Rim domain loops of staphylococcal β-pore forming bi-component toxin S-components recognize target human erythrocytes in a coordinated manner

Rim domain loops of staphylococcal β-pore forming bi-component toxin S-components recognize... Abstract Staphylococcus aureus bi-component pore-forming toxins consist of S- and F-components, and form hetero-octameric beta-barrel pores on target blood cell membranes. Among them, γ-haemolysin (Hlg2 and F-component of Luk (LukF)) and LukED (LukE and LukD) possess haemolytic activity, whereas the Panton-Valentine leukocidin (LukS-PV and LukF-PV) does not lyse human erythrocytes. Here, we focussed on four loop structures in the rim domain of S-component, namely loops -1, -2, -3 and -4, and found that replacement of Loop-4 in both Hlg2 and LukE with that of LukS-PV abolished their haemolytic activity. Furthermore, LukS-PV gained haemolytic activity by Loop-4 exchange with Hlg2 or LukE, suggesting that Loop-4 of these S-components determined erythrocyte specificity. LOOP-1 and -2 enhanced the erythrocytes-binding ability of both components. Although Hlg2 and LukE recognize Duffy antigen receptor for chemokines on human erythrocytes, the ability of Loop-4 was not complementary between Hlg2 and LukE. Exchange of Hlg2 with LukE Loop-4 showed weaker activity than intact Hlg2, and LukE mutant with Hlg2 Loop-4 lost its haemolytic activity in combination of LukD. Interestingly, the haemolytic activities of these Loop-4 exchange mutants were affected by F-component, namely LukF enhanced haemolytic activities of these Hlg2 and LukE Loop-4 mutants, and also haemolytic activity of LukS-PV mutant with LukE Loop-4. protein, bacterial, protein-protein interactions, toxins, drugs, xenobiotics, staphylococcal bi-component toxin, S-components, CCR5 Staphylococcus aureus produces several types of leukocidin family bi-component toxins belonging to the β-barrel pore-forming toxins (PFTs). Leukocidin family toxins comprise the F- and S-component proteins, and form pores containing both components in a 1:1 ratio on the surface membrane of target white blood cells (1). Almost all isolates of S. aureus possess the leukocidin (Luk) cluster consisting of genes encoding LukF (F-component) and LukS (S-component), alternatively designated as HlgB and HlgC, respectively, on their chromosome. Hlg2 (also called HlgA), another S-component with its gene located just upstream of lukS, shows strong haemolytic activity against human erythrocytes in combination with LukF. Therefore, LukF and Hlg2 are called γ-haemolysin (Hlg). At an extremely low dose of 104 molecules of each component per cell, Hlg lyses human erythrocytes within 10 min at 37°C, whereas Luk (with LukS) does not show haemolytic activity under the same conditions, suggesting that the S-component plays an important role in host cell specificity (1–3). F- and S-components of Hlg/Luk have similar structures, and form hetero-octameric beta-barrel pores with the F- and S-components being arranged alternately on the target cell membrane (4, 5). To date, a number of staphylococcal leukocidin variants, closely related to Hlg/Luk, such as Panton-Valentine leukocidin (PVL LukF-PV and LukS-PV) (6) and LukED (LukD and LukE) (7, 8), have been identified. Among them, LukED has been reported to possess haemolytic activity against human erythrocytes (8, 9), whereas PVL specifically acts on human leukocytes, but has no haemolytic activity against human and rabbit erythrocytes (6). Components of bi-component toxins are secreted by the bacterium as soluble proteins, and form insoluble membrane pore on the target cell surface. The pore-forming event by toxins begins with the binding of soluble protein components to the target cell. Alpha-haemolysin (Hla), a staphylococcal β-PFT forming homo-heptamers, has a phosphatidylcholine (PC)-binding site in the rim domain that is involved in interacting with the erythrocyte surface (10). The monomeric structure of LukF is similar to that of Hla in the cap and rim domains, and its PC-binding site comprising of W177, E192 and R198, which directly interact with the choline head of PC or sphingomyelin in the plasma membrane, is confirmed to be present in the rim domain (11–13). In addition, three aromatic residues, namely Y72, F260 and Y261, which are important for binding to erythrocytes, are located in the loop structure at the bottom of the rim domain of LukF (1, 12). These observations suggest that the loop structures in the rim domain of Hla and LukF play an important role in cell recognition and cell binding. On the other hand, details of the mechanism involving binding of the S-component to target cells remain to be investigated. The K243RST246 sequence of LukS (Z-region; R241RTT244 for LukS-PV) near the LukS loop region is essential for its cytotoxic activity along with LukF (14). Recently, several receptor molecules for leukocidin S-components on target cells have been identified (15). For the S-component of PVL (LukS-PV), binding affinity to its leukocyte receptor C5a receptor (C5aR) was dramatically reduced by alanine substitution of residues T244, H245 and Y250 (16). Among them, T244 is correspond to Thr244 of LukS, and Y250 is located in the loop region. However, little is known about erythrocyte recognition and binding of S-components. Here, we constructed exchange mutants using the loops of the rim domains in Hlg2, LukE and LukS-PV, and investigated the role of each loop region in Hlg2 and LukE on haemolytic activity and erythrocyte binding. Materials and Methods Preparation of Hlg2, LukE, LukS-PV and their mutants The toxin component genes were amplified by polymerase chain reaction using PrimeSTAR MAX DNA Polymerase (Takara Bio, Kusatsu, Japan) and the genomic DNA of S. aureus Mu50 (Hlg2 and LukF) (17), Newman (LukE and LukD) (18), or ATCC49775 (LukS-PV) (19) strains as the template. Next, NcoI and XhoI-digested fragments were cloned into the same restriction sites of a pET26b-derived expression vector (4). Using expression plasmids for intact S-components as templates, site-directed mutagenesis for Hlg2, LukE and LukS-PV was performed using the PrimeSTAR Mutagenesis Basal kit (Takara Bio). Expression and purification of these mutant proteins was performed as described previously in (4). Briefly, Escherichia coli BL21(DE3) cells harbouring the expression plasmids were grown at 37°C in Luria-Bertani medium supplemented with 30 µg/ml kanamycin until the optical density at 600 nm reached 0.6–0.8. Next isopropyl-β-D-thiogalactopyranoside at a final concentration of 0.5 mM was added to induce the expression of desired proteins. After 24 h of cultivation at 25°C, the cells were collected by centrifugation and suspended in phosphate-buffered saline without calcium and magnesium (PBS (−)) for sonication. The N-terminal His6-tagged proteins present in the cell lysates were purified using a His-Trap HP column (GE Healthcare Biosciences AB, Uppsala Sweden), according to the manufacturer’s instructions. For all purified toxin components used in this study, circular dichroism (CD) analysis was performed using the J-720W1 spectropolarimeter (JASCO, Tokyo, Japan) to confirm their secondary structures. Quantification of proteins was performed using the BCA Protein Assay kit (Thermo Fisher Scientific Inc, Waltham, MA, USA). In silico analysis Multiple sequence alignment analysis for protein sequence of S-components was performed by ClustalW using the DDBJ server. Protein structure homology modelling was performed in the SWISS-MODEL workspace. Analysis of haemolytic activity and binding to human erythrocytes Haemolytic activity and binding of toxin components to human erythrocytes were assayed as described previously (3). Briefly, human erythrocytes were isolated from heparinized venous blood of healthy volunteers and washed with PBS (−). Each purified recombinant toxin was incubated with 108 human erythrocytes in PBS (−) at 37°C for different reaction times. Next, the samples were centrifuged at 8,000 × g for 5 min at 4°C, and haemoglobin concentration in the supernatant was measured using a spectrophotometer at 541 nm. Relative activity was calculated using hypotonic haemolysis with 5 mM phosphate buffer as 100%. For quantification of the membrane-binding components, the precipitate was washed with 5 mM Na-phosphate buffer, and the resulting membrane sample was analysed using western-blotting following sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) (3). To detect the pore complex, SDS-treatment was performed at 20°C (3). The F- and S-components were analysed using the anti-LukF and -Hlg2 anti sera, respectively, since these antisera will cross-react with same class components. Band intensity was quantified using the ImageJ software, and bound component levels were calculated in terms of relative intensity with respect to the intensity of the corresponding bound wild-type components (Hlg2, LukF or LukD) that blotted onto the same PVDF membrane. Results and Discussion Determination of S-component loops involved in binding to human erythrocytes According to the model of membrane pore formation by staphylococcal β-PFTs, toxin components bind to the target cell surface and assemble as a ring-shaped complex (1). The crystal structure of α- and γ-haemolysin membrane pore and the pre-pore indicated that the mushroom-shaped pore remained in contact with the target cell surface via its rim domain (4, 5, 10, 20), suggesting that the interaction between the rim domain of toxin components and the target cell surface was essential for the formation of the ring-shaped complex. In the LukF component, all amino acid residues important for erythrocyte binding were located at loops in the rim domain (e.g. Y72 in Loop-1; W177 and R198 in Loop-2; and F260 and Y261 in Loop-4) (1, 12). From the structure of the Hlg pore, four loop structures in the rim domain of Hlg2, designated as loops -1, -2, -3 and -4, were defined and the residues K64-Y67, N167, T183-A187 and R242-H243 were observed to be located at the tip of each loop structure, respectively, that extruded from the surface of the rim domain (Fig. 1A and B). Comparison of amino acid sequences among class S-components showed that the loop regions had low amino acid identity, indicating the involvement of these regions in cell surface recognition (Fig. 1A). Comparison with the LukF rim domain showed that Loop-1 corresponded to the loop containing Y72. Although Hlg2 lacked a phosphatidylcholine (PC)-binding domain, the location of N167 in Loop-2 was similar to that of W177 in the PC-binding site of LukF. F260 and Y261 of LukF were located in Loop-4. In addition, the Z-region (K243RST246 for LukS, RRTT for LukS-PV) was located just before Loop-4 of LukS, but was absent in Hlg2 and LukE. We presumed that the loop structures located at the rim domain of S-components were important for their target cell-specific membrane binding followed by pore formation, since monomeric structures of β-PFT components were highly similar to each other. Therefore, we focussed on these loops in the rim domain of S-components and analysed their contribution towards haemolytic activity. Fig. 1 View largeDownload slide Loop structures in S-component. (A) Alignment of S-components amino acid sequences. Loops 1–4 and Z-region in the rim domain and D-region in the cap domain are indicated. Positions of N167 in Hlg2 Loop-2 and N185 in LukE Loop-3 are underlined. (B) Side