Serum apolipoprotein A-I potentiates the therapeutic efficacy of lysocin E against Staphylococcus aureus

Hamamoto, Hiroshi; Panthee, Suresh; Paudel, Atmika; Ishii, Kenichi; Yasukawa, Jyunichiro; Su, Jie; Miyashita, Atsushi; Itoh, Hiroaki; Tokumoto, Kotaro; Inoue, Masayuki; Sekimizu, Kazuhisa

doi:10.1038/s41467-021-26702-0

Download PDF

Article
Open access
Published: 04 November 2021

Serum apolipoprotein A-I potentiates the therapeutic efficacy of lysocin E against Staphylococcus aureus

Nature Communications volume 12, Article number: 6364 (2021) Cite this article

3943 Accesses
10 Citations
31 Altmetric
Metrics details

Subjects

Abstract

Lysocin E is a lipopeptide with antibiotic activity against methicillin-resistant Staphylococcus aureus. For unclear reasons, the antibacterial activity of lysocin E in a mouse systemic infection model is higher than expected from in vitro results, and the in vitro activity is enhanced by addition of bovine serum. Here, we confirm that serum from various species, including humans, increases lysocin E antimicrobial activity, and identify apolipoprotein A-I (ApoA-I) as an enhancing factor. ApoA-I increases the antibacterial activity of lysocin E when added in vitro, and the antibiotic displays reduced activity in ApoA-I gene knockout mice. Binding of ApoA-I to lysocin E is enhanced by lipid II, a cell-wall synthesis precursor found in the bacterial membrane. Thus, the antimicrobial activity of lysocin E is potentiated through interactions with host serum proteins and microbial components.

Human serum triggers antibiotic tolerance in Staphylococcus aureus

Article Open access 19 April 2022

High density lipoproteins mediate in vivo protection against staphylococcal phenol-soluble modulins

Article Open access 28 July 2021

The lipid components of high-density lipoproteins (HDL) are essential for the binding and transportation of antimicrobial peptides in human serum

Article Open access 16 February 2022

Introduction

Since the discovery of penicillin, antimicrobial agents have been screened on the basis of their in vitro antimicrobial activity, which does not accurately reflect the host environment. Protein binding affects the activity of antimicrobials. For example, some host factors, such as serum albumin, bind to some antimicrobials, which influences their activity. Indeed, the activity of many antimicrobials is decreased by adding blood serum to the culture medium due to the decreased free-drug concentration resulting from large distribution¹. On the other hand, the activities of some antimicrobials, such as azithromycin, are enhanced by serum factors in the blood². Daptomycin and telavancin exhibit high protein binding capacities, although these high protein binding capacities do not affect their therapeutic efficacy^3,4. Evaluation of the therapeutic efficacy of antimicrobials in vivo is therefore crucial.

Our unique strategy to search for antimicrobial agents, utilising silkworms as a host for evaluating the in vivo effects of antimicrobials at early stages of screening^5,6, allows for the exclusion of compounds with reduced antimicrobial activity due to host protein binding^7,8, and led to the discovery of a novel antibiotic, lysocin E⁹ (Supplementary Fig. 1a). Lysocin E targets menaquinone, a cofactor in the electron transport chain, in the cell membrane of Staphylococcus aureus and disrupts the membrane, which results in rapid bactericidal activity. The total synthesis of lysocin E has been achieved¹⁰ and comprehensive amino acid substitution by the “one-bead-one-compound” method revealed that the membrane disruption of liposomes containing menaquinone is well correlated with the antimicrobial activity of lysocin E derivatives against S. aureus¹¹. In a previous study, we demonstrated that lysocin E is therapeutically effective against the methicillin-sensitive S. aureus (MSSA) Smith strain at lower doses than vancomycin, despite its lower antibacterial activity compared with vancomycin⁹. In addition, we found that the antimicrobial activity of lysocin E in vitro is highly increased by adding bovine serum to the medium¹¹. We, therefore, hypothesised that the antimicrobial activity of lysocin E is promoted by factors in the host environment.

Here, we report the identification of ApoA-I and ApoA-II as enhancing factors in serum for lysocin E antimicrobial and therapeutic activity. We further demonstrate that ApoA-I promotes membrane-damaging and bactericidal activity of lysocin E at a sub-minimum inhibitory concentration (MIC) against S. aureus and lipid II mediates the binding of lysocin E to ApoA-I. Furthermore, the antimicrobial activity of nisin, which is also lipid II-binding antimicrobial, is enhanced by ApoA-I and lipid II interaction. Overall, our study demonstrates the interaction of host factor(s) and the bacterial component can potentiate these antimicrobial activities.

Results

Serum ApoA-I enhances lysocin E activity

We found that the ratio of the median effective dose to the MIC (ED₅₀/MIC) is more than 16-fold smaller for lysocin E compared with the other anti-MRSA drugs, indicating that lysocin E has a high therapeutic efficacy compared with the other anti-MRSA drugs (Table 1). In contrast to clinically used anti-MRSA microbials such as vancomycin and linezolid, lysocin E (1) has high protein binding capacity (Table 1). The antimicrobial activity of lysocin E against MSSA and MRSA strains, however, could be increased by adding sera from various species, including human or silkworm hemolymph (Fig. 1a, Table 1 and Supplementary Table 1). The increased antimicrobial activity of lysocin E induced by bovine serum was also observed against other Gram-positive bacteria such as Staphylococcus spp. and Bacillus spp. (Supplementary Table 1). To understand the mechanism by which serum enhances the bactericidal activity of lysocin E, we purified the responsible factors in bovine serum by measuring the increase in antimicrobial activity of lysocin E due to exposure to each of the various serum components with activity guided fractionation (Supplementary Fig. 2a). Ethanol extraction, octadecyl silica (ODS) column chromatography and gel filtration chromatography were performed. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) revealed that the elution patterns of 32 and 10 kDa proteins from gel filtration column chromatography were consistent with the antimicrobial-enhancing activity of lysocin E (Supplementary Fig. 2b, c). Peptide mass fingerprinting analysis revealed that the proteins were bovine ApoA-I (bApoA-I) and ApoA-II (Supplementary Fig. 2d). Adding recombinant human ApoA-I (rhApoA-I) or purified human ApoA-II (hApoA-II) to the culture medium reduced the MIC of lysocin E (Fig. 1b). In addition, the specific activity of rhApoA-I and recombinant mouse ApoA-I (rmApoA-I) was similar to that of bApoA-I (Supplementary Fig. 2e), confirming that ApoA-I and ApoA-II are serum factors responsible for enhancing the antimicrobial activity of lysocin E. rhApoA-I decreased the MIC against various MSSA and MRSA strains, including the highly virulent USA300 JE2 strain (Supplementary Table 2). Although ApoA-I exhibits antimicrobial activity against Gram-negative bacteria such as Escherichia coli and Klebsiella pneumoniae^12,13, the addition of up to 300 or 600 µg ml⁻¹ of purified bovine and human ApoA-I or ApoA-II did not inhibit S. aureus growth (Supplementary Fig. 2f), as previously reported¹³. Thus, the enhanced antimicrobial activity of lysocin E induced by ApoA-I and ApoA-II is not due to the direct antimicrobial activity of ApoA-I and ApoA-II against S. aureus.

