Serratia ATP-binding cassette protein exporter, Lip, recognizes a protein region upstream of C-terminus for specific secretion

Serratia marcescens ATP-binding cassette exporter, Lip system, secretes lipase (LipA SM ), metalloproteases, and cell surface layer protein homologue but not a heme-acquisition protein, HasA (HasA SM ). Secretion of HasA SM is limited to Has SM system. However, HasA proteins from Pseudomonas fluorescens (HasA PF ) and Pseudomonas aeruginosa were exported through Lip and Has SM systems. In order to investigate the specificity in Lip exporter-mediated secretion, secretion analysis was performed using chimeras containing HasA PF and HasA SM sequences. The segment Val-Ala-Leu (designated R1 to R3 sites), which is present close to the C-terminus of HasA PF but not HasA SM , was revealed to be involved in the substrate specificity of Lip exporter. Introduction of amino acid substitutions into the region R1 to R5 demonstrated that R1, R3, R4, and R5 sites require some specific amino acid residues for Lip-mediated secretion. The amino acid sequence of the region was conserved considerably among the proteins secreted by Lip exporter.


INTRODUCTION
motifs are analogous, and therefore, the secretion specificity of HasA SM cannot be elucidated simply by the motif.
Specific recognition by ABC exporter is an essential process for translocation of secretory proteins, and therefore, the specificity of HasA SM secretion is one of important subjects for understanding the secretion mechanism of ABC exporter. In this paper, we aim to describe the characteristics of the sequence recognized by ABC exporter using HasA SM and LipA SM mutants as a substrate and Lip system as an exporter. We reveal involvement of a protein region close to but not at the C-terminus in Lip exporter-mediated secretion and in recognition by ABC protein, LipB. Furthermore, we demonstrate that a typical C-terminal motif, which is considered as a secretion signal so far, is not essential for secretion. Our findings provide new important information for understanding the secretion mechanism and protein structure involved in the secretion specificity of ABC exporter.

EXPERIMENTAL PROCEDURES
Strains and media-E. coli K-12 DH5 was used as a host and LB medium was used as a rich medium (22). S. marcescens 413 that is a LipA SM -deficient strain (11) was used for a host of the secretion analysis of LipA SM mutants. Antibiotics were added at the following concentrations: ampicillin, 50 µg/ml; kanamycin, 50 µg/ml; and chloramphenicol, 20 µg/ml for E. coli, and ampicillin, 1000 µg/ml for S. marcescens

413.
General methods-DNA manipulations were carried out according to standard procedures (21). PCR was carried out using ExTaq purchased from Takara Shuzo (Kyoto, Japan). Nucleotide sequence was determined by dideoxy-chain-termination for 40h with shaking vigorously. After TCA-precipitation, the proteins in the supernatants of the cultured media were loaded on the gel as described above, and then electrophoretically transferred to an Immobilon P filter (Millipore). The blots were blocked by soaking in Block Ace (Dainippon Pharmaceutical, Japan) overnight at 4°C and incubated with an anti HasA SM antibody (a kind gift from Dr. Cecile Wandersman) at room temperature for two hours (diluted 1:5000 in PBS containing 0.1% Tween20).
They were washed and incubated with horseradish peroxidase-conjugated anti-rabbit immunoglobulin G, and the bound antibody was detected with the enhanced chemiluminescence systems (Amersham). the presence of segments, composed of 12 and 14 amino acid residues, close to but not at the C-termini of HasA PF and HasA PA , respectively ( Fig. 2A). Involvement of the segment in the Lip-mediated secretion was tested using HasA chimeras containing the HasA SM and HasA PF sequences (Fig. 2B). Secretion of all HasA chimeras through Has SM exporter has been already confirmed (21). As shown in Fig Secretion of these HasA mutants through Lip exporter provided us several interesting observations. First, importance of R1 residue for secretion was exhibited ( Fig. 4A). Replacement with Ile did not alter the secretion level, whereas other substitutions including hydrophobic residues, Leu, Phe, and Met, reduced it to undetectable levels. These observations implied that the presence of a hydrophobic residue at R1 is necessary but not sufficient for Lip-promoted secretion. Val and Ile at R1 are likely to play a specific role for secretion. L ip exporter secreted HasA mutants with replacements at R2 but except for Pro (Fig. 4A), indicating that Ala residue at R2 is not essential. D eletion of the R2 residue (pMFM-ATDVL), however, disturbed Lip-mediated secretion (Fig. 3B), suggesting that the residue functions as a spacer to allocate the hydrophobic R1 at a proper position necessary for the secretion. R3 residue was replaceable with Ile ( Fig. 4B). Decrease in the secretion levels (but at a detectable level) was observed by replacement with Val at R3. Thr-substitution by guest on March 23, 2020 http://www.jbc.org/ Downloaded from allowed secretion of the mutants at a low level. HasA mutants containing other residues at R3 were not secreted via Lip system at a detectable level.

