35, 1042–1051.[CrossRef][Medline]

Ravn, P., Arnau, J., Madsen, S., Vrang, A. & Israelsen, H. (2000). The development of TnNuc and its use for the isolation of novel secretion signals in Lactococcus lactis. Gene 242, 347–356.[CrossRef][Medline]

Rusch, S. L. & Kendall, D. A. (1992). Signal sequences containing multiple aromatic residues. J Mol Biol 224, 77–85.[Medline]

Rusch, S. L. & Kendall, D. A. (1994). Transport of an export-defective protein by a highly hydrophobic signal peptide. J Biol Chem 269, 1243–1248.[Abstract/Free Full Text]

Rusch, S. L., Chen, H., Izard, J. W. & Kendall, D. A. (1994). Signal peptide hydrophobicity is finely tailored for function. J Cell Biochem 55, 209–217.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Steidler, L., Hans, W., Schotte, L., Neirynck, S., Obermeier, F., Falk, W., Fiers, W. & Remaut, E. (2000). Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 289, 1352–1355.[Abstract/Free Full Text]

Takimura, Y., Kato, M., Ohta, T., Yamagata, H. & Udaka, S. (1997). Secretion of human interleukin-2 in biologically active form by Bacillus brevis directly into culture medium. Biosci Biotechnol Biochem 61, 1858–1861.[Medline]

van Asseldonk, M., de Vos, W. M. & Simons, G. (1993). Functional analysis of the Lactococcus lactis usp45 secretion signal in the secretion of a homologous proteinase and a heterologous alpha-amylase. Mol Gen Genet 240, 428–434.[Medline]

van Wely, K. H. M., Swaving, J., Freudl, R. & Driessen, A. J. M. (2001). Translocation of proteins across the cell envelope of gram-positive bacteria. FEMS Microbiol Rev 25, 437–454.[CrossRef][Medline]

von Heijne, G. (1983). Patterns of amino acids near signal-sequence cleavage sites. Eur J Biochem 133, 17–21.[Abstract]

von Heijne, G. (1990). The signal peptide. J Mol Biol 115, 195–201.

Received 12 February 2003; revised 24 April 2003; accepted 8 May 2003.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Ravn, P.
Articles by Israelsen, H.
Articles citing this Article
PubMed
PubMed Citation
Articles by Ravn, P.
Articles by Israelsen, H.
Agricola
Articles by Ravn, P.
Articles by Israelsen, H.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2003 Society for General Microbiology.