Table 1 Therapeutic effect in a mouse systemic MRSA infection model and antimicrobial activity of anti-MRSA agent against MRSA.

Full size table

**Fig. 1: Apolipoprotein A enhances the antimicrobial and therapeutic activity of lysocin E.**

To obtain genetic evidence for the involvement of ApoA in the therapeutic effects of lysocin E, we examined the therapeutic effect of lysocin E using gene knockout mice¹⁴. In mouse plasma, the ApoA-I concentration is ~7-fold higher than that of ApoA-II¹⁵. In addition, bovine and human ApoA-I and ApoA-II similarly enhanced the antimicrobial activity of lysocin E (Supplementary Fig. 2a, e). Further, we confirmed that mixing rhApoA-I and hApoA-II in a similar concentration ratio in human blood had no synergistic effects (Supplementary Fig. 2g), suggesting that these factors have similar antimicrobial-enhancing effects and work in an independent manner. Therefore, we performed further analysis using Apoa1 gene knockout mice. The enhanced antimicrobial activity of lysocin E in the serum of Apoa1 homozygous-deficient mice was 6-fold lower than that in wild-type mice (Fig. 1c). The remaining activity of the serum in the Apoa1 gene-deficient mice could be explained by ApoA-II activity because ApoA-II is present in the serum of Apoa1 homozygous-deficient mice at the same level as in wild-type mice¹⁶. Next, we examined the therapeutic efficacy of lysocin E in an S. aureus systemic infection model using Apoa1 homozygous-deficient mice. Treatment with lysocin E did not reduce the number of surviving bacteria in the kidney of Apoa1 homozygous-deficient mice, whereas it did reduce the number of bacteria in the kidney of wild-type mice (Fig. 1d). These results suggest that ApoA-I contributes to the therapeutic effect of lysocin E in mice.

ApoA-I activity is mediated via lipid II

ApoA-I and high-density lipoprotein (HDL) containing ApoA-I bind to the surface of bacteria^17,18 and free lipoteichoic acid (LTA)¹⁹. We confirmed that rhApoA-I binds to the surface of S. aureus in a concentration-dependent manner (Fig. 2a and Supplementary Fig. 3a). Because adding rhApoA-I did not increase the amount of lysocin E bound to the surface of S. aureus (Supplementary Fig. 3b), the enhancing effect of ApoA-I cannot be explained by an increased accumulation of lysocin E on the cell surface. To elucidate the lysocin E-enhancing functions of ApoA-I, we searched for lysocin E derivatives that did not respond to the addition of rhApoA-I, but the antimicrobial activity did not differ from that of the natural type in the absence of rhApoA-I. We found that the antimicrobial activity of Bu-type lysocin E (2) (Supplementary Fig. 1a)¹⁰, which has a modified fatty acid chain moiety, was not enhanced by rhApoA-I (Supplementary Table 3). We assumed that the interaction of lysocin E Bu-type with rhApoA-I differs from that of the natural-type lysocin E. We compared the affinities of the natural- and Bu-type lysocin E for rhApoA-I and found that their binding constants were similar, 3.6 and 3.4 µM, respectively (Supplementary Fig. 4a). The results suggested that lysocin E interacts with ApoA-I, although this interaction does not explain the difference in the ApoA-I response between natural- and Bu-type lysocin E. We then tested the binding affinity of the natural- and Bu-type lysocin E to menaquinone, a cofactor of the electron transport chain of S. aureus that is a target of lysocin E⁹. The binding constant of natural lysocin E with menaquinone was 4.0 µM, a value similar to the previous report⁹, and that of the Bu-type lysocin E was 5.0 µM (Supplementary Fig. 4b). Further, the binding capacity of lysocin E to ApoA-I in a rhApoA-I pull-down assay did not significantly differ between the natural- and Bu-type lysocin E in the presence of menaquinone (Fig. 2b), suggesting that the presence of menaquinone does not affect the affinities of natural- and Bu-type lysocin E for ApoA-I.

**Fig. 2: Lipid II enhances the interaction of lysocin E with ApoA-I.**

Lipid II (6, Supplementary Fig. 1c) was recently reported to interact with lysocin E, and treatment of S. aureus with sub-MIC of lysocin E leads to the accumulation of lipid II in the S. aureus membrane²⁰. We hypothesised that lipid II is involved in the lysocin E antimicrobial-enhancing effect of ApoA-I, and tested this hypothesis in the following experiments. We found that natural-type lysocin E had a significantly higher binding capacity to lipid II than Bu-type lysocin E (Supplementary Fig. 4c). We further demonstrated that lipid II increased the binding of the natural-type lysocin E to rhApoA-I (Fig. 2b) in a pull-down assay in the presence of menaquinone, and this phenomenon was not observed in the absence of menaquinone (Fig. 2c). In addition, other ApoA-I binding lipids, such as phosphatidylglycerol (PG)²¹, cardiolipin (CL)²² and LTA¹⁹, which are present in the S. aureus membrane, did not increase natural-type lysocin E binding to rhApoA-I in the presence or absence of menaquinone (Fig. 2b, c). To further confirm the importance of the lipid II–ApoA-I interaction, we constructed rhApoA-I with deletion of amino acids from the N-terminal (1–65 AA; ∆N), middle (146–160 AA; ∆M) and C-terminal (185–243 AA; ∆C) regions, known to be involved in ApoA-I function^23,24 (Supplementary Fig. 5a). All these variants retained at least one-third of their lysocin E-enhancing activity (Supplementary Fig. 5b). Next, we constructed ΔNM and ΔNC variants and found that the ΔNM variant lost the majority of the lysocin E-enhancing activity (Fig. 2d and Supplementary Fig. 5b). We observed little difference in the binding affinity of lysocin E to rhApoA-I ∆NM compared to rhApoA-I wild-type (Supplementary Fig. 4a), while, compared with wild-type rhApoA-I, rhApoA-I ΔNM had lower lipid II-dependent lysocin E binding capacity (Fig. 2e). These findings suggested that lipid II enhanced the recruitment of the lysocin E complex to ApoA-I.