view of the Hlg2 protomer extracted from the octameric pore structure (pdb: 3B07). The cap, rim, and stem domains are indicated. Positions of loops 1–4 (L1–L4) and D- region (D) are depicted. (C) Bottom and side views of the LukE monomer (pdb: 3ROH). Prestem is indicated using an arrow. Fig. 1 View largeDownload slide Loop structures in S-component. (A) Alignment of S-components amino acid sequences. Loops 1–4 and Z-region in the rim domain and D-region in the cap domain are indicated. Positions of N167 in Hlg2 Loop-2 and N185 in LukE Loop-3 are underlined. (B) Side view of the Hlg2 protomer extracted from the octameric pore structure (pdb: 3B07). The cap, rim, and stem domains are indicated. Positions of loops 1–4 (L1–L4) and D- region (D) are depicted. (C) Bottom and side views of the LukE monomer (pdb: 3ROH). Prestem is indicated using an arrow. In addition to Hlg, only LukED showed haemolytic activity against human erythrocyte among leukocidins. To elucidate the effects of loop regions of Hlg2 and LukE on human erythrocytes haemolysis, we constructed a series of loop-replaced mutants of each component with corresponding amino acids of each loop of LukS-PV, which is the S-component of PVL that cannot lyse human erythrocytes (Table I). Considering the differences of the amino acid sequences between loops in Hlg2 and LukE, their exchange mutants were also constructed (Table I). Table I. Mutants of S-components used in this study S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  -, not exchanged Table I. Mutants of S-components used in this study S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  -, not exchanged The secondary structures of all recombinant S-components mutant proteins used in this study were assessed using CD spectrum analysis. Unfortunately, in some samples, marked changes in the secondary structure were observed, especially in Loop-3 mutants, and thus, these mutants were excluded from further analysis. Effects of Hlg2 component loop mutations replaced with amino acids of LukS-PV Under standard assay conditions (37°C, 10 min), 5 pmol Hlg2 completely lysed 108 human erythrocytes by combining with the same amount of LukF component. Substitution of Hlg2 Loop-4 with that of LukS-PV [Hlg2(L4)SPV] resulted in complete loss of haemolytic activity, even after the addition of 15 pmol of each component to the reaction mixture (Fig. 2A). Loop-1 substitution [Hlg2(L1)SPV] reduced the activity to approximately 60% of that of the wild type at 5 pmol, whereas only a slight decrease in the activity was observed by Loop-2 mutation [Hlg2(L2)SPV]. Double mutation in Loop-1 and -2 [Hlg2(L1, 2)SPV] markedly reduced the activity; however, haemolytic activity was still observed following the addition of 15 pmol of each component. In contrast, Loop-3 mutation [Hlg2(L3)SPV] did not affect the haemolytic activity at all (data not shown). Fig. 2 View largeDownload slide Effect of Hlg2 loop-replacement mutation with corresponding LukS-PV residue(s). (A) Haemolytic activity of Hlg2 loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated with varying amounts of loop-replaced Hlg2 mutants indicated in the figure, in the presence of 5 pmol LukF in 350 µl PBS(–), at 37°C. After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding ability of Hlg2 loop-replaced mutants. The amount of membrane-bound Hlg2 mutants on the erythrocyte membrane usingt 5 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (*P ≤ 0.05, **P ≤ 0.01). nd: not detected. Fig. 2 View largeDownload slide Effect of Hlg2 loop-replacement mutation with corresponding LukS-PV residue(s). (A) Haemolytic activity of Hlg2 loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated with varying amounts of loop-replaced Hlg2 mutants indicated in the figure, in the presence of 5 pmol LukF in 350 µl PBS(–), at 37°C. After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding ability of Hlg2 loop-replaced mutants. The amount of membrane-bound Hlg2 mutants on the erythrocyte membrane usingt 5 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (*P ≤ 0.05, **P ≤ 0.01). nd: not detected. The toxin components bound on the treated erythrocyte membrane at 5 pmol were analysed using western blotting. The haemolytic activity of these Hlg2 mutants was decreased in parallel with their binding amount on the erythrocytes (Fig. 2B), whereas a constant amount of LukF was detected on the erythrocyte membrane for all combinations of Hlg2 and its mutants. These results showed that the loop replacement mutation of Hlg2 lowered the haemolytic activity, due to the reduced binding ability of these mutants to human erythrocytes in the presence of LukF. Finally, the haemolytic activity and binding ability of Hlg2 were lost upon replacement of R242HR244 in Loop-4 with N248SY250 of LukS-PV [Hlg2(L4RHR)SPV] (Fig. 2), suggesting that the basic triad in Loop-4 is crucial for the binding of Hlg2 to cells in combination with LukF. A previous report indicated that T246 of LukS in the Z-region (K243RSTH246) was an essential residue required for the leukocytolytic activity (14). In addition, T244 and H245 in LukS-PV have been shown to be involved in binding to the leukocyte receptor C5aR1 and 2 (16), and these residues were located within the Z-region (RRTT244H245) (Fig. 1). To examine the effect of the Z-region adjoining Loop-4 of LukS and LukS-PV on the haemolytic activity of Hlg2, RRTTH was inserted at the corresponding site of Hlg2. The resultant mutant Hlg2insZ still showed 80% haemolysis at 5 pmol, but did not show 100% haemolysis even after the addition of an excess amount to the reaction mixture (Fig. 3A). In addition to T244 and H245 in the Z-region, alanine substitution of residues, R73, Y184 and Y250 of LukS-PV dramatically reduced its binding affinity for its leukocyte receptor C5aR1 and 2 (16). Among these, Y184 was located in Loop-3; Y250 in Loop-4; whereas R73 was present between the rim and cap domain near Loop-1. Structural analysis revealed that the residues Y184 and Y250 contributed towards LukS-PV binding by providing structural flexibility (16). As described earlier, substitution of R242HR244 in Hlg2 Loop-4 with N248SY250 of LukS-PV [Hlg2(L4RHR)SPV] abolished the haemolytic activity and erythrocyte-binding ability of Hlg2 (Fig. 2A). Although the haemolytic activity of Hlg2 reduced to 61.2% at a concentration of 15 pmol/108 cells by a single R244Y mutation, this value was similar to that of the R242N (58.6%) and H243S (64.0%) mutants, suggesting that Y250 of LukS-PV did not affect alone in the binding of Hlg2 to human erythrocytes. Thus, the Z-region may provide a hindrance for the haemolytic activity of Hlg2, although it is required for leukocytolysis (14) and possibly acts as an element for C5aR1- and 2-dependent cytotoxicity (16). Fig. 3 View largeDownload slide Analyses of Hlg2 mutation in D- and Z-regions. (A) Haemolytic activity of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of Hlg2 mutants, in which the D-region was replaced with the corresponding region of LukS-PV [Hlg2(D)SPV] and the Z -region was inserted just before Loop-4 (Hlg2insZ), in the presence of 5 pmol LukF in 350 µl of PBS (–). After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding and pore -forming ability of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with Hlg2 (15 pmol) and 5 pmol LukF in 350 µl PBS. After 60 min, haemolytic activity was measured, and the membrane pores formed were detected by western blot analysis using the anti-Hlg2 antiserum. Fig. 3 View largeDownload slide Analyses of Hlg2 mutation in D- and Z-regions. (A) Haemolytic activity of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of Hlg2 mutants, in which the D-region was replaced with the corresponding region of LukS-PV [Hlg2(D)SPV] and the Z -region was inserted just before Loop-4 (Hlg2insZ), in the presence of 5 pmol LukF in 350 µl of PBS (–). After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding and pore -forming ability of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with Hlg2 (15 pmol) and 5 pmol LukF in 350 µl PBS. After 60 min, haemolytic activity was measured, and the membrane pores formed were detected by western blot analysis using the anti-Hlg2 antiserum. Loops of Hlg2 confers haemolytic activity to LukS-PV Among Luk, PVL and LukED, their F- and S-components are interchangeable with each other, and LukS-PV forms membrane pores on the human leukocytes and shows leukocytolytic activity in combination with LukF (7). However, LukS-PV and LukF even at a high dose of 500 pmol/108 cells, do not show haemolytic activity (Fig. 4A), indicating that LukS-PV lacks the erythrocyte-binding potential. Therefore, the effect of substitution of LukS-PV loop regions with those of Hlg2 was investigated with respect to haemolytic activity against human erythrocytes. Fig. 4 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding Hlg2 residue(s). (A) Human erythrocytes (108 cells) were incubated at 37°C with 500 pmol of both the loop-replaced LukS-PV mutant and LukF (in a 1:1 ratio) in 350 µl PBS(−) for varying time intervals. After incubation, the amount of haemoglobin in the supernatants was measured. The Z -region was deleted in all Loop-4 mutants. Error bars represent standard deviations of triplicate measurements. The lower panel presents the enlarged view of the boxed section. (B) Effect of D- and Z -regions on triple mutants. Haemolysis by the LukS-PV triple mutant [SPV(L1, 2, 4)Hlg2], its Z -region deleted mutant [SPV(L1, 2, 4)Hlg2delZ], and further D-region-replaced mutant [SPV(D, L1, 2, 4)Hlg2delZ], was analysed as described above. Error bars represent standard deviations of triplicate measurements. Fig. 4 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding Hlg2 residue(s). (A) Human erythrocytes (108 cells) were incubated at 37°C with 500 pmol of both the loop-replaced LukS-PV mutant and LukF (in a 1:1 ratio) in 350 µl PBS(−) for varying time intervals. After incubation, the amount of haemoglobin in the supernatants was measured. The Z -region was deleted in all Loop-4 mutants. Error bars represent standard deviations of triplicate measurements. The lower panel presents the enlarged view of the boxed section. (B) Effect of D- and Z -regions on triple mutants. Haemolysis by the LukS-PV triple mutant [SPV(L1, 2, 4)Hlg2], its Z -region deleted mutant [SPV(L1, 2, 4)Hlg2delZ], and further D-region-replaced mutant [SPV(D, L1, 2, 4)Hlg2delZ], was analysed as described above. Error bars represent standard deviations of triplicate measurements. Since a higher amount of toxin component and long incubation time were required for the haemolysis by LukS-PV mutants in combination with LukF, we used a dose of 500 pmol/108 cells to compare the activities of the mutants. This dose was higher than that used for Hlg (5 pmol/108 cells). LOOP-4 mutation of LukS-PV with that of Hlg2 [SPV(L4)Hlg2] showed considerably less activity; however the mutant showed 40% haemolysis after deletion of the Z-region (RRTTH segment) [SPV(L4)Hlg2delZ] (Fig. 4A). Loop-1 mutant [SPV(L1)Hlg2] showed considerably less haemolysis, whereas Loop-2 mutant [SPV(L2)Hlg2] showed no activity. However, the haemolytic activity of SPV(L4)Hlg2delZ significantly increased by an additional mutation in Loop-1 [SPV(L1, 4)Hlg2delZ] or Loop-2 [SPV(L2, 4)Hlg2delZ], indicating that Loop-1 and -2 of Hlg2 play an auxiliary role in haemolysis. In contrast, activity of the Loop-1 and -2 double