Secretion competition analysis-The
Thus, the (Val/Ile)-X-(Leu/Ile/Val) sequence upstream of the C-terminus was demonstrated to be essential for secretion through Lip exporter. Because the sequence was unrelated to Has SM -mediated secretion, the region R1 to R3 was suggested to be involved in the recognition or specific transport process by Lip exporter. In conclusion, the amino acid residues at R1, R3, R4, and R5 were strongly related to Lip-mediated HasA secretion. The residue Gly at the R5 site also played an important role in high-level secretion through Has SM . Ala-substitution at R3) was secreted by Lip and hybrid exporters. HasD SM -LipC-LipD has been already known inactive for secretion (11), and an exporter lacking ABC protein did not direct secretion. Thus, the sequence Val-Ala-Leu was essential for secretion through exporters including the ABC protein LipB, confirming that LipB is a determinant of the substrate specificity of the exporter. In other words, LipB might recognize the region R1 to R3 of the secretion protein.

C-terminal secretion signal is not essential for secretion via ABC
exporters-Proteins secreted by Lip exporter also contain a conserved motif consisting of a negatively charged amino acid residue followed by several hydrophobic residues at the C-terminus (Fig. 5). Involvement of the motif in secretion was investigated using HasA SM -VAL mutants (Fig. 7A). Interestingly, Ala-substitutions of 184 Glu or/and These findings indicated that the C-terminal motif so far considered as a secretion signal, is not essential for secretion though ABC exporter. Although the motif may have a role for something other than secretion, the importance of the region R1 to R5 in secretion via ABC exporters except for Has SM was generally recognized. Influence of location of the region R1 to R5 from the C-terminus on secretion was examined (Fig. 8C). As shown in Fig. 8D, LipA SM mutants, having Gly (R5) 13 to 17 amino acid residue upstream from the C-terminus, were secreted but at various levels, whereas the secretion levels were low on mutants containing Gly (R5) 11-12 and in the E. coli cells without exporter exhibited intracellular lipase activity, indicating production of lipase proteins (data not shown). Position of the R1 to R5 region, but not a specific sequence of the C-terminal tail, was demonstrated to be critical for Lip-mediated secretion based on the similarity of the C-terminal tails of proteins secreted by Lip system and the results from the secretion analysis of LipA SM mutants. observed, that is, proteins retaining an R1 to R5 motif inactive for Lip-mediated secretion, did not cause a blockade of secretion. These findings indicate that proteins including the inactive R1 to R5 motif do not have high affinity to ABC protein, which leads to an impediment to the secretion process. The protein region carrying an active R1 to R5 motif is recognized by ABC protein with higher affinity than that containing inactive one. Secretion competition between LipA SM and PrtA in S. marcescens, which is previously reported in ref 25, is explained by competition in the process of recognizing the R1 to R5 regions of these proteins by Lip system. Thus, a new determinant within secretory proteins, which enables secretion through Lip system, in by guest on March 23, 2020 http://www.jbc.org/ Downloaded from other words, which Lip system recognizes, was revealed. Secretion profile of LipA SM mutants with Ala-replacements differed from that of HasA SM mutants, for example, a conserved Gly residue at R5 was demonstrated necessary for HasA mutant secretion via Has SM and Lip exporters but not essential for LipA SM secretion via Lip system. These observations indicate that a motif in the region is not rigorous and that amino acid residues in the region R1 to R5 possibly affect secretion through Lip exporter each other.