Nisin is another antimicrobial compound that binds lipid II with membrane-damaging and bactericidal activities against S. aureus²⁵. We found that the antimicrobial activity of nisin was enhanced by adding rhApoA-I as in the case of lysocin E (Supplementary Table 3). We then demonstrated in a pull-down assay that the binding of nisin to rhApoA-I was increased by lipid II, but not by other ApoA-I binding lipids (Fig. 3a). These findings suggest that the binding of nisin to ApoA-I is mediated by lipid II, and results in enhanced antimicrobial activity. We further tested whether the bactericidal activities of lysocin E and nisin are enhanced by ApoA-I, and found that rhApoA-I enhanced the bactericidal activity of both antimicrobials in sub-MIC conditions (Fig. 3b). We also evaluated whether the membrane-damaging activity and loss of membrane potential induced by lysocin E and nisin were increased by the addition of ApoA-I. The membrane-damaging activity and loss of membrane potential in S. aureus were partially induced by both antimicrobials at sub-MICs and increased by the addition of rhApoA-I (Fig. 3c, d). In addition, externally added lipid II did not affect the antimicrobial activity of lysocin E without rhApoA-I, while it attenuated the lysocin E-enhancing activity of rhApoA-I, suggesting that the interaction of lipid II and ApoA-I is essential for lysocin E function at a sub-MIC (Supplementary Table 3). These findings implied that lysocin E and nisin induce loss of membrane integrity at sub-MIC and that ApoA-I promotes the membrane-disrupting activity of both antimicrobials via lipid II, thereby enhancing the bactericidal actions of both antimicrobials.

**Fig. 3: Nisin and lysocin E exhibit common characteristics under sub-MIC.**

A role of menaquinone in the membrane disruption activity of lysocin E

We further examined the role of menaquinone in the presence of lipid II in the membrane disruptive activity of lysocin E. Santiago et al.²⁰ reported that lysocin E can be eluted from immobilised lipid II bound to lysocin E by the addition of lipid II, but not by menaquinone. Our finding that lysocin E bound to immobilised menaquinone was eluted by the addition of lipid II, but not by menaquinone, confirms this observation (Fig. 4a). These results suggested that both lipid II and menaquinone have an affinity for lysocin E, and that lipid II has a higher affinity than menaquinone for lysocin E.

**Fig. 4: The role of menaquinone and lipid II in membrane-damaging activity.**

The ∆menA and ∆menB strains having a menaquinone content <0.1% than that in wild type⁹ exhibit resistance against lysocin E. We confirmed that the lipid II content of these mutants was reduced to ~13% compared with wild type (Fig. 4b), but the antimicrobial activity of nisin against these mutants was indistinguishable from that of the wild type (Fig. 4c). In addition, externally added lipid II in the medium did not affect the antimicrobial activity of lysocin E, whereas the antimicrobial activity of nisin was significantly decreased (Supplementary Table 3) by the addition of lipid II. Furthermore, we demonstrated that lipid II in lipid vesicles is not involved in the membrane disruptive activity of lysocin E (Fig. 4d), while nisin disrupts liposomes containing lipid II as previously reported²⁶, but not liposomes containing menaquinone (Fig. 4e). We assume that lysocin E could not interact with lipid II in the membrane, which is supported by the results that delipidated lipid II (Supplementary Fig. 1d) does not elute lysocin E immobilised on menaquinone (Fig. 4a). On the other hand, menaquinone is required for loading lysocin E onto/into the membrane and has essential functions for the membrane disruptive activity of lysocin E. Subsequently, lysocin E might transfer to lipid II and disrupt the membrane further in the presence of ApoA-I (Fig. 4f).

Discussion

The findings of the present study demonstrated that the antimicrobial activity of lysocin E is enhanced by apolipoproteins in sera, contributing to the therapeutic efficacy of lysocin E (Fig. 1). To our knowledge, this is the first study providing genetic evidence that specific host factors enhance the activity of an antimicrobial, thereby contributing to its therapeutic effects. Lysocin E was identified by screening based on its therapeutic efficacy using a silkworm S. aureus infection model⁹. Serum and ApoA-I and ApoA-II from various species had enhancing effects on the antimicrobial activity of lysocin E (Fig. 1a and Supplementary Fig. 2a, e). In addition, the antimicrobial activity of lysocin E was also enhanced by silkworm hemolymph (Fig. 1a), suggesting that this phenomenon is conserved widely among animals. Insect hemolymph contains apolipophorin I, II and III, which are involved in the transport of lipids, like mammalian ApoA-I^27,28 and apolipophorin III and human apolipoprotein A-I have similar properties²⁹. These insect apolipophorins bind the surface of S. aureus^30,31. We consider that the use of the silkworm infection model to screen for compounds with therapeutic effectiveness led to the discovery of the antibiotic lysocin E, which has unique features.

Here, we present a presumed model in which small amounts of membrane damage induced by lysocin E and nisin trigger a coordinated action between ApoA-I and lipid II, which finally leads to massive membrane disruption, resulting in bactericidal action (Fig. 4f). ApoA-I constitutes a major protein of HDLs in mammals and HDL carries lipids such as cholesterol. In addition, PG and CL, which are acidic phospholipids constituting the cell membrane of S. aureus³², bind ApoA-I^21,22. Furthermore, a previous study suggested that the C-terminal region of ApoA-I, required for binding lipopolysaccharides, contributes to complement-dependent bactericidal activity against Yersinia enterocolitica²³. From the present study, the addition of sole purified ApoA-I showed enhancing the antimicrobial activity of lysocin E and nisin (Fig. 1b), and PG and CL did not increase binding of lysocin E and nisin to ApoA-I (Figs. 2b, c and 3a), suggesting ApoA-I directly interacted with lipid II and enhanced antimicrobial activity of antimicrobials. Further, C-terminal region had less of an effect on lysocin E antimicrobial activity (Supplementary Fig. 5b), suggesting that functional sites for the lipid II-mediated antimicrobial-enhancing activity of lysocin E differ from this region. In addition, the ApoA-I ∆NM variant had reduced binding capacity against the cell surface of S. aureus compared with wild-type ApoA-I (Fig. 2a), suggesting that these regions are involved in the interaction of ApoA-I with the bacterial surface and contribute to the antimicrobial-enhancing activity of lysocin E. Among ten tandem repeat helices present in ApoA-I, H1 (residues 44–65) and H10 (residues 222–243) possess the highest lipid-binding activity³³. In addition, helix 6 (residues 143–165) is involved in activating lecithin cholesterol acyltransferase and has a low lipid-binding affinity³³. ∆NM variant, lacking helices 1 and 6, had a reduced lysocin E antimicrobial activity-enhancing effect, whereas the ∆C variant, lacking helix 10, did not exhibit a reduced lysocin E antimicrobial activity-enhancing effect. This finding suggests a complex relationship between the lipid-binding nature of ApoA-I and the induction of antimicrobial activity mediated by lipid II. On the basis of the recent structural analysis³⁴, it can be expected that deletion of the N and M regions would severely affect the structure and function of the variant, although the underlying process by which ApoA-I variants form a complex with lipid II and lysocin E, including their structural elucidation, requires further investigation. In this study, we revealed that ApoA-I also interacts with lysocin E (Supplementary Fig. 4a). Even though ApoA-I did not increase lysocin E binding to the cell surface of S. aureus (Supplementary Fig. 3b), we cannot deny the possibility that lysocin E forms a complex with ApoA-I before it encounters the microbial surface because ApoA-I is a major lipoprotein of HDL in the blood. ApoA-II is also found in HDL in the blood, and thus these factors may work independently, while how HDL interacts with S. aureus and lysocin E and contributes to enhancing the antimicrobial activity of lysocin E were not elucidated in this study. Furthermore, the equilibrium dissociation constant (K_d) values of lysocin E for menaquinone and ApoA-I were similar; thus, we assume that a state of equilibrium among them has been established, although it is important to point out that menaquinone is embedded in the membrane and lysocin E binds to lipids. Thus, further detailed analysis is required to elucidate how ApoA-I and ApoA-II impact the pharmacokinetics of lysocin E in the host.