mutant [SPV(L1, 2)Hlg2] was similar to that of SPV(L1)Hlg2 (Fig. 4A). Similar to the Hlg2 mutant with an inserted Z-region from LukS-PV (Fig. 3, Hlg2insZ), Deletion of the Z-region from triple mutant SPV(L1, 2, 4)Hlg2 having the Z-region [SPV(L1, 2, 4)Hlg2delZ] resulted in 100% haemolysis after 60 min. Taken together, we suggest that while Loop-4 of Hlg2 potentially plays a role in determining erythrocyte specificity, loops-1 and -2 assist in binding. The D-region, K23RLAI27 segment in the cap domain, has been identified as a region with a pivotal role in haemolytic activity (21) (Fig. 1), and Hlg2 mutant where KRLAI was substituted with DKWGV of LukS-PV [Hlg2(D)SPV] showed reduced haemolytic activity (Fig. 3A). When 5 pmol Hlg2(D)SPV was incubated with 108 cells for 10 min, the amount of bound Hlg2(D)SPV was estimated to be only 40% of that of the wild-type Hlg2 using western blotting. Thus, the amount of Hlg2(D)SPV added to the standard reaction mixture was increased to 15 pmol, and after a 10-min incubation, the amount of bound Hlg2(D)SPV mutant increased to 60% of that of the wild-type Hlg2, and weak haemolysis was observed (Fig. 3A). Further incubation with 15 pmol Hlg2(D)SPV mutant showed 62.3% haemolysis after 60 min, and weaker pore-forming ability than the wild type was observed (Fig. 3B), suggesting D-region plays a role in the haemolytic activity via its pore-forming ability or stability. Indeed, further substitution of loop-D with KRLAI of Hlg2 in SPV(L1, 2, 4)Hlg2delZ [SPV(D, L1, 2, 4)Hlg2delZ] resulted in more than 90% haemolysis after a 10-min incubation (Fig. 4B). However, SPV(D, L1, 2, 4)Hlg2delZ had to be used at a high dose in combination with LukF (500 pmol/108 cells) to complete haemolysis within a short incubation time. Additional factors responsible for the strong haemolytic activity of Hlg2 at low dose remain to be investigated. Effect of loop-replacement mutations in LukE component with LukS-PV To gain insights into the erythrocyte recognition mechanism of LukE, the effects of loop replacement in LukE on its haemolytic activity were investigated in the same way as those investigated for Hlg2. In the presence of 100 pmol each of LukD and LukE components with 108 human erythrocytes, about 65% lysis was observed; however, this toxin dose used was 20-fold more than that used for Hlg (Fig. 5A). Loop-4 mutation of LukE resulted in complete loss of the activity with 100 pmol of both the mutant and LukD. Unlike Hlg2, Loop-1 and -2 mutations in LukE markedly decreased its haemolytic activity (Fig. 5A), and it was attributed to the loss of binding ability to human erythrocytes (Fig. 5B). These results suggested that Loop-4 of LukE was important for binding to human erythrocytes in combination with LukD. In contrast, the Loop-3 mutant [LukE(L3)SPV] showed slightly higher haemolytic activity than the wild-type LukE. Fig. 5 View largeDownload slide Effect of loop-replacement mutation between Hlg2 and LukE. Haemolytic activity (A) and binding ability (B) of LukE loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both loop-replaced LukE mutants and LukD (in a 1:1 ratio) in 350 µl PBS, as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured. In parallel, the amount of membrane-bound LukE mutants on the erythrocyte membrane using 100 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Haemolytic activity (C) and binding ability (D) of LukE loop-replaced mutants with Hlg2 loops were analysed in the same manner. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**; P ≤ 0.01). Fig. 5 View largeDownload slide Effect of loop-replacement mutation between Hlg2 and LukE. Haemolytic activity (A) and binding ability (B) of LukE loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both loop-replaced LukE mutants and LukD (in a 1:1 ratio) in 350 µl PBS, as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured. In parallel, the amount of membrane-bound LukE mutants on the erythrocyte membrane using 100 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Haemolytic activity (C) and binding ability (D) of LukE loop-replaced mutants with Hlg2 loops were analysed in the same manner. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**; P ≤ 0.01). To confirm the haemolytic activity of LukE rim domain loops, a series of LukS-PV mutants (similar to Hlg2 mutants constructed above) with loops replaced by corresponding regions from LukE were constructed, and their haemolytic activities were tested. Unlike LukS-PV mutants having Hlg2 Loop-4, LukS-PV mutants with LukE Loop-4 showed 60% haemolysis in combination with LukF, with or without the Z-region [SPV(L4)LukE and SPV(L4)LukEdelZ, respectively] (Fig. 6). In addition, further mutations in Loop-2 [SPV(L2, 4)LukE] and in loops-1 and -2 [SPV(L1, 2, 4)LukE] did not enhance the haemolytic activity of SPV(L4)LukE, suggesting that in combination with LukF, LukS-PV loops-1 and -2 behave similar to those of LukE. Fig. 6 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding LukE residue(s). Human erythrocytes (108 cells) were incubated at 37°C for 120 min with 500 pmol of both loop-replaced LukS-PV mutants and LukF or LukD (in a 1:1 ratio) in 350 µl PBS. The amount of haemoglobin in the supernatants was measured after 1-h incubation, and haemolytic activity combined with LukF or LukD was indicated by open and closed bar, respectively. Error bars represent the standard deviations of triplicate measurements. Fig. 6 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding LukE residue(s). Human erythrocytes (108 cells) were incubated at 37°C for 120 min with 500 pmol of both loop-replaced LukS-PV mutants and LukF or LukD (in a 1:1 ratio) in 350 µl PBS. The amount of haemoglobin in the supernatants was measured after 1-h incubation, and haemolytic activity combined with LukF or LukD was indicated by open and closed bar, respectively. Error bars represent the standard deviations of triplicate measurements. Effect of loop-replacement mutations between Hlg2 and LukE Loop-4 of both Hlg2 and LukE was suggested to play an important role in mediating haemolytic activity. Loop-1 and -2 of both components assisted in the binding of these components. However, the degree of contribution of these loops, which conferred haemolytic activity on LukS-PV, towards haemolysis was different for Hlg2 and LukE mutants. Thus, we investigated the effect of mutual loop substitution between LukE and Hlg2. Surprisingly, in combination with LukD, the haemolytic activity of LukE was abolished by Loop-4 replacement mutation with Hlg2 Loop-4 [LukE(L4)Hlg2] (Fig. 5C). The haemolytic activity of LukE also decreased following Loop-2 mutation [LukE(L2)Hlg2], whereas Loop-1, -3 or D-region substitutions with those of the corresponding Hlg2 regions [LukE(L1)Hlg2, LukE(L3)Hlg2 and LukE(D)Hlg2] slightly increased the activity (Fig. 5C). These changes correlated with the human erythrocyte-binding abilities of these mutants (Fig. 5D), indicating that Hlg2 Loop-4 did not function in the human erythrocyte-binding ability of LukE in combination with LukD. Effect of partner F-components for the haemolytic activities of mutant S-components For further investigating Loop-4 compatibility between LukE and Hlg2, a Hlg2 mutant with LukE Loop-4 [Hlg2(L4)LukE] was constructed. Unlike the LukE mutant with Hlg2 Loop-4 [LukE(L4)Hlg2], Hlg2(L4)LukE showed haemolytic activity in combination with LukD, but it required higher dose than intact Hlg2 (Fig. 7B). It is known that LukF enhances the binding of Hlg2 toward human erythrocytes (22). Therefore, we examined the synergistic actions of F-components on the haemolytic activities of the S-components mutants. Fig. 7 View largeDownload slide Combinatorial effect of S-component Loop-4 mutants and F-components on haemolysis. Haemolytic activity and binding ability of S- and F-components with LukF and with LukD. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both S- and F-components (in a 1:1 ratio) in 350 µl PBS(–), as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured (A and B). In parallel, the amount of membrane-bound S- and F-components on the erythrocyte membrane using 50 pmol of each component was quantified by western blot analysis using the anti-Hlg2 and anti-LukF antisera (C–F). The amount of LukE(L4)Hlg2 bound to the erythrocyte membrane at 100 pmol was quantified by western blot analysis using the anti-Hlg2 antiserum (G and H). Error bars represent the standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**P ≤ 0.01). Fig. 7 View largeDownload slide Combinatorial effect of S-component Loop-4 mutants and F-components on haemolysis. Haemolytic activity and binding ability of S- and F-components with LukF and with LukD. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both S- and F-components (in a 1:1 ratio) in 350 µl PBS(–), as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured (A and B). In parallel, the amount of membrane-bound S- and F-components on the erythrocyte membrane using 50 pmol of each component was quantified by western blot analysis using the anti-Hlg2 and anti-LukF antisera (C–F). The amount of LukE(L4)Hlg2 bound to the erythrocyte membrane at 100 pmol was quantified by western blot analysis using the anti-Hlg2 antiserum (G and H). Error bars represent the standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**P ≤ 0.01). Interestingly, the haemolytic activities of Loop-4 exchange mutants of Hlg2 and LukE differed in combination with the F-component. Hlg2 mutant having Loop-4 from LukE [Hlg2(L4)LukE] showed almost same haemolytic activity in combination with LukF (Fig. 7A), and it markedly increased from haemolytic activity in combination with LukD (Fig. 7B). Almost same amounts of LukF and LukD bound to human erythrocytes in the combination of any S-components (Fig. 7E and F), indicating that this reduction of haemolytic activity was not due to the binding of F-components. Meanwhile, compared with intact Hlg2, binding of mutant Hlg2(L4)LukE to human erythrocytes increased in combination with both LukD and LukF used at a concentration of 50 pmol/108 cells (Fig. 7C and D). These observations suggested that LukF enhanced the haemolytic activity of low doses of the Hlg2 Loop-4 mutant; however, the induction of haemolytic activity was not solely due to toxin binding. Moreover, LukF also induced haemolytic activity with LukE mutant with Hlg2 Loop-4 [LukE(L4)Hlg2] at 100 pmol, which did not show any activity with LukD (Fig. 7A and B), and the increase of haemolytic activity was according to the binding ability of LukE(L4)Hlg2 (Fig. 7G and H). This result suggested that the binding of the LukE(L4)Hlg2 having Hlg2 Loop-4 was stimulated by LukF, as observed in Hlg2 (22). Similarly, LukS-PV loop-replaced mutants with LukE segments showed markedly higher haemolytic activity in combination with LukF, than with LukD (Fig. 6). This ‘partner effect’ by LukF may be involved in enhancement of S-component binding toward human erythrocytes, and also mediating inter-molecular interactions between the S- and F-components during formation of ring-shaped complex. In conclusion, rim loops of both Hlg2 and LukE components, especially Loop-4, play an important role in erythrocyte binding, whereas their amino acid sequence were different each other (Fig. 1A and Table I). Our results provide novel insight into the different target receptor recognition mechanism of Hlg2 and LukE using a combination of