In addition, LipA SM mutants containing Gly (R5) at positions -14 to -17 from the C-terminus were secreted through Lip exporter, whereas those with a longer or shorter C-terminal tail were a poor substrate. Any specific feature was not found in the sequence between R5 and the C-terminus, and the alteration of the C-terminal tail sequence by insertion or deletion of one or two amino acid residues did not change the secretion level severely, indicating that location of the region R1 to R5 from the C-terminus is another important factor.
One report (13) supports this idea. In E. chrysanthemi PrtG, the C-terminal 15 amino acid sequence Val-Asn-Ile-Val-Gly-Ala-Ala-Leu-Gln-Pro-Ser-Asp-Val-Ile-Val-COOH, which includes the region R1 to R5 (underlined) at position -11 from the C-terminus, has been shown to be a minimum functional part for Prt-mediated secretion. Furthermore, PrtG mutant lacking the C-terminal 14 amino acid residues but exhibiting a newly created typical C-terminal motif, was secreted, and therefore, a conserved C-terminal motif was regarded as a secretion signal. The deletion, however, located the sequence Leu-His-Leu-Ser-Gly, which is similar to the region R1 to R5, at position -9. Their previous findings might imply the importance of the region R1 to R5 at a proper position in secretion.
What is a function of the conserved C-terminal motif? At least, the motif is not by guest on March 23, 2020 http://www.jbc.org/ Downloaded from necessary for secretion through Lip, Prt, and Has SM exporters. Considering that the motif is conserved among proteins secreted through ABC exporters including highly conserved repeat toxin family exporters, the motif may be necessary for protein-folding after secretion through ABC exporters. However, its real function has not been defined yet.
In order to clarify recognition mechanism by Lip system, the analysis of structural features laying the region R1 to R5 is necessary. Although 3D-structure of LipA SM is unknown presently, crystallographic analyses of AprA PA and PrtA, both of which are secreted by Lip system and possess the sequence homologous to the region R1 to R5, show very similar C-terminal structure (30, 31). The conserved C-terminal motif and a protein region corresponding to the region R1 to R5 are β-sheet structured.
On the contrary, conformation of the HasA SM C-terminus produced in E. coli, which is the last 15 residues containing the region R4 to R5, has been studied and revealed to be highly flexible and unstructured (32), suggesting that the C-terminal structure of HasA SM -VAL having the region R1 to R5 is unstructured. Probably, C-terminal tails of secretory proteins are unstructured during a process of secretion, and therefore, no rigid conformation seems to be necessary for secretion through Lip system although the region R1 to R5 plays an important role for the secretion.
Our findings reported here opened a new avenue for investigating a secretion mechanism of ABC exporter. Besides the crystallographic analysis of LipA SM mutants or HasA chimeras, studies on a protein region of the ABC protein LipB, which is involved in the specific recognition by Lip system, are intriguing. Although 3D-structure of the E. coli outermembrane protein TolC, which demonstrates that a tunnel connected to the external environment is constituted with trimeric TolC protein, has been reported recently (33, 34). However, 3D-structure of ABC protein and mechanism of secretion is still unknown. Chimeric ABC proteins containing the LipB and HasD SM sequences or mutants of ABC proteins will be helpful and informative for the exploration of the secretion specificity of Lip and Has SM exporters. Through these approaches, a mechanism of the substrate recognition by ABC exporter will be totally understood.