It is important to elucidate the essentiality of lipid II for the ApoA-I-enhancing effect on the antimicrobial activity of lysocin E, but it is difficult to verify using gene disruption strains because lipid II is an essential component for bacterial growth. Competition assays using immobilised menaquinone (Fig. 4a) or lipid II²⁰ suggested that the binding capacity of lipid II to lysocin E is higher than menaquinone and these compounds could not bind simultaneously. Nevertheless, menaquinone is an essential component for the antimicrobial activity of lysocin E, because the ∆menA and ∆menB strains were highly resistant to lysocin E⁹ (Fig. 4c). In addition, biochemical analysis using lipid II-containing artificial liposomes suggested that menaquinone, but not lipid II, is required for the membrane disruptive activity of lysocin E (Fig. 4d). Delipidated lipid II did not interact with lysocin E (Fig. 4a), also suggesting that lysocin E could not interact with lipid II embedded in the lipid bilayer. Thus, we assume that menaquinone is required for loading lysocin E into the cell membrane and exerting small membrane disruption activity at sub-MIC. Subsequently, lysocin E transfers to lipid II in the cell membrane, and ApoA-I interacts with the lipid II–lysocin E complex and enhances the membrane damage to kill S. aureus (Fig. 4f). This model is also consistent with the findings that lipid II increased lysocin E binding to ApoA-I in the presence of menaquinone (Fig. 2b). Nisin disrupted liposomes containing lipid II as previously reported²⁶, but not liposomes containing menaquinone (Fig. 4e). Therefore, although the roles of lipid II in the antimicrobial activity of lysocin E and nisin seem to differ, they share a common mechanism by which the antimicrobial activity modulated by ApoA-I via lipid II is enhanced, implying the existence of a series of similar antimicrobials. Vancomycin³⁵ and teixobactin³⁶ are well-known antimicrobials that also bind lipid II. The antimicrobial activity of these antibiotics, however, are not increased by the addition of serum (Supplementary Table 3 and ref. ³⁶), thus factors other than lipid II binding may be required for ApoA-I-dependent antimicrobial-enhancing activity. Nisin and lysocin E have potent membrane disruptive activity, but vancomycin and teixobactin do not³⁷, suggesting that the membrane disruption activity might be required for ApoA-I function in antimicrobial binding with lipid II. In addition to ApoA-I, host proteins such as albumin³⁸, complement³⁹ and lipids⁴⁰ interact with bacteria. Serum proteins are thought to act as inhibitory molecules against the antimicrobial activity of antimicrobial agents. On the other hand, the complement system may increase the susceptibility of E. coli to vancomycin³⁹, and arachidonic acid may enhance the effect of beta-lactam antibiotics and amikacin⁴⁰, suggesting that host factors other than ApoA-I contribute to the therapeutic efficacy of antimicrobials. On the basis of our results, we propose that the screening of antimicrobials whose activity is enhanced by host proteins that interact with the bacterial surface is an effective way to discover novel and therapeutically effective antimicrobials with different mechanisms from those of previously identified antimicrobials.

Methods

Bacteria, animals, and reagents

The S. aureus RN4220⁴¹, Newman and MSSA1⁴² strains were cultured in tryptic soy broth (TSB; Becton Dickinson and Co., Franklin, NJ, USA) overnight at 37 °C with shaking. Various kinds of serum were purchased from MilliporeSigma (St. Louis, MO, USA), heat-inactivated by treatment at 56 °C for 30 min and stored at −20 °C until use. rhApoA-I was purified as described below and the human ApoA-II was purchased from Calbiochem (Merck Millipore, Billerica, MA, USA). Mice (ICR and C57BL/6J) were obtained from and CLEA Japan, Inc. Apoa1 gene knockout mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA) and bred to the age of 8–12 weeks. Lysocin E was prepared as previously described⁹ and lysocin E derivatives were synthesised as previously reported¹⁰. Lipids, such as PG from egg yolk lecithin, CL from bovine heart and LTA from S. aureus were purchased from MilliporeSigma. The other reagents were purchased from Fujifilm Wako Pure Chemicals (Tokyo, Japan) or MilliporeSigma except where noted.

Measurement of the antimicrobial, bactericidal and protein-binding activities of lysocin E

The antimicrobial activities of the compounds were measured using the microdilution method⁴³. Measurement of the bactericidal activity of lysocin E (0.25 µg ml⁻¹) and nisin (2 µg ml⁻¹) with or without ApoA-I was performed according to the National Committee for Clinical Laboratory Standards method⁴⁴. We chose the concentration of 25 µg ml⁻¹ ApoA-I as the enhancing effect of ApoA-1 on lysocin E was observed in a linear range. For the ∆menA and ∆menB strains, TSB medium was used instead of cation-adjusted-Mueller Hinton broth (CA-MHB) in the microdilution method. Sera and ApoA-I, II were added to the bacterial culture before distributing them into 96-well plates. ApoA-I and ApoA-II antimicrobial activity were evaluated using the RN4220 strain. The protein binding rate of lysocin E against human plasma was determined by a rapid equilibrium dialysis device (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions.