multiple loops in the rim domain. It has been suggested that Hlg2 and LukE recognize the same receptor via different site of Duffy antigen receptor for chemokines (DARC), on human erythrocytes (9). Thus, our findings about incompatibility in the erythrocyte binding between the rim domain loops of Hlg2 and LukE may reflect the difference in their target regions. In fact, unlike Hlg2 Loop-4, LukE Loop-4 segment conferred haemolytic activity to LukS-PV without deletion of the Z-region, indicating that the interaction between LukE Loop-4 and erythrocyte receptor was not disturbed by the Z-region (Fig. 6). LukE and Hlg2 also share CXC chemokine receptor 1 (CXCR1) and CXCR 2 on neutrophils (23–25) as their leukocyte receptors, and LukE Loop-3-containing region was required for targeting this receptor, whereas the Loop-4-containing region was involved in cell killing (24). The different behaviour of LukE Loop-4 towards CXCRs and DARC may be explained by the difference in the interaction between LukE and its target molecules. Furthermore, LukE uses the chemokine receptor C-C chemokine receptor type 5(CCR5) on T lymphocytes (23, 26). Tam et al suggested that the amino acid segment K64GSGYE69 in the rim domain, which corresponded to Loop-1 in LukE, was required for CCR5 targeting and cytotoxicity using an in-frame deletion mutant (27). Our preliminary experiments against hCCR5/ Chinese hamster ovary (CHO)-K1 cells (28) showed that Loop-4 replacement with Hlg2 segment [LukE(L4)Hlg2] reduced cytotoxic activity of LukE in combination with LukD, (Data not shown), suggesting that Loop-4 of LukE may also play a role in hCCR5-mediated cytotoxicity. It is difficult to understand how Hlg2 and LukE components having different rim domain amino acid sequences share several different receptor molecules, including DARC on their target cells. For understanding the detailed mechanism underlying recognition of different receptors, further structural studies on interaction between S-component loops and extracellular loops of receptors are required. Acknowledgements We would like to thank Editage (www.editage.jp) for English language editing. Funding This work was supported by JSPS Grant-in-Aid for Challenging Exploratory Research (KAKENHI) Grant Number JP16K14897 (to J. K). Conflict of Interest None declared. References 1 Kaneko J., Kamio Y. ( 2004) Bacterial two-component and hetero-heptameric pore-forming cytolytic toxins: structures, pore-forming mechanism, and organization of the genes. Biosci. Biotechnol. Biochem . 68, 981– 1003 Google Scholar CrossRef Search ADS PubMed  2 Kamio Y., Rahman A., Nariya H., Ozawa T., Izaki K. ( 1993) The two Staphylococcal bi-component toxins, leukocidin and gamma-hemolysin, share one component in common. FEBS Lett . 321, 15– 18 Google Scholar CrossRef Search ADS PubMed  3 Kaneko J., Ozawa T., Tomita T., Kamio Y. ( 1997) Sequential binding of Staphylococcal gamma-hemolysin to human erythrocytes and complex formation of the hemolysin on the cell surface. Biosci. Biotechnol. Biochem . 61, 846– 851 Google Scholar CrossRef Search ADS PubMed  4 Yamashita K., Kawai Y., Tanaka Y., Hirano N., Kaneko J., Tomita N., Ohta M., Kamio Y., Yao M., Tanaka I. ( 2011) Crystal structure of the octameric pore of staphylococcal γ-hemolysin reveals the β-barrel pore formation mechanism by two components. Proc. Nat.l Acad. Sci. U. S. A . 108, 17314– 17319 Google Scholar CrossRef Search ADS   5 Yamashita D., Sugawara T., Takeshita M., Kaneko J., Kamio Y., Tanaka I., Tanaka Y., Yao M. ( 2014) Molecular basis of transmembrane beta-barrel formation of staphylococcal pore-forming toxins. Nat. Commun.  5, 4897 Google Scholar CrossRef Search ADS PubMed  6 Prévost G., Cribier B., Couppié P., Petiau P., Supersac G., Finck-Barbançon V., Monteil H., Piemont Y. ( 1995) Panton-Valentine leucocidin and gamma-hemolysin from Staphylococcus aureus ATCC 49775 are encoded by distinct genetic loci and have different biological activities. Infect. Immun . 3, 4121– 4129 7 Gravet A., Colin D.A., Keller D., Giradot R., Monteil H., Prévost G. ( 1998) Characterization of a novel structural member, LukE-LukD, of the bi-component staphylococcal leucotoxins family. FEBS Lett . 436, 202– 208 Google Scholar CrossRef Search ADS PubMed  8 Morinaga N., Kaihou Y., Noda M. ( 2003) Purification, cloning and characterization of variant LukE-LukD with strong leukocidal activity of staphylococcal bi-component leukotoxin family. Microbiol. Immunol . 47, 81– 90 Google Scholar CrossRef Search ADS PubMed  9 Spaan A.N., Reyes-Robles T., Badiou C., Cochet S., Boguslawski K.M., Yoong P., Day C.J., de Haas C.C.J., van Kessel K.P.M., Vandenesch F., Jennings M.P., Le Van Kim C., Colin Y., van Strijp J.A., Henry T., Torres V.J. ( 2015) Staphylococcus aureus Targets the Duffy Antigen Receptor for Chemokines (DARC) to Lyse Erythrocytes. Cell Host Microbe  18, 363– 370 Google Scholar CrossRef Search ADS PubMed  10 Song L., Hobaugh M.R., Shustak C., Cheley S., Bayley H., Gouaux J.E. ( 1996) Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science  274, 1859– 1866 Google Scholar CrossRef Search ADS PubMed  11 Olson R., Nariya H., Yokota K., Kamio Y., Gouaux J.E. ( 1999) Crystal structure of staphylococcal LukF delineates conformational changes accompanying formation of a transmembrane channel. Nat. Struct. Biol . 6, 134– 140 Google Scholar CrossRef Search ADS PubMed  12 Yokota K., Kamio Y. ( 2000) Tyrosine72 residue at the bottom of rim domain in LukF crucial for the sequential binding of the staphylococcal gamma-hemolysin to human erythrocytes. Biosci. Biotechnol. Biochem . 64, 2744– 2747 Google Scholar CrossRef Search ADS PubMed  13 Monma N., Nguyen V.T., Kaneko J., Higuchi H., Kamio Y. ( 2004) Essential residues, W177 and R198, of LukF for phosphatidylcholine-binding and pore-formation by staphylococcal gamma-hemolysin on human erythrocyte membranes. J. Biochem . 136, 427– 431 Google Scholar CrossRef Search ADS PubMed  14 Nishiyama A., Nariya H., Kamio Y. ( 1998) Phosphorylation of LukS by protein kinase A is crucial for the LukS-specific function of the staphylococcal leukocidin on human polymorphonuclear leukocytes. Biosci. Biotechnol. Biochem . 62, 1834– 1838 Google Scholar CrossRef Search ADS PubMed  15 Spaan A.N., Henry T., van Rooijen W.J., Perret M., Badiou C., Aerts P.C., Kemmink J., de Haas C.J., van Kessel K.P., Vandenesch F., Lina G., van Strijp J.A. ( 2013) The staphylococcal toxin Panton-Valentine Leukocidin targets human C5a receptors. Cell Host Microbe  13, 584– 594 Google Scholar CrossRef Search ADS PubMed  16 Laventie B.J., Guérin F., Mourey L., Tawk M.Y., Jover E., Maveyraud L., Prévost G. ( 2014) Residues essential for Panton-Valentine leukocidin S component binding to its cell receptor suggest both plasticity and adaptability in its interaction surface. PLoS One  9, e92094 Google Scholar CrossRef Search ADS PubMed  17 Kuroda M., Ohta T., Uchiyama I., Baba T., Yuzawa H., Kobayashi I., Cui L., Oguchi A., Aoki K., Nagai Y., Lian J., Ito T., Kanamori M., Matsumaru H., Maruyama A., Murakami H., Hosoyama A., Mizutani-Ui Y., Takahashi N.K., Sawano T., Inoue R., Kaito C., Sekimizu K., Hirakawa H., Kuhara S., Goto S., Yabuzaki J., Kanehisa M., Yamashita A., Oshima K., Furuya K., Yoshino C., Shiba T., Hattori M., Ogasawara N., Hayashi H., Hiramatsu K. ( 2001) Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet  357, 1225– 1240 Google Scholar CrossRef Search ADS PubMed  18 Baba T., Bae T., Schneewind O., Takeuchi F., Hiramatsu K. ( 2008) Genome sequence of Staphylococcus aureus strain Newman and comparative analysis of staphylococcal genomes: polymorphism and evolution of two major pathogenicity islands. J. Bacteriol . 190, 300– 310 Google Scholar CrossRef Search ADS PubMed  19 Kaneko J., Kimura T., Kawakami Y., Tomita T., Kamio Y. ( 1997) Panton-valentine leukocidin genes in a phage-like particle isolated from mitomycin C-treated Staphylococcus aureus V8 (ATCC 49775). Biosci. Biotechnol. Biochem . 61, 1960– 1962 Google Scholar CrossRef Search ADS PubMed  20 Sugawara T., Yamashita D., Kato K., Peng Z., Ueda J., Kaneko J., Kamio Y., Tanaka Y., Yao M. ( 2015) Structural basis for pore-forming mechanism of α-hemolysin. Toxicon  108, 226– 231 Google Scholar CrossRef Search ADS PubMed  21 Nariya H., Kamio Y. ( 1997) Identification of the minimum segment essential for the H gamma II-specific function of staphylococcal gamma-hemolysin. Biosci. Biotechnol. Biochem.  61, 1786– 1788 Google Scholar CrossRef Search ADS PubMed  22 Nguyen V.T., Kamio Y., Higuchi H. ( 2003) Single-molecule imaging of cooperative assembly of gamma-hemolysin on erythrocyte membranes. embo J . 22, 4968–4679 23 Spaan A.N., van Strijp J.A.G., Torres V.J. ( 2017) Leukocidins: staphylococcal bi-component pore-forming toxins find their receptors. Nat. Rev. Micro.  15, 435– 447 Google Scholar CrossRef Search ADS   24 Reyes-Robles T., Alonzo F.3rd., Kozhaya L., Lacy D.B., Unutmaz D., Torres V.J. ( 2013) Staphylococcus aureus leukotoxin ED targets the chemokine receptors CXCR1 and CXCR2 to kill leukocytes and promote infection. Cell Host Microbe  14, 453– 459 Google Scholar CrossRef Search ADS PubMed  25 Spaan A.N., Vrieling M., Wallet P., Badiou C., Reyes-Robles T., Ohneck E.A., Benito Y., de Haas C.J., Day C.J., Jennings M.P., Lina G., Vandenesch F., van Kessel K.P., Torres V.J., van Strijp J.A., Henry T. ( 2014) The staphylococcal toxins γ-haemolysin AB and CB differentially target phagocytes by employing specific chemokine receptors. Nat. Commun.  5, 5438 Google Scholar CrossRef Search ADS PubMed  26 Alonzo F.3rd., Kozhaya L., Rawlings S.A., Reyes-Robles T., DuMont A.L., Myszka D.G., Landau N.R., Unutmaz D., Torres V.J. ( 2012) CCR5 is a receptor for Staphylococcus aureus leukotoxin ED. Nature  493, 51– 55 Google Scholar CrossRef Search ADS PubMed  27 Tam K., Schultz M., Reyes-Robles T., Vanwalscappel B., Horton J., Alonzo F.3rd., Wu B., Landau N.R., Torres V.J. ( 2016) Staphylococcus aureus Leukocidin LukED and HIV-1 gp120 Target Different Sequence Determinants on CCR5. MBiol  7, e02024-16 pii Google Scholar CrossRef Search ADS   28 Maeda K., Yoshimura K., Shibayama S., Habashita H., Tada H., Sagawa K., Miyakawa T., Aoki M., Fukushima D., Mitsuya H. ( 2001) Novel low molecular weight spirodiketopiperazine derivatives potently inhibit R5 HIV-1 infection through their antagonistic effects on CCR5. J. Biol. Chem.  276, 35194– 35200 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations C5aR C5a receptor CCR5 C-C chemokine receptor type 5 CD circular dichroism CHO Chinese hamster ovary CXCR CXC chemokine receptor DARC Duffy antigen receptor for chemokines Hla α-haemolysin Hlg γ-Haemolysin Hlg2 (alias HlgA) S-component of Hlg Luk leukocidin LukD F-component of LukED LukE S-component of LukED LukF (alias HlgB) F-component of Luk LukS S-component of Luk LukS-PV S-component of PVL PBS(−) phosphate-buffered saline without calcium and magnesium PC phosphatidylcholine PFT pore-forming toxin PVL Panton-Valentine leukocidin SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Biochemistry Oxford University Press

Rim domain loops of staphylococcal β-pore forming bi-component toxin S-components recognize target human erythrocytes in a coordinated manner

Loading next page...
 
/lp/ou_press/rim-domain-loops-of-staphylococcal-pore-forming-bi-component-toxin-s-P0EYOISjui
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved
ISSN
0021-924X
eISSN
1756-2651
D.O.I.