Measurement of membrane-damaging activity

Membrane-damaging activity of lysocin E and nisin with or without 25 µg ml⁻¹ rhApoA-I against S. aureus MSSA1 strain was measured using a LIVE/DEAD™ BacLight™ Bacterial Viability kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. Membrane potential was measured according to a fluorescence-based assay⁹ with slight modification. Briefly, S. aureus MSSA1 or Newman strain was grown in TSB at 37 °C with shaking at 200 r.p.m. overnight. The overnight culture was diluted 100-fold in CA-MHB and grown until the A₆₀₀ reached 1.0. The cells were then collected by centrifugation at 7500 × g for 10 min at 4 °C, washed with HEPES buffer (pH 7.0) containing 5 mM HEPES, 50 mM glucose and 5 mM EDTA, and resuspended in the same buffer to an A₆₀₀ of 0.05. A membrane potential-sensitive dye, 3,3′-dipropylthiadicarbocyanine iodide, was added to the cells at a final concentration of 250 nM, and the cells were incubated in the dark at 37 °C for 30 min. The fluorescence intensities (excitation 622 nm; emission 670 nm), after adding the respective antimicrobials, were acquired using Spectra Manager ver. 1.55.00 on an FP-6200 spectrofluorometer (JASCO, Tokyo, Japan) equipped with a heated chamber at 37 °C after adding the respective antimicrobials. All experiments were performed under minimal light conditions.

Purification of lysocin E antimicrobial activity-enhancing factors from calf serum

Calf serum (10 ml) was added to 15 ml EtOH (60% v/v) and the mixture was centrifuged for 15 min. The collected supernatant was evaporated and suspended in 8 ml Milli-Q water (fraction [fr.] II). Fr. II with formic acid (0.1%) was applied to 10 ml of ODS resin (Waters Corporation, Milford, MA, USA). After washing with 25% EtOH and then 50% EtOH, the resin was eluted with 75% EtOH containing 0.1% formic acid. The eluted sample was evaporated and dissolved in 0.8 ml buffer A (10 mM phosphate buffer [pH 7.1], 6 M urea; Fr. III). Fr. III was applied to the Superose 12 10/300 GL gel filtration column (GE Healthcare) and eluted with the AKTApurifier system (GE Healthcare). The elution was performed in buffer A at a flow rate of 0.5 ml min⁻¹ and 60 fractions (0.5 ml per fr.) were collected in 2-ml tubes. All fractions, except for Fr. I, were dialysed in 10 mM phosphate buffer (pH 7.1) before assessment of their activities. Protein concentrations were measured using Coomassie Plus^TM protein assay reagent (Thermo Fisher) with bovine serum albumin as the standard. Amino acid sequence analysis was performed by in-gel digestion and mass spectrometric analysis⁴⁵. One unit was defined as the activity that decreases the MIC of lysocin E to 1 µg ml⁻¹.

Plasmid construction, expression and purification of ApoA-I

The oligonucleotide sequence of the human and mouse apolipoprotein A-I open-reading frames (ORFs), modified for optimised codon usage in E. coli (Supplementary Fig. 6) according to a previous report⁴⁶, were synthesised at Eurofins Genomics and cloned in the pET20b vector (Novagen). The plasmids, designated pET-hApoWT and pET-mApoWT, were introduced into E. coli BL21(DE3)/pLysS and cells were selected with 100 µg ml⁻¹ ampicillin and 12.5 µg ml⁻¹ chloramphenicol. To construct rhApoA-I variants, we first amplified the two regions within the ORF (nucleotide nos. 1–435 and 481–729) using degenerated primers with linker sequences (Supplementary Table 4) and the pET-hApoWT plasmid as a template. The two products were then connected by another round of PCR and ligated to the EcoRV-digested pET20b vector, resulting in pET-hApoΔM. Next, the internal ORF region (nucleotide nos. 196–729) was amplified by PCR using degenerated primers (Supplementary Table 4) and the pET-hApoΔM plasmid as a template. The PCR product was digested with NdeI and HindIII and then ligated with the pET20b vector linearised with the same restriction enzymes, resulting in pET-hApo∆NM. To construct pET-hApoΔC, the plasmid with C-terminal region-deleted rhApoA-I, we used the primer pair Apo F and ΔC R and pET-hApoWT as a template. Next, we used this template and primers ΔN F and ΔC R to construct pET-hApoΔNC. Colonies were inoculated with NZCYM broth (10 g pancreatic digest casein, 1 g casamino acids, 5 g yeast extract, 5 g NaCl, 1 g magnesium sulfate anhydrous, dissolved in 1 L water and autoclaved at 121 °C for 15 min) in the presence of the above-mentioned antibiotics. After overnight incubation, 0.5 ml of the full-growth cultures were added to 50 ml of the fresh NZCYM broth containing the above-mentioned antibiotics, followed by aerobic incubation at 37 °C to OD₆₀₀ = 0.6. Then, 50 µl of 1 M isopropyl β-d-1-thiogalactopyranoside (final concentration 1 mM) was added and the cultures were further incubated for 3 h. Cells were collected by centrifugation, resuspended in 1xGnPB (50 mM phosphate buffer pH 8, 0.5 M NaCl, 3 M guanidine hydrochloride), sonicated and centrifuged at maximum speed for 10 min. The supernatants were collected and applied to Probond, a nickel-chelating resin (Life Technologies). The column was washed successively with 1xGnPB and 20 mM imidazole in 1xGnPB and then eluted with 200 mM imidazole in 1xGnPB. The eluted fractions were dialysed against 50 mM phosphate buffer (pH 8) and 100 mM NaCl. Approximately 5–6 mg of the recombinant protein was obtained from 50 ml of the initial E. coli culture.

Mouse infection experiments

Mouse protocols followed the Regulations for Animal Care and Use of the University of Tokyo under approval by the Graduate School of Pharmaceutical Science at the University of Tokyo (approval nos. 24–55). Mice were housed in a group of at most five in a cage, kept in a room maintained at 23–24 °C, 55% humidity, and 12 h light/12 h dark cycle. For Table 1, clinically isolated MRSA MR6 strain⁴⁷ was cultured overnight, suspended in 7% type II porcine stomach mucin supplemented with 0.2 mM FeNH₄-citrate and intraperitoneally administered to mice (ICR, female, 4 weeks of age) at 5.7 × 10⁷ colony-forming units. Lysocin E (4–0.25 mg kg⁻¹), vancomycin (8–0.5 mg kg⁻¹), linezolid (8–0.5 mg kg⁻¹), or daptomycin (4–0.25 mg kg⁻¹) were administered subcutaneously to 5 animals/group at 1 h and 6 h after infection. Mouse survival was assessed after 5 days and ED₅₀ values were calculated by multiple logistic regression analysis. For Fig. 1d, the Newman strain was cultured in TSB overnight and cells (4.0 × 10⁷ colony-forming units) were washed in phosphate-buffered saline (PBS) were injected into the tail vein of 8–12-week-old female C57BL/6J wild-type mice or Apoa1 homozygous-deficient mice. After 2 h, lysocin E (0.5 mg kg⁻¹) was administered subcutaneously. After 24 h, the kidneys were harvested and crushed with a BioMasher II (Nippi Incorporated, Tokyo, Japan). The suspensions were diluted with PBS(−), plated on TSB agar medium and the number of colonies was counted.

rhApoA-I pull-down assay

Menaquinone 4 (MK-4; 100 µl of 1.25 mM) was mixed with N-dodecyl β-d-maltoside (250 µl of 100 mM) in the presence or absence of purified lipid II or lipids (300 µl of 70 µM). Lipid II fraction was prepared from S. aureus RN4220 treated with moenomycin, followed by park nucleotide removal⁴⁸. The mixture was dried in vacuo and dissolved in 2 ml of water. The micelles were then filtered through a 0.45-µm polyvinylidene difluoride membrane filter and a 200 µl aliquot of the filtered micelles were used to evaluate the binding of the antimicrobials in the presence of rhApoA-I. Briefly, 5 µg of antimicrobials was added to the aliquots and the mixture was incubated for 15 min on ice in the presence of 4 µg rhApoA-I. The mixture was then transferred to His-Tag magnetic beads (Dynabeads, Thermo Fisher). The bound protein was then eluted with 250 mM imidazole and then eluted antimicrobials were quantified using an ACQUITY ultra-performance liquid chromatography (UPLC) H-Class PLUS attached to Xevo G2-XS QTof system (Waters Corporation). After injecting the sample into a 2.1 × 100 mm Acquity UPLC HSS T3 1.8 μm column (Waters Corporation) equilibrated with 0.3 mL min^–1 of 30% acetonitrile and 0.1% formic acid in water, the column was developed with a linear gradient to 60% acetonitrile and 0.1% formic acid in water over 10 min. The eluate was continuously applied to a Waters Xevo G2-XS QTof mass spectrometer and the spectra were obtained using MassLynx 4.1 (Waters Corporation) in electrospray ionisation (ESI)-positive mode that was analysed by UNIFI Scientific Information System ver. 1.9 (Waters Corporation).

Lysocin E pull-down assay

The pull-down assay for the binding of lysocin E and lipid II was performed using Streptavidin beads (Dynabeads^TM M-280, Thermo Fisher) and 0.05% [v/v] Tween-20 in PBS. The beads were first washed with the buffer and loaded with premixed 2.5 µg biotinylated lysocin in 30 µl of 50 µM lipid II. After mixing and incubating at room temperature for 15 min, the beads were washed twice with the buffer and eluted with 6 M guanidine thiocyanate. Control experiments, without lysocin E, were performed to control for the possibility of nonspecific binding. The amount of lipid II in the eluted sample was quantified using UPLC-MS.

Bio-layer interferometry analysis

Bio-layer interferometry (BLI) measurements were performed on a BLItz system (ForteBio, Fremont, CA, USA) maintained at room temperature. Kinetic analyses were performed using BLItz Pro ver. 1.2.1.5 (ForteBio). The biotinylated ligand, diluted to 0.1 mg ml⁻¹ in BLItz buffer (0.05% [v/v] Tween-20 in PBS), was immobilised on a streptavidin biosensor and the kinetic analysis was performed by running BLItz buffer for 300 s each, in the presence and absence of analyte for association and dissociation, respectively. Biotinylated menaquinone (5) and biotinylated lysocin E derivatives (3 and 4) were synthesised as described in the Supplementary Information (Supplementary Figs. 7–15).

Amount of rhApoA-I binding on the surface of S. aureus

Staphylococcus aureus RN4220 strain was grown overnight in TSB at 37 °C with shaking at 150 r.p.m. The A₆₀₀ was adjusted to 0.5 with CA-MHB, 1 ml of which was treated with 0, 10, 20, or 40 µg ml⁻¹ of rhApoA-I and further incubated at 37 °C with shaking at 150 r.p.m. for 5 min. The cells were then centrifuged at 900 × g at 4 °C for 30 min, washed twice with PBS, and then the pellets were suspended in SDS-PAGE sample buffer. Under these conditions, the S. aureus cells were not lysed. The supernatants were collected, boiled, and applied to 12.5% SDS-PAGE. The gel was stained with Coomassie brilliant blue, and the band intensities were quantified using ImageJ software (1.47v, NIH, USA)⁴⁹.

Liposome disruption assay

We prepared liposomes by making some modifications to the previously described procedure⁹. To 30 µl PG (50 mM, dissolved in methanol/chloroform) contained in a 10 mL round-bottom flask, 7.5 µl of CL (50 mM, dissolved in the same solvent) and 500 µL of chloroform were added. To this mixture, lipids (1.0 mol % MK-4 and/or 0.1 mol % lipid II) were added to prepare liposomes. Control liposomes without lipids were also prepared using the same protocol. A lipid film prepared was rehydrated with 5 mM sodium HEPES (pH 7.5) containing 2.5 mM calcein. The liposome suspension was freeze-thawed five times and filtered through a Sephadex G-50 column with 20 mM sodium HEPES (pH 7.5) containing 1 mM EDTA to remove the free calcein. The obtained liposomes contained 40 ± 10% lipid II. The liposome fraction was diluted 5-fold with buffer, and the increase in fluorescence intensity (excitation: 310 nm; emission: 530 nm) was measured by a spectrofluorometer after adding the antimicrobials. The fluorescence value for 100% disruption was estimated by adding Triton X-100 to each sample.