10.1093/jb/mvy030
Publisher site
See Article on Publisher Site

Abstract

Abstract Staphylococcus aureus bi-component pore-forming toxins consist of S- and F-components, and form hetero-octameric beta-barrel pores on target blood cell membranes. Among them, γ-haemolysin (Hlg2 and F-component of Luk (LukF)) and LukED (LukE and LukD) possess haemolytic activity, whereas the Panton-Valentine leukocidin (LukS-PV and LukF-PV) does not lyse human erythrocytes. Here, we focussed on four loop structures in the rim domain of S-component, namely loops -1, -2, -3 and -4, and found that replacement of Loop-4 in both Hlg2 and LukE with that of LukS-PV abolished their haemolytic activity. Furthermore, LukS-PV gained haemolytic activity by Loop-4 exchange with Hlg2 or LukE, suggesting that Loop-4 of these S-components determined erythrocyte specificity. LOOP-1 and -2 enhanced the erythrocytes-binding ability of both components. Although Hlg2 and LukE recognize Duffy antigen receptor for chemokines on human erythrocytes, the ability of Loop-4 was not complementary between Hlg2 and LukE. Exchange of Hlg2 with LukE Loop-4 showed weaker activity than intact Hlg2, and LukE mutant with Hlg2 Loop-4 lost its haemolytic activity in combination of LukD. Interestingly, the haemolytic activities of these Loop-4 exchange mutants were affected by F-component, namely LukF enhanced haemolytic activities of these Hlg2 and LukE Loop-4 mutants, and also haemolytic activity of LukS-PV mutant with LukE Loop-4. protein, bacterial, protein-protein interactions, toxins, drugs, xenobiotics, staphylococcal bi-component toxin, S-components, CCR5 Staphylococcus aureus produces several types of leukocidin family bi-component toxins belonging to the β-barrel pore-forming toxins (PFTs). Leukocidin family toxins comprise the F- and S-component proteins, and form pores containing both components in a 1:1 ratio on the surface membrane of target white blood cells (1). Almost all isolates of S. aureus possess the leukocidin (Luk) cluster consisting of genes encoding LukF (F-component) and LukS (S-component), alternatively designated as HlgB and HlgC, respectively, on their chromosome. Hlg2 (also called HlgA), another S-component with its gene located just upstream of lukS, shows strong haemolytic activity against human erythrocytes in combination with LukF. Therefore, LukF and Hlg2 are called γ-haemolysin (Hlg). At an extremely low dose of 104 molecules of each component per cell, Hlg lyses human erythrocytes within 10 min at 37°C, whereas Luk (with LukS) does not show haemolytic activity under the same conditions, suggesting that the S-component plays an important role in host cell specificity (1–3). F- and S-components of Hlg/Luk have similar structures, and form hetero-octameric beta-barrel pores with the F- and S-components being arranged alternately on the target cell membrane (4, 5). To date, a number of staphylococcal leukocidin variants, closely related to Hlg/Luk, such as Panton-Valentine leukocidin (PVL LukF-PV and LukS-PV) (6) and LukED (LukD and LukE) (7, 8), have been identified. Among them, LukED has been reported to possess haemolytic activity against human erythrocytes (8, 9), whereas PVL specifically acts on human leukocytes, but has no haemolytic activity against human and rabbit erythrocytes (6). Components of bi-component toxins are secreted by the bacterium as soluble proteins, and form insoluble membrane pore on the target cell surface. The pore-forming event by toxins begins with the binding of soluble protein components to the target cell. Alpha-haemolysin (Hla), a staphylococcal β-PFT forming homo-heptamers, has a phosphatidylcholine (PC)-binding site in the rim domain that is involved in interacting with the erythrocyte surface (10). The monomeric structure of LukF is similar to that of Hla in the cap and rim domains, and its PC-binding site comprising of W177, E192 and R198, which directly interact with the choline head of PC or sphingomyelin in the plasma membrane, is confirmed to be present in the rim domain (11–13). In addition, three aromatic residues, namely Y72, F260 and Y261, which are important for binding to erythrocytes, are located in the loop structure at the bottom of the rim domain of LukF (1, 12). These observations suggest that the loop structures in the rim domain of Hla and LukF play an important role in cell recognition and cell binding. On the other hand, details of the mechanism involving binding of the S-component to target cells remain to be investigated. The K243RST246 sequence of LukS (Z-region; R241RTT244 for LukS-PV) near the LukS loop region is essential for its cytotoxic activity along with LukF (14). Recently, several receptor molecules for leukocidin S-components on target cells have been identified (15). For the S-component of PVL (LukS-PV), binding affinity to its leukocyte receptor C5a receptor (C5aR) was dramatically reduced by alanine substitution of residues T244, H245 and Y250 (16). Among them, T244 is correspond to Thr244 of LukS, and Y250 is located in the loop region. However, little is known about erythrocyte recognition and binding of S-components. Here, we constructed exchange mutants using the loops of the rim domains in Hlg2, LukE and LukS-PV, and investigated the role of each loop region in Hlg2 and LukE on haemolytic activity and erythrocyte binding. Materials and Methods Preparation of Hlg2, LukE, LukS-PV and their mutants The toxin component genes were amplified by polymerase chain reaction using PrimeSTAR MAX DNA Polymerase (Takara Bio, Kusatsu, Japan) and the genomic DNA of S. aureus Mu50 (Hlg2 and LukF) (17), Newman (LukE and LukD) (18), or ATCC49775 (LukS-PV) (19) strains as the template. Next, NcoI and XhoI-digested fragments were cloned into the same restriction sites of a pET26b-derived expression vector (4). Using expression plasmids for intact S-components as templates, site-directed mutagenesis for Hlg2, LukE and LukS-PV was performed using the PrimeSTAR Mutagenesis Basal kit (Takara Bio). Expression and purification of these mutant proteins was performed as described previously in (4). Briefly, Escherichia coli BL21(DE3) cells harbouring the expression plasmids were grown at 37°C in Luria-Bertani medium supplemented with 30 µg/ml kanamycin until the optical density at 600 nm reached 0.6–0.8. Next isopropyl-β-D-thiogalactopyranoside at a final concentration of 0.5 mM was added to induce the expression of desired proteins. After 24 h of cultivation at 25°C, the cells were collected by centrifugation and suspended in phosphate-buffered saline without calcium and magnesium (PBS (−)) for sonication. The N-terminal His6-tagged proteins present in the cell lysates were purified using a His-Trap HP column (GE Healthcare Biosciences AB, Uppsala Sweden), according to the manufacturer’s instructions. For all purified toxin components used in this study, circular dichroism (CD) analysis was performed using the J-720W1 spectropolarimeter (JASCO, Tokyo, Japan) to confirm their secondary structures. Quantification of proteins was performed using the BCA Protein Assay kit (Thermo Fisher Scientific Inc, Waltham, MA, USA). In silico analysis Multiple sequence alignment analysis for protein sequence of S-components was performed by ClustalW using the DDBJ server. Protein structure homology modelling was performed in the SWISS-MODEL workspace. Analysis of haemolytic activity and binding to human erythrocytes Haemolytic activity and binding of toxin components to human erythrocytes were assayed as described previously (3). Briefly, human erythrocytes were isolated from heparinized venous blood of healthy volunteers and washed with PBS (−). Each purified recombinant toxin was incubated with 108 human erythrocytes in PBS (−) at 37°C for different reaction times. Next, the samples were centrifuged at 8,000 × g for 5 min at 4°C, and haemoglobin concentration in the supernatant was measured using a spectrophotometer at 541 nm. Relative activity was calculated using hypotonic haemolysis with 5 mM phosphate buffer as 100%. For quantification of the membrane-binding components, the precipitate was washed with 5 mM Na-phosphate buffer, and the resulting membrane sample was analysed using western-blotting following sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) (3). To detect the pore complex, SDS-treatment was performed at 20°C (3). The F- and S-components were analysed using the anti-LukF and -Hlg2 anti sera, respectively, since these antisera will cross-react with same class components. Band intensity was quantified using the ImageJ software, and bound component levels were calculated in terms of relative intensity with respect to the intensity of the corresponding bound wild-type components (Hlg2, LukF or LukD) that blotted onto the same PVDF membrane. Results and Discussion Determination of S-component loops involved in binding to human erythrocytes According to the model of membrane pore formation by staphylococcal β-PFTs, toxin components bind to the target cell surface and assemble as a ring-shaped complex (1). The crystal structure of α- and γ-haemolysin membrane pore and the pre-pore indicated that the mushroom-shaped pore remained in contact with the target cell surface via its rim domain (4, 5, 10, 20), suggesting that the interaction between the rim domain of toxin components and the target cell surface was essential for the formation of the ring-shaped complex. In the LukF component, all amino acid residues important for erythrocyte binding were located at loops in the rim domain (e.g. Y72 in Loop-1; W177 and R198 in Loop-2; and F260 and Y261 in Loop-4) (1, 12). From the structure of the Hlg pore, four loop structures in the rim domain of Hlg2, designated as loops -1, -2, -3 and -4, were defined and the residues K64-Y67, N167, T183-A187 and R242-H243 were observed to be located at the tip of each loop structure, respectively, that extruded from the surface of the rim domain (Fig. 1A and B). Comparison of amino acid sequences among class S-components showed that the loop regions had low amino acid identity, indicating the involvement of these regions in cell surface recognition (Fig. 1A). Comparison with the LukF rim domain showed that Loop-1 corresponded to the loop containing Y72. Although Hlg2 lacked a phosphatidylcholine (PC)-binding domain, the location of N167 in Loop-2 was similar to that of W177 in the PC-binding site of LukF. F260 and Y261 of LukF were located in Loop-4. In addition, the Z-region (K243RST246 for LukS, RRTT for LukS-PV) was located just before Loop-4 of LukS, but was absent in Hlg2 and LukE. We presumed that the loop structures located at the rim domain of S-components were important for their target cell-specific membrane binding followed by pore formation, since monomeric structures of β-PFT components were highly similar to each other. Therefore, we focussed on these loops in the rim domain of S-components and analysed their contribution towards haemolytic activity. Fig. 1 View largeDownload slide Loop structures in S-component. (A) Alignment of S-components amino acid sequences. Loops 1–4 and Z-region in the rim domain and D-region in the cap domain are indicated. Positions of N167 in Hlg2 Loop-2 and N185 in LukE Loop-3 are underlined. (B) Side view of the Hlg2 protomer extracted from the octameric pore structure (pdb: 3B07). The cap, rim, and stem domains are indicated. Positions of loops 1–4 (L1–L4) and D- region (D) are depicted. (C) Bottom and side views of the LukE monomer (pdb: 3ROH). Prestem is indicated using an arrow. Fig. 1 View largeDownload slide Loop structures in S-component. (A) Alignment of S-components amino acid sequences. Loops 1–4 and Z-region in the rim domain and D-region in the cap domain are indicated. Positions of N167 in Hlg2 Loop-2 and N185 in LukE Loop-3 are underlined. (B) Side view of the Hlg2 protomer extracted from the octameric pore structure (pdb: 3B07). The cap, rim, and stem domains are indicated. Positions of loops 1–4 (L1–L4) and D- region (D) are depicted. (C) Bottom and side views of the LukE monomer (pdb: 3ROH). Prestem is indicated using an arrow. In addition to Hlg, only LukED showed haemolytic activity against human erythrocyte among leukocidins. To elucidate the effects of loop regions of Hlg2 and LukE on human erythrocytes haemolysis, we constructed a series of loop-replaced mutants of each component with corresponding amino acids of each loop of LukS-PV, which is the S-component of PVL that cannot lyse human erythrocytes (Table I). Considering the differences of the amino acid sequences between loops in Hlg2 and LukE, their exchange mutants were also constructed (Table I). Table I. Mutants of S-components used in this study S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  -, not exchanged Table I. Mutants of S-components used in this study S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  S-components  Name  D-region  Loop-1  Loop-2  Loop-3  Z-region  Loop-4  Hlg2  wild type  K23RLAI27  K64YPY67  N167  T183GPAA187  no  V240TRHR244  Hlg2(D)SPV  DKWGV  -  -  -  no  -  Hlg2(L1)SPV  -  NTDH  -  -  no  -  Hlg2(L2)SPV  -  -  N167L  -  no  -  Hlg2(L3)SPV  -  -  -  YSQNP  no  -  Hlg2(L4)SPV  -  -  -  -  no  YGNSY  Hlg2(L1, 2)SPV  -  NTDH  N167L  -  no  -  Hlg2(L4RHR)SPV  -  -  -  -  no  VTNSY  Hlg2insZ  -  -  -  -  RRTTH  -  Hlg2(L4)LukE  -  -  -  -  no  FPRTG  LukE  wild type  K23KWGV27  G64SDYEL69  D169  N185GPTGSA191  no  F245PRTG249  LukE(L1)SPV  -  NTDH  -  -  no  -  LukE(L2)SPV  -  -  D169L  -  no  -  LukE(L3)SPV  -  -  -  N185P  no    LukE(L4)SPV  -  -  -  -  no  YGNSY  LukE(L1)Hlg2  -  KYPY  -  -  no  -  LukE(L2)Hlg2  -  -  D169N  -  no  -  LukE(L3)Hlg2  -  -  -  N185T  no  -  LukE(L4)Hlg2  -  -  -  -  no  VTRHR  LukE(D)Hlg2  KRLAI  -  -  -  No  -  LukS-PV (SPV)  wild type  D23KWGV27  N65TDH68  L167  P183YSQNP188  R241RTTH245  Y246GNSY250  SPV(L1)Hlg2  -  KYPY  -  -  -  -  SPV(L2)Hlg2  -  -  L167N  -  -  -  SPV(L4)Hlg2  -  -  -  -  -  VTRHR  SPV(L4)Hlg2delZ  -  -  -  -  deleted  VTRHR  SPV(L1, 2)Hlg2  -  KYPY  L167N  -  -  -  SPV(L1, 4)Hlg2delZ  -  KYPY  -  -  deleted  VTRHR  SPV(L2, 4)Hlg2delZ  -  -  L167N  -  deleted  VTRHR  SPV(L1, 2, 4)Hlg2  -  KYPY  L167N  -  -  VTRHR  SPV(L1, 2, 4)Hlg2delZ  -  KYPY  L167N  -  deleted  VTRHR  SPV(D, L1, 2, 4)Hlg2delZ  KRLAI  KYPY  L167N  -  deleted  VTRHR  SPV(L4)LukE  -  -  -  -  -  FPRTG  SPV(L4)LukEdelZ  -  -  -  -  deleted  FPRTG  SPV(L2, 4)LukE  -  -  L167D  -  -  FPRTG  SPV(L1, 2, 4)LukE  -  GSDYEL  L167D  -  -  FPRTG  -, not exchanged The secondary structures of all recombinant S-components mutant proteins used in this study were assessed using CD spectrum analysis. Unfortunately, in some samples, marked changes in the secondary structure were observed, especially in Loop-3 mutants, and thus, these mutants were excluded from further analysis. Effects of Hlg2 component loop mutations replaced with amino acids of LukS-PV Under standard assay conditions (37°C, 10 min), 5 pmol Hlg2 completely lysed 108 human erythrocytes by combining with the same amount of LukF component. Substitution of Hlg2 Loop-4 with that of LukS-PV [Hlg2(L4)SPV] resulted in complete loss of haemolytic activity, even after the addition of 15 pmol of each component to the reaction mixture (Fig. 2A). Loop-1 substitution [Hlg2(L1)SPV] reduced the activity to approximately 60% of that of the wild type at 5 pmol, whereas only a slight decrease in the activity was observed by Loop-2 mutation [Hlg2(L2)SPV]. Double mutation in Loop-1 and -2 [Hlg2(L1, 2)SPV] markedly reduced the activity; however, haemolytic activity was still observed following the addition of 15 pmol of each component. In contrast, Loop-3 mutation [Hlg2(L3)SPV] did not affect the haemolytic activity at all (data not shown). Fig. 2 View largeDownload slide Effect of Hlg2 loop-replacement mutation with corresponding LukS-PV residue(s). (A) Haemolytic activity of Hlg2 loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated with varying amounts of loop-replaced Hlg2 mutants indicated in the figure, in the presence of 5 pmol LukF in 350 µl PBS(–), at 37°C. After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding ability of Hlg2 loop-replaced mutants. The amount of membrane-bound Hlg2 mutants on the erythrocyte membrane usingt 5 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (*P ≤ 0.05, **P ≤ 0.01). nd: not detected. Fig. 2 View largeDownload slide Effect of Hlg2 loop-replacement mutation with corresponding LukS-PV residue(s). (A) Haemolytic activity of Hlg2 loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated with varying amounts of loop-replaced Hlg2 mutants indicated in the figure, in the presence of 5 pmol LukF in 350 µl PBS(–), at 37°C. After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding ability of Hlg2 loop-replaced mutants. The amount of membrane-bound Hlg2 mutants on the erythrocyte membrane usingt 5 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (*P ≤ 0.05, **P ≤ 0.01). nd: not detected. The toxin components bound on the treated erythrocyte membrane at 5 pmol were analysed using western blotting. The haemolytic activity of these Hlg2 mutants was decreased in parallel with their binding amount on the erythrocytes (Fig. 2B), whereas a constant amount of LukF was detected on the erythrocyte membrane for all combinations of Hlg2 and its mutants. These results showed that the loop replacement mutation of Hlg2 lowered the haemolytic activity, due to the reduced binding ability of these mutants to human erythrocytes in the presence of LukF. Finally, the haemolytic activity and binding ability of Hlg2 were lost upon replacement of R242HR244 in Loop-4 with N248SY250 of LukS-PV [Hlg2(L4RHR)SPV] (Fig. 2), suggesting that the basic triad in Loop-4 is crucial for the binding of Hlg2 to cells in combination with LukF. A previous report indicated that T246 of LukS in the Z-region (K243RSTH246) was an essential residue required for the leukocytolytic activity (14). In addition, T244 and H245 in LukS-PV have been shown to be involved in binding to the leukocyte receptor C5aR1 and 2 (16), and these residues were located within the Z-region (RRTT244H245) (Fig. 1). To examine the effect of the Z-region adjoining Loop-4 of LukS and LukS-PV on the haemolytic activity of Hlg2, RRTTH was inserted at the corresponding site of Hlg2. The resultant mutant Hlg2insZ still showed 80% haemolysis at 5 pmol, but did not show 100% haemolysis even after the addition of an excess amount to the reaction mixture (Fig. 3A). In addition to T244 and H245 in the Z-region, alanine substitution of residues, R73, Y184 and Y250 of LukS-PV dramatically reduced its binding affinity for its leukocyte receptor C5aR1 and 2 (16). Among these, Y184 was located in Loop-3; Y250 in Loop-4; whereas R73 was present between the rim and cap domain near Loop-1. Structural analysis revealed that the residues Y184 and Y250 contributed towards LukS-PV binding by providing structural flexibility (16). As described earlier, substitution of R242HR244 in Hlg2 Loop-4 with N248SY250 of LukS-PV [Hlg2(L4RHR)SPV] abolished the haemolytic activity and erythrocyte-binding ability of Hlg2 (Fig. 2A). Although the haemolytic activity of Hlg2 reduced to 61.2% at a concentration of 15 pmol/108 cells by a single R244Y mutation, this value was similar to that of the R242N (58.6%) and H243S (64.0%) mutants, suggesting that Y250 of LukS-PV did not affect alone in the binding of Hlg2 to human erythrocytes. Thus, the Z-region may provide a hindrance for the haemolytic activity of Hlg2, although it is required for leukocytolysis (14) and possibly acts as an element for C5aR1- and 2-dependent cytotoxicity (16). Fig. 3 View largeDownload slide Analyses of Hlg2 mutation in D- and Z-regions. (A) Haemolytic activity of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of Hlg2 mutants, in which the D-region was replaced with the corresponding region of LukS-PV [Hlg2(D)SPV] and the Z -region was inserted just before Loop-4 (Hlg2insZ), in the presence of 5 pmol LukF in 350 µl of PBS (–). After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding and pore -forming ability of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with Hlg2 (15 pmol) and 5 pmol LukF in 350 µl PBS. After 60 min, haemolytic activity was measured, and the membrane pores formed were detected by western blot analysis using the anti-Hlg2 antiserum. Fig. 3 View largeDownload slide Analyses of Hlg2 mutation in D- and Z-regions. (A) Haemolytic activity of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of Hlg2 mutants, in which the D-region was replaced with the corresponding region of LukS-PV [Hlg2(D)SPV] and the Z -region was inserted just before Loop-4 (Hlg2insZ), in the presence of 5 pmol LukF in 350 µl of PBS (–). After 10 min, the amount of haemoglobin in the supernatants was measured. Error bars represent standard deviations of triplicate measurements. (B) Binding and pore -forming ability of Hlg2 mutants. Human erythrocytes (108 cells) were incubated at 37°C with Hlg2 (15 pmol) and 5 pmol LukF in 350 µl PBS. After 60 min, haemolytic activity was measured, and the membrane pores formed were detected by western blot analysis using the anti-Hlg2 antiserum. Loops of Hlg2 confers haemolytic activity to LukS-PV Among Luk, PVL and LukED, their F- and S-components are interchangeable with each other, and LukS-PV forms membrane pores on the human leukocytes and shows leukocytolytic activity in combination with LukF (7). However, LukS-PV and LukF even at a high dose of 500 pmol/108 cells, do not show haemolytic activity (Fig. 4A), indicating that LukS-PV lacks the erythrocyte-binding potential. Therefore, the effect of substitution of LukS-PV loop regions with those of Hlg2 was investigated with respect to haemolytic activity against human erythrocytes. Fig. 4 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding Hlg2 residue(s). (A) Human erythrocytes (108 cells) were incubated at 37°C with 500 pmol of both the loop-replaced LukS-PV mutant and LukF (in a 1:1 ratio) in 350 µl PBS(−) for varying time intervals. After incubation, the amount of haemoglobin in the supernatants was measured. The Z -region was deleted in all Loop-4 mutants. Error bars represent standard deviations of triplicate measurements. The lower panel presents the enlarged view of the boxed section. (B) Effect of D- and Z -regions on triple mutants. Haemolysis by the LukS-PV triple mutant [SPV(L1, 2, 4)Hlg2], its Z -region deleted mutant [SPV(L1, 2, 4)Hlg2delZ], and further D-region-replaced mutant [SPV(D, L1, 2, 4)Hlg2delZ], was analysed as described above. Error bars represent standard deviations of triplicate measurements. Fig. 4 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding Hlg2 residue(s). (A) Human erythrocytes (108 cells) were incubated at 37°C with 500 pmol of both the loop-replaced LukS-PV mutant and LukF (in a 1:1 ratio) in 350 µl PBS(−) for varying time intervals. After incubation, the amount of haemoglobin in the supernatants was measured. The Z -region was deleted in all Loop-4 mutants. Error bars represent standard deviations of triplicate measurements. The lower panel presents the enlarged view of the boxed section. (B) Effect of D- and Z -regions on triple mutants. Haemolysis by the LukS-PV triple mutant [SPV(L1, 2, 4)Hlg2], its Z -region deleted mutant [SPV(L1, 2, 4)Hlg2delZ], and further D-region-replaced mutant [SPV(D, L1, 2, 4)Hlg2delZ], was analysed as described above. Error bars represent standard deviations of triplicate measurements. Since a higher amount of toxin component and long incubation time were required for the haemolysis by LukS-PV mutants in combination with LukF, we used a dose of 500 pmol/108 cells to compare the activities of the mutants. This dose was higher than that used for Hlg (5 pmol/108 cells). LOOP-4 mutation of LukS-PV with that of Hlg2 [SPV(L4)Hlg2] showed considerably less activity; however the mutant showed 40% haemolysis after deletion of the Z-region (RRTTH segment) [SPV(L4)Hlg2delZ] (Fig. 4A). Loop-1 mutant [SPV(L1)Hlg2] showed considerably less haemolysis, whereas Loop-2 mutant [SPV(L2)Hlg2] showed no activity. However, the haemolytic activity of SPV(L4)Hlg2delZ significantly increased by an additional mutation in Loop-1 [SPV(L1, 4)Hlg2delZ] or Loop-2 [SPV(L2, 4)Hlg2delZ], indicating that Loop-1 and -2 of Hlg2 play an auxiliary role in haemolysis. In contrast, activity of the Loop-1 and -2 double mutant [SPV(L1, 2)Hlg2] was similar to that of SPV(L1)Hlg2 (Fig. 4A). Similar to the Hlg2 mutant with an inserted Z-region from LukS-PV (Fig. 3, Hlg2insZ), Deletion of the Z-region from triple mutant SPV(L1, 2, 4)Hlg2 having the Z-region [SPV(L1, 2, 4)Hlg2delZ] resulted in 100% haemolysis after 60 min. Taken together, we suggest that while Loop-4 of Hlg2 potentially plays a role in determining erythrocyte specificity, loops-1 and -2 assist in binding. The D-region, K23RLAI27 segment in the cap domain, has been identified as a region with a pivotal role in haemolytic activity (21) (Fig. 1), and Hlg2 mutant where KRLAI was substituted with DKWGV of LukS-PV [Hlg2(D)SPV] showed reduced haemolytic activity (Fig. 3A). When 5 pmol Hlg2(D)SPV was incubated with 108 cells for 10 min, the amount of bound Hlg2(D)SPV was estimated to be only 40% of that of the wild-type Hlg2 using western blotting. Thus, the amount of Hlg2(D)SPV added to the standard reaction mixture was increased to 15 pmol, and after a 10-min incubation, the amount of bound Hlg2(D)SPV mutant increased to 60% of that of the wild-type Hlg2, and weak haemolysis was observed (Fig. 3A). Further incubation with 15 pmol Hlg2(D)SPV mutant showed 62.3% haemolysis after 60 min, and weaker pore-forming ability than the wild type was observed (Fig. 3B), suggesting D-region plays a role in the haemolytic activity via its pore-forming ability or stability. Indeed, further substitution of loop-D with KRLAI of Hlg2 in SPV(L1, 2, 4)Hlg2delZ [SPV(D, L1, 2, 4)Hlg2delZ] resulted in more than 90% haemolysis after a 10-min incubation (Fig. 4B). However, SPV(D, L1, 2, 4)Hlg2delZ had to be used at a high dose in combination with LukF (500 pmol/108 cells) to complete haemolysis within a short incubation time. Additional factors responsible for the strong haemolytic activity of Hlg2 at low dose remain to be investigated. Effect of loop-replacement mutations in LukE component with LukS-PV To gain insights into the erythrocyte recognition mechanism of LukE, the effects of loop replacement in LukE on its haemolytic activity were investigated in the same way as those investigated for Hlg2. In the presence of 100 pmol each of LukD and LukE components with 108 human erythrocytes, about 65% lysis was observed; however, this toxin dose used was 20-fold more than that used for Hlg (Fig. 5A). Loop-4 mutation of LukE resulted in complete loss of the activity with 100 pmol of both the mutant and LukD. Unlike Hlg2, Loop-1 and -2 mutations in LukE markedly decreased its haemolytic activity (Fig. 5A), and it was attributed to the loss of binding ability to human erythrocytes (Fig. 5B). These results suggested that Loop-4 of LukE was important for binding to human erythrocytes in combination with LukD. In contrast, the Loop-3 mutant [LukE(L3)SPV] showed slightly higher haemolytic activity than the wild-type LukE. Fig. 5 View largeDownload slide Effect of loop-replacement mutation between Hlg2 and LukE. Haemolytic activity (A) and binding ability (B) of LukE loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both loop-replaced LukE mutants and LukD (in a 1:1 ratio) in 350 µl PBS, as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured. In parallel, the amount of membrane-bound LukE mutants on the erythrocyte membrane using 100 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Haemolytic activity (C) and binding ability (D) of LukE loop-replaced mutants with Hlg2 loops were analysed in the same manner. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**; P ≤ 0.01). Fig. 5 View largeDownload slide Effect of loop-replacement mutation between Hlg2 and LukE. Haemolytic activity (A) and binding ability (B) of LukE loop-replaced mutants with LukS-PV loops. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both loop-replaced LukE mutants and LukD (in a 1:1 ratio) in 350 µl PBS, as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured. In parallel, the amount of membrane-bound LukE mutants on the erythrocyte membrane using 100 pmol of each component was quantified by western blot analysis using the anti-Hlg2 antiserum. Haemolytic activity (C) and binding ability (D) of LukE loop-replaced mutants with Hlg2 loops were analysed in the same manner. Error bars represent standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**; P ≤ 0.01). To confirm the haemolytic activity of LukE rim domain loops, a series of LukS-PV mutants (similar to Hlg2 mutants constructed above) with loops replaced by corresponding regions from LukE were constructed, and their haemolytic activities were tested. Unlike LukS-PV mutants having Hlg2 Loop-4, LukS-PV mutants with LukE Loop-4 showed 60% haemolysis in combination with LukF, with or without the Z-region [SPV(L4)LukE and SPV(L4)LukEdelZ, respectively] (Fig. 6). In addition, further mutations in Loop-2 [SPV(L2, 4)LukE] and in loops-1 and -2 [SPV(L1, 2, 4)LukE] did not enhance the haemolytic activity of SPV(L4)LukE, suggesting that in combination with LukF, LukS-PV loops-1 and -2 behave similar to those of LukE. Fig. 6 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding LukE residue(s). Human erythrocytes (108 cells) were incubated at 37°C for 120 min with 500 pmol of both loop-replaced LukS-PV mutants and LukF or LukD (in a 1:1 ratio) in 350 µl PBS. The amount of haemoglobin in the supernatants was measured after 1-h incubation, and haemolytic activity combined with LukF or LukD was indicated by open and closed bar, respectively. Error bars represent the standard deviations of triplicate measurements. Fig. 6 View largeDownload slide Analyses of LukS-PV loop-replacement mutation with corresponding LukE residue(s). Human erythrocytes (108 cells) were incubated at 37°C for 120 min with 500 pmol of both loop-replaced LukS-PV mutants and LukF or LukD (in a 1:1 ratio) in 350 µl PBS. The amount of haemoglobin in the supernatants was measured after 1-h incubation, and haemolytic activity combined with LukF or LukD was indicated by open and closed bar, respectively. Error bars represent the standard deviations of triplicate measurements. Effect of loop-replacement mutations between Hlg2 and LukE Loop-4 of both Hlg2 and LukE was suggested to play an important role in mediating haemolytic activity. Loop-1 and -2 of both components assisted in the binding of these components. However, the degree of contribution of these loops, which conferred haemolytic activity on LukS-PV, towards haemolysis was different for Hlg2 and LukE mutants. Thus, we investigated the effect of mutual loop substitution between LukE and Hlg2. Surprisingly, in combination with LukD, the haemolytic activity of LukE was abolished by Loop-4 replacement mutation with Hlg2 Loop-4 [LukE(L4)Hlg2] (Fig. 5C). The haemolytic activity of LukE also decreased following Loop-2 mutation [LukE(L2)Hlg2], whereas Loop-1, -3 or D-region substitutions with those of the corresponding Hlg2 regions [LukE(L1)Hlg2, LukE(L3)Hlg2 and LukE(D)Hlg2] slightly increased the activity (Fig. 5C). These changes correlated with the human erythrocyte-binding abilities of these mutants (Fig. 5D), indicating that Hlg2 Loop-4 did not function in the human erythrocyte-binding ability of LukE in combination with LukD. Effect of partner F-components for the haemolytic activities of mutant S-components For further investigating Loop-4 compatibility between LukE and Hlg2, a Hlg2 mutant with LukE Loop-4 [Hlg2(L4)LukE] was constructed. Unlike the LukE mutant with Hlg2 Loop-4 [LukE(L4)Hlg2], Hlg2(L4)LukE showed haemolytic activity in combination with LukD, but it required higher dose than intact Hlg2 (Fig. 7B). It is known that LukF enhances the binding of Hlg2 toward human erythrocytes (22). Therefore, we examined the synergistic actions of F-components on the haemolytic activities of the S-components mutants. Fig. 7 View largeDownload slide Combinatorial effect of S-component Loop-4 mutants and F-components on haemolysis. Haemolytic activity and binding ability of S- and F-components with LukF and with LukD. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both S- and F-components (in a 1:1 ratio) in 350 µl PBS(–), as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured (A and B). In parallel, the amount of membrane-bound S- and F-components on the erythrocyte membrane using 50 pmol of each component was quantified by western blot analysis using the anti-Hlg2 and anti-LukF antisera (C–F). The amount of LukE(L4)Hlg2 bound to the erythrocyte membrane at 100 pmol was quantified by western blot analysis using the anti-Hlg2 antiserum (G and H). Error bars represent the standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**P ≤ 0.01). Fig. 7 View largeDownload slide Combinatorial effect of S-component Loop-4 mutants and F-components on haemolysis. Haemolytic activity and binding ability of S- and F-components with LukF and with LukD. Human erythrocytes (108 cells) were incubated at 37°C with varying amounts of both S- and F-components (in a 1:1 ratio) in 350 µl PBS(–), as indicated in the figure. After 1 h, the amount of haemoglobin in the supernatants was measured (A and B). In parallel, the amount of membrane-bound S- and F-components on the erythrocyte membrane using 50 pmol of each component was quantified by western blot analysis using the anti-Hlg2 and anti-LukF antisera (C–F). The amount of LukE(L4)Hlg2 bound to the erythrocyte membrane at 100 pmol was quantified by western blot analysis using the anti-Hlg2 antiserum (G and H). Error bars represent the standard deviations of triplicate measurements, and asterisks indicate the statistical significance (**P ≤ 0.01). Interestingly, the haemolytic activities of Loop-4 exchange mutants of Hlg2 and LukE differed in combination with the F-component. Hlg2 mutant having Loop-4 from LukE [Hlg2(L4)LukE] showed almost same haemolytic activity in combination with LukF (Fig. 7A), and it markedly increased from haemolytic activity in combination with LukD (Fig. 7B). Almost same amounts of LukF and LukD bound to human erythrocytes in the combination of any S-components (Fig. 7E and F), indicating that this reduction of haemolytic activity was not due to the binding of F-components. Meanwhile, compared with intact Hlg2, binding of mutant Hlg2(L4)LukE to human erythrocytes increased in combination with both LukD and LukF used at a concentration of 50 pmol/108 cells (Fig. 7C and D). These observations suggested that LukF enhanced the haemolytic activity of low doses of the Hlg2 Loop-4 mutant; however, the induction of haemolytic activity was not solely due to toxin binding. Moreover, LukF also induced haemolytic activity with LukE mutant with Hlg2 Loop-4 [LukE(L4)Hlg2] at 100 pmol, which did not show any activity with LukD (Fig. 7A and B), and the increase of haemolytic activity was according to the binding ability of LukE(L4)Hlg2 (Fig. 7G and H). This result suggested that the binding of the LukE(L4)Hlg2 having Hlg2 Loop-4 was stimulated by LukF, as observed in Hlg2 (22). Similarly, LukS-PV loop-replaced mutants with LukE segments showed markedly higher haemolytic activity in combination with LukF, than with LukD (Fig. 6). This ‘partner effect’ by LukF may be involved in enhancement of S-component binding toward human erythrocytes, and also mediating inter-molecular interactions between the S- and F-components during formation of ring-shaped complex. In conclusion, rim loops of both Hlg2 and LukE components, especially Loop-4, play an important role in erythrocyte binding, whereas their amino acid sequence were different each other (Fig. 1A and Table I). Our results provide novel insight into the different target receptor recognition mechanism of Hlg2 and LukE using a combination of multiple loops in the rim domain. It has been suggested that Hlg2 and LukE recognize the same receptor via different site of Duffy antigen receptor for chemokines (DARC), on human erythrocytes (9). Thus, our findings about incompatibility in the erythrocyte binding between the rim domain loops of Hlg2 and LukE may reflect the difference in their target regions. In fact, unlike Hlg2 Loop-4, LukE Loop-4 segment conferred haemolytic activity to LukS-PV without deletion of the Z-region, indicating that the interaction between LukE Loop-4 and erythrocyte receptor was not disturbed by the Z-region (Fig. 6). LukE and Hlg2 also share CXC chemokine receptor 1 (CXCR1) and CXCR 2 on neutrophils (23–25) as their leukocyte receptors, and LukE Loop-3-containing region was required for targeting this receptor, whereas the Loop-4-containing region was involved in cell killing (24). The different behaviour of LukE Loop-4 towards CXCRs and DARC may be explained by the difference in the interaction between LukE and its target molecules. Furthermore, LukE uses the chemokine receptor C-C chemokine receptor type 5(CCR5) on T lymphocytes (23, 26). Tam et al suggested that the amino acid segment K64GSGYE69 in the rim domain, which corresponded to Loop-1 in LukE, was required for CCR5 targeting and cytotoxicity using an in-frame deletion mutant (27). Our preliminary experiments against hCCR5/ Chinese hamster ovary (CHO)-K1 cells (28) showed that Loop-4 replacement with Hlg2 segment [LukE(L4)Hlg2] reduced cytotoxic activity of LukE in combination with LukD, (Data not shown), suggesting that Loop-4 of LukE may also play a role in hCCR5-mediated cytotoxicity. It is difficult to understand how Hlg2 and LukE components having different rim domain amino acid sequences share several different receptor molecules, including DARC on their target cells. For understanding the detailed mechanism underlying recognition of different receptors, further structural studies on interaction between S-component loops and extracellular loops of receptors are required. Acknowledgements We would like to thank Editage (www.editage.jp) for English language editing. Funding This work was supported by JSPS Grant-in-Aid for Challenging Exploratory Research (KAKENHI) Grant Number JP16K14897 (to J. K). Conflict of Interest None declared. References 1 Kaneko J., Kamio Y. ( 2004) Bacterial two-component and hetero-heptameric pore-forming cytolytic toxins: structures, pore-forming mechanism, and organization of the genes. Biosci. Biotechnol. Biochem . 68, 981– 1003 Google Scholar CrossRef Search ADS PubMed  2 Kamio Y., Rahman A., Nariya H., Ozawa T., Izaki K. ( 1993) The two Staphylococcal bi-component toxins, leukocidin and gamma-hemolysin, share one component in common. FEBS Lett . 321, 15– 18 Google Scholar CrossRef Search ADS PubMed  3 Kaneko J., Ozawa T., Tomita T., Kamio Y. ( 1997) Sequential binding of Staphylococcal gamma-hemolysin to human erythrocytes and complex formation of the hemolysin on the cell surface. Biosci. Biotechnol. Biochem . 61, 846– 851 Google Scholar CrossRef Search ADS PubMed  4 Yamashita K., Kawai Y., Tanaka Y., Hirano N., Kaneko J., Tomita N., Ohta M., Kamio Y., Yao M., Tanaka I. ( 2011) Crystal structure of the octameric pore of staphylococcal γ-hemolysin reveals the β-barrel pore formation mechanism by two components. Proc. Nat.l Acad. Sci. U. S. A . 108, 17314– 17319 Google Scholar CrossRef Search ADS   5 Yamashita D., Sugawara T., Takeshita M., Kaneko J., Kamio Y., Tanaka I., Tanaka Y., Yao M. ( 2014) Molecular basis of transmembrane beta-barrel formation of staphylococcal pore-forming toxins. Nat. Commun.  5, 4897 Google Scholar CrossRef Search ADS PubMed  6 Prévost G., Cribier B., Couppié P., Petiau P., Supersac G., Finck-Barbançon V., Monteil H., Piemont Y. ( 1995) Panton-Valentine leucocidin and gamma-hemolysin from Staphylococcus aureus ATCC 49775 are encoded by distinct genetic loci and have different biological activities. Infect. Immun . 3, 4121– 4129 7 Gravet A., Colin D.A., Keller D., Giradot R., Monteil H., Prévost G. ( 1998) Characterization of a novel structural member, LukE-LukD, of the bi-component staphylococcal leucotoxins family. FEBS Lett . 436, 202– 208 Google Scholar CrossRef Search ADS PubMed  8 Morinaga N., Kaihou Y., Noda M. ( 2003) Purification, cloning and characterization of variant LukE-LukD with strong leukocidal activity of staphylococcal bi-component leukotoxin family. Microbiol. Immunol . 47, 81– 90 Google Scholar CrossRef Search ADS PubMed  9 Spaan A.N., Reyes-Robles T., Badiou C., Cochet S., Boguslawski K.M., Yoong P., Day C.J., de Haas C.C.J., van Kessel K.P.M., Vandenesch F., Jennings M.P., Le Van Kim C., Colin Y., van Strijp J.A., Henry T., Torres V.J. ( 2015) Staphylococcus aureus Targets the Duffy Antigen Receptor for Chemokines (DARC) to Lyse Erythrocytes. Cell Host Microbe  18, 363– 370 Google Scholar CrossRef Search ADS PubMed  10 Song L., Hobaugh M.R., Shustak C., Cheley S., Bayley H., Gouaux J.E. ( 1996) Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science  274, 1859– 1866 Google Scholar CrossRef Search ADS PubMed  11 Olson R., Nariya H., Yokota K., Kamio Y., Gouaux J.E. ( 1999) Crystal structure of staphylococcal LukF delineates conformational changes accompanying formation of a transmembrane channel. Nat. Struct. Biol . 6, 134– 140 Google Scholar CrossRef Search ADS PubMed  12 Yokota K., Kamio Y. ( 2000) Tyrosine72 residue at the bottom of rim domain in LukF crucial for the sequential binding of the staphylococcal gamma-hemolysin to human erythrocytes. Biosci. Biotechnol. Biochem . 64, 2744– 2747 Google Scholar CrossRef Search ADS PubMed  13 Monma N., Nguyen V.T., Kaneko J., Higuchi H., Kamio Y. ( 2004) Essential residues, W177 and R198, of LukF for phosphatidylcholine-binding and pore-formation by staphylococcal gamma-hemolysin on human erythrocyte membranes. J. Biochem . 136, 427– 431 Google Scholar CrossRef Search ADS PubMed  14 Nishiyama A., Nariya H., Kamio Y. ( 1998) Phosphorylation of LukS by protein kinase A is crucial for the LukS-specific function of the staphylococcal leukocidin on human polymorphonuclear leukocytes. Biosci. Biotechnol. Biochem . 62, 1834– 1838 Google Scholar CrossRef Search ADS PubMed  15 Spaan A.N., Henry T., van Rooijen W.J., Perret M., Badiou C., Aerts P.C., Kemmink J., de Haas C.J., van Kessel K.P., Vandenesch F., Lina G., van Strijp J.A. ( 2013) The staphylococcal toxin Panton-Valentine Leukocidin targets human C5a receptors. Cell Host Microbe  13, 584– 594 Google Scholar CrossRef Search ADS PubMed  16 Laventie B.J., Guérin F., Mourey L., Tawk M.Y., Jover E., Maveyraud L., Prévost G. ( 2014) Residues essential for Panton-Valentine leukocidin S component binding to its cell receptor suggest both plasticity and adaptability in its interaction surface. PLoS One  9, e92094 Google Scholar CrossRef Search ADS PubMed  17 Kuroda M., Ohta T., Uchiyama I., Baba T., Yuzawa H., Kobayashi I., Cui L., Oguchi A., Aoki K., Nagai Y., Lian J., Ito T., Kanamori M., Matsumaru H., Maruyama A., Murakami H., Hosoyama A., Mizutani-Ui Y., Takahashi N.K., Sawano T., Inoue R., Kaito C., Sekimizu K., Hirakawa H., Kuhara S., Goto S., Yabuzaki J., Kanehisa M., Yamashita A., Oshima K., Furuya K., Yoshino C., Shiba T., Hattori M., Ogasawara N., Hayashi H., Hiramatsu K. ( 2001) Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet  357, 1225– 1240 Google Scholar CrossRef Search ADS PubMed  18 Baba T., Bae T., Schneewind O., Takeuchi F., Hiramatsu K. ( 2008) Genome sequence of Staphylococcus aureus strain Newman and comparative analysis of staphylococcal genomes: polymorphism and evolution of two major pathogenicity islands. J. Bacteriol . 190, 300– 310 Google Scholar CrossRef Search ADS PubMed  19 Kaneko J., Kimura T., Kawakami Y., Tomita T., Kamio Y. ( 1997) Panton-valentine leukocidin genes in a phage-like particle isolated from mitomycin C-treated Staphylococcus aureus V8 (ATCC 49775). Biosci. Biotechnol. Biochem . 61, 1960– 1962 Google Scholar CrossRef Search ADS PubMed  20 Sugawara T., Yamashita D., Kato K., Peng Z., Ueda J., Kaneko J., Kamio Y., Tanaka Y., Yao M. ( 2015) Structural basis for pore-forming mechanism of α-hemolysin. Toxicon  108, 226– 231 Google Scholar CrossRef Search ADS PubMed  21 Nariya H., Kamio Y. ( 1997) Identification of the minimum segment essential for the H gamma II-specific function of staphylococcal gamma-hemolysin. Biosci. Biotechnol. Biochem.  61, 1786– 1788 Google Scholar CrossRef Search ADS PubMed  22 Nguyen V.T., Kamio Y., Higuchi H. ( 2003) Single-molecule imaging of cooperative assembly of gamma-hemolysin on erythrocyte membranes. embo J . 22, 4968–4679 23 Spaan A.N., van Strijp J.A.G., Torres V.J. ( 2017) Leukocidins: staphylococcal bi-component pore-forming toxins find their receptors. Nat. Rev. Micro.  15, 435– 447 Google Scholar CrossRef Search ADS   24 Reyes-Robles T., Alonzo F.3rd., Kozhaya L., Lacy D.B., Unutmaz D., Torres V.J. ( 2013) Staphylococcus aureus leukotoxin ED targets the chemokine receptors CXCR1 and CXCR2 to kill leukocytes and promote infection. Cell Host Microbe  14, 453– 459 Google Scholar CrossRef Search ADS PubMed  25 Spaan A.N., Vrieling M., Wallet P., Badiou C., Reyes-Robles T., Ohneck E.A., Benito Y., de Haas C.J., Day C.J., Jennings M.P., Lina G., Vandenesch F., van Kessel K.P., Torres V.J., van Strijp J.A., Henry T. ( 2014) The staphylococcal toxins γ-haemolysin AB and CB differentially target phagocytes by employing specific chemokine receptors. Nat. Commun.  5, 5438 Google Scholar CrossRef Search ADS PubMed  26 Alonzo F.3rd., Kozhaya L., Rawlings S.A., Reyes-Robles T., DuMont A.L., Myszka D.G., Landau N.R., Unutmaz D., Torres V.J. ( 2012) CCR5 is a receptor for Staphylococcus aureus leukotoxin ED. Nature  493, 51– 55 Google Scholar CrossRef Search ADS PubMed  27 Tam K., Schultz M., Reyes-Robles T., Vanwalscappel B., Horton J., Alonzo F.3rd., Wu B., Landau N.R., Torres V.J. ( 2016) Staphylococcus aureus Leukocidin LukED and HIV-1 gp120 Target Different Sequence Determinants on CCR5. MBiol  7, e02024-16 pii Google Scholar CrossRef Search ADS   28 Maeda K., Yoshimura K., Shibayama S., Habashita H., Tada H., Sagawa K., Miyakawa T., Aoki M., Fukushima D., Mitsuya H. ( 2001) Novel low molecular weight spirodiketopiperazine derivatives potently inhibit R5 HIV-1 infection through their antagonistic effects on CCR5. J. Biol. Chem.  276, 35194– 35200 Google Scholar CrossRef Search ADS PubMed  Abbreviations Abbreviations C5aR C5a receptor CCR5 C-C chemokine receptor type 5 CD circular dichroism CHO Chinese hamster ovary CXCR CXC chemokine receptor DARC Duffy antigen receptor for chemokines Hla α-haemolysin Hlg γ-Haemolysin Hlg2 (alias HlgA) S-component of Hlg Luk leukocidin LukD F-component of LukED LukE S-component of LukED LukF (alias HlgB) F-component of Luk LukS S-component of Luk LukS-PV S-component of PVL PBS(−) phosphate-buffered saline without calcium and magnesium PC phosphatidylcholine PFT pore-forming toxin PVL Panton-Valentine leukocidin SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis © The Author(s) 2018. Published by Oxford University Press on behalf of the Japanese Biochemical Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The Journal of BiochemistryOxford University Press

Published: Feb 17, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off