Quantification of lipid II

The lipid II fraction was prepared from 5 ml cultures as previously described²⁰ and hydrolysed by incubating at 99 °C for 30 min in 100 µl of 10 mM ammonium acetate (pH 4.2). The samples were freeze-dried and dissolved in 30 µl water. The supernatant, after centrifugation at 16,000 × g for 10 min, was analysed by a quantitative liquid chromatography-mass spectrometry. The liquid chromatography-mass spectrometry system comprised an ACQUITY UPLC H-Class PLUS and Xevo G2-XS QTof system and CORTECS UPLC T3 1.6 µm 2.1 × 150 mm columns (Waters Corporation) maintained at 40 °C with a flow rate of 0.3 ml min⁻¹. Solvent A was 0.1% aqueous formic acid and solvent B was 0.1% formic acid in acetonitrile. After 8 µl of the sample was applied to the column equilibrated with solvent A, the column was developed with the same solvent for 2 min, followed by a linear gradient of 100–80% of A over 15 min. Mass spectra were collected in ESI-negative mode using MassLynx 4.1 (Waters Corporation) and analysed by UNIFI Scientific Information System (ver. 1.9). The peak area of delipidated lipid II (Supplementary Fig. 1d; 7) to assess the exact mass m/z 664.255 [M-2H]²⁻ (calculated mass m/z 664.256 [M-2H]²⁻) was quantified and the structure was confirmed by the fragment pattern generated from MS^E analysis using the UNIFI Scientific Information System. To evaluate the lipid II purity of the fraction, we purified lipid II referring to a previous report⁵⁰. Briefly, a lipid II fraction prepared as above was dissolved in 75% methanol with 25% 50 mM ammonium bicarbonate and separated using an Alliance HPLC system with a PDA detector 2998 and Empower 3 software (Waters Corporation) and PEGASIL C4 SP100 column (4.6 × 250 mm, Senshu Scientific, Tokyo, Japan). Solvent A was 50 mM ammonium bicarbonate and solvent B was 100% methanol. After injecting the lipid II fraction prepared as above, the column was developed with a linear gradient of 85–90% of B over 15 min, followed by 90–100% of B over 5 min, and isocratic 100% B for 5 min. The lipid II peak appeared at 11.3 min and was collected. The purity of this fraction was analysed by developing 70–100% of A over 30 min and isocratic 100% B for 10 min and found to be >95% (Supplementary Fig. 16a), and the structure was confirmed by in-fusion MS/MS (Supplementary Fig. 16b, c).

Statistical analysis

Statistical analysis was performed using GraphPad Prism 8.4.3 (GraphPad Software, San Diego, CA, USA). P values of 0.05 or less were considered significant.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request. Source data are provided with this paper.

References

Dalhoff, A. Seventy-five years of research on protein binding. Antimicrob. Agents Chemother. 62, e01663–01617 (2018).
Article PubMed PubMed Central Google Scholar
Pruul, H. & McDonald, P. J. Potentiation of antibacterial activity of azithromycin and other macrolides by normal human serum. Antimicrob. Agents Chemother. 36, 10–16 (1992).
Article CAS PubMed PubMed Central Google Scholar
Safdar, N., Andes, D. & Craig, W. A. In vivo pharmacodynamic activity of daptomycin. Antimicrob. Agents Chemother. 48, 63–68 (2004).
Article CAS PubMed PubMed Central Google Scholar
Hegde, S. S. et al. Pharmacodynamics of telavancin (TD-6424), a novel bactericidal agent, against gram-positive bacteria. Antimicrob. Agents Chemother. 48, 3043–3050 (2004).
Article CAS PubMed PubMed Central Google Scholar
Hamamoto, H. et al. Quantitative evaluation of the therapeutic effects of antibiotics using silkworms infected with human pathogenic microorganisms. Antimicrob. Agents Chemother. 48, 774–779 (2004).
Article CAS PubMed PubMed Central Google Scholar
Panthee, S., Paudel, A., Hamamoto, H. & Sekimizu, K. Advantages of the silkworm as an animal model for developing novel antimicrobial agents. Front. Microbiol. 8, 373 (2017).
Article PubMed PubMed Central Google Scholar
Fujiyuki, T., Imamura, K., Hamamoto, H. & Sekimizu, K. Evaluation of therapeutic effects and pharmacokinetics of antibacterial chromogenic agents in a silkworm model of Staphylococcus aureus infection. Drug Discov. Ther. 4, 349–354 (2010).
CAS PubMed Google Scholar
Hamamoto, H., Horie, R. & Sekimizu, K. Pharmacokinetics of anti-infectious reagents in silkworms. Sci. Rep. 9, 9451 (2019).
Article ADS PubMed PubMed Central CAS Google Scholar
Hamamoto, H. et al. Lysocin E is a new antibiotic that targets menaquinone in the bacterial membrane. Nat. Chem. Biol. 11, 127–133 (2015).
Article CAS PubMed Google Scholar
Kaji, T. et al. Total synthesis and functional evaluation of fourteen derivatives of lysocin E: importance of cationic, hydrophobic, and aromatic moieties for antibacterial activity. Chem. Eur. J. 22, 16912–16919 (2016).
Article CAS PubMed Google Scholar
Itoh, H. et al. Development of a high-throughput strategy for discovery of potent analogues of antibiotic lysocin E. Nat. Commun. 10, 2992 (2019).
Article ADS PubMed PubMed Central CAS Google Scholar
Beck, W. H. et al. Apolipoprotein A-I binding to anionic vesicles and lipopolysaccharides: role for lysine residues in antimicrobial properties. Biochim. Biophys. Acta 1828, 1503–1510 (2013).
Article CAS PubMed PubMed Central Google Scholar
Motizuki, M., Itoh, T., Yamada, M., Shimamura, S. & Tsurugi, K. Purification, primary structure, and antimicrobial activities of bovine apolipoprotein A-II. J. Biochem. 123, 675–679 (1998).
Article CAS PubMed Google Scholar
Williamson, R., Lee, D., Hagaman, J. & Maeda, N. Marked reduction of high density lipoprotein cholesterol in mice genetically modified to lack apolipoprotein A-I. Proc. Natl Acad. Sci. USA 89, 7134–7138 (1992).
Article ADS CAS PubMed PubMed Central Google Scholar
Hedrick, C. C., Castellani, L. W., Warden, C. H., Puppione, D. L. & Lusis, A. J. Influence of mouse apolipoprotein A-II on plasma lipoproteins in transgenic mice. J. Biol. Chem. 268, 20676–20682 (1993).
Article CAS PubMed Google Scholar
Li, H., Reddick, R. L. & Maeda, N. Lack of apoA-I is not associated with increased susceptibility to atherosclerosis in mice. Arterioscler. Thromb. 13, 1814–1821 (1993).
Article CAS PubMed Google Scholar
Osland, A. Binding of human serum high density lipoprotein to Staphylococcus capitis. Acta Pathol. Microbiol. Immunol. Scand. B 93, 371–376 (1985).
CAS PubMed Google Scholar
Ichimura, N. et al. High-density lipoprotein binds to Mycobacterium avium and affects the infection of THP-1 macrophages. J. Lipids 2016, 4353620 (2016).
Article PubMed PubMed Central CAS Google Scholar
Jiao, Y. L. & Wu, M. P. Apolipoprotein A-I diminishes acute lung injury and sepsis in mice induced by lipoteichoic acid. Cytokine 43, 83–87 (2008).
Article CAS PubMed Google Scholar
Santiago, M. et al. Genome-wide mutant profiling predicts the mechanism of a Lipid II binding antibiotic. Nat. Chem. Biol. 14, 601–608 (2018).
Article CAS PubMed PubMed Central Google Scholar
Surewicz, W. K., Epand, R. M., Pownall, H. J. & Hui, S. W. Human apolipoprotein A-I forms thermally stable complexes with anionic but not with zwitterionic phospholipids. J. Biol. Chem. 261, 16191–16197 (1986).
Article CAS PubMed Google Scholar
Fox, C. A., Ellison, P., Ikon, N. & Ryan, R. O. Calcium-induced transformation of cardiolipin nanodisks. Biochim. Biophys. Acta 1861, 1030–1036 (2019).
Article CAS Google Scholar
Biedzka-Sarek, M. et al. Apolipoprotein A-I exerts bactericidal activity against Yersinia enterocolitica serotype O:3. J. Biol. Chem. 286, 38211–38219 (2011).
Article CAS PubMed PubMed Central Google Scholar
Tanaka, M. et al. Contributions of the N- and C-terminal helical segments to the lipid-free structure and lipid interaction of apolipoprotein A-I. Biochemistry 45, 10351–10358 (2006).
Article CAS PubMed Google Scholar
Dosler, S. & Gerceker, A. A. In vitro activities of antimicrobial cationic peptides; melittin and nisin, alone or in combination with antibiotics against Gram-positive bacteria. J. Chemother. 24, 137–143 (2012).
Article CAS PubMed Google Scholar
Martinez, B. et al. Specific interaction of the unmodified bacteriocin Lactococcin 972 with the cell wall precursor lipid II. Appl. Environ. Microbiol. 74, 4666–4670 (2008).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhao, Y. et al. Apolipophorin-II/I contributes to cuticular hydrocarbon transport and cuticle barrier construction in Locusta migratoria. Front. Physiol. 11, 790 (2020).
Article PubMed PubMed Central Google Scholar
Hanada, Y., Sekimizu, K. & Kaito, C. Silkworm apolipophorin protein inhibits Staphylococcus aureus virulence. J. Biol. Chem. 286, 39360–39369 (2011).
Article CAS PubMed PubMed Central Google Scholar
Malik, Z. A. & Amir, S. An apolipophorin III protein from the hemolymph of desert locust, Schistocerca gregaria. Appl. Biochem. Biotechnol. 165, 1779–1788 (2011).
Article CAS PubMed Google Scholar
Omae, Y., Hanada, Y., Sekimizu, K. & Kaito, C. Silkworm apolipophorin protein inhibits hemolysin gene expression of Staphylococcus aureus via binding to cell surface lipoteichoic acids. J. Biol. Chem. 288, 25542–25550 (2013).
Article CAS PubMed PubMed Central Google Scholar
Yu, H. Z. et al. Molecular characterisation of Apolipophorin-III gene in Samia cynthia ricini and its roles in response to bacterial infection. J. Invertebr. Pathol. 159, 61–70 (2018).
Article CAS PubMed Google Scholar
Ichihashi, N., Kurokawa, K., Matsuo, M., Kaito, C. & Sekimizu, K. Inhibitory effects of basic or neutral phospholipid on acidic phospholipid-mediated dissociation of adenine nucleotide bound to DnaA protein, the initiator of chromosomal DNA replication. J. Biol. Chem. 278, 28778–28786 (2003).
Article CAS PubMed Google Scholar
Mei, X. & Atkinson, D. Lipid-free apolipoprotein A-I structure: insights into HDL formation and atherosclerosis development. Arch. Med. Res. 46, 351–360 (2015).
Article CAS PubMed PubMed Central Google Scholar
Melchior, J. T. et al. A consensus model of human apolipoprotein A-I in its monomeric and lipid-free state. Nat. Struct. Mol. Biol. 24, 1093–1099 (2017).
Article CAS PubMed PubMed Central Google Scholar
Malin, J. J. & de Leeuw, E. Therapeutic compounds targeting Lipid II for antibacterial purposes. Infect. Drug Resist. 12, 2613–2625 (2019).
Article CAS PubMed PubMed Central Google Scholar
Ling, L. L. et al. A new antibiotic kills pathogens without detectable resistance. Nature 517, 455–459 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Parmar, A. et al. Design and syntheses of highly potent teixobactin analogues against Staphylococcus aureus, Methicillin-resistant Staphylococcus aureus (MRSA), and Vancomycin-resistant Enterococci (VRE) in Vitro and in Vivo. J. Med. Chem. 61, 2009–2017 (2018).
Article CAS PubMed Google Scholar
Ioannou, P. et al. Albumin enhances caspofungin activity against Aspergillus species by facilitating drug delivery to germinating hyphae. Antimicrob. Agents Chemother. 60, 1226–1233 (2015).
Article PubMed CAS Google Scholar
Heesterbeek, D. A. C. et al. Complement-dependent outer membrane perturbation sensitizes Gram-negative bacteria to Gram-positive specific antibiotics. Sci. Rep. 9, 3074 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Giamarellos-Bourboulis, E. J. et al. Ex vivo synergy of arachidonate-enriched serum with ceftazidime and amikacin on multidrug-resistant Pseudomonas aeruginosa. J. Antimicrob. Chemother. 51, 423–426 (2003).
Article CAS PubMed Google Scholar
Peng, H. L., Novick, R. P., Kreiswirth, B., Kornblum, J. & Schlievert, P. Cloning, characterization, and sequencing of an accessory gene regulator (agr) in Staphylococcus aureus. J. Bacteriol. 170, 4365–4372 (1988).
Article CAS PubMed PubMed Central Google Scholar
Akimitsu, N. et al. Increase in resistance of methicillin-resistant Staphylococcus aureus to beta-lactams caused by mutations conferring resistance to benzalkonium chloride, a disinfectant widely used in hospitals. Antimicrob. Agents Chemother. 43, 3042–3043 (1999).
Article CAS PubMed PubMed Central Google Scholar
Clinical and Laboratory Standards Institute. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically; Approved Standard—Eighth edition CLSI Document M07-A8 (Clinical and Laboratory Standards Institute, 2009).
National Committee for Clinical Laboratory Standards. Methods for Determining Bactericidal Activity of Antimicrobial Agents; Approved Guideline NCCLS Document M26-A (National Committee for Clinical Laboratory Standards, 1999).
Ishii, K., Adachi, T., Hamamoto, H. & Sekimizu, K. Serratia marcescens suppresses host cellular immunity via the production of an adhesion-inhibitory factor against immunosurveillance cells. J. Biol. Chem. 289, 5876–5888 (2014).
Article CAS PubMed PubMed Central Google Scholar
Ryan, R. O., Forte, T. M. & Oda, M. N. Optimized bacterial expression of human apolipoprotein A-I. Protein Expr. Purif. 27, 98–103 (2003).
Article CAS PubMed Google Scholar
Ishii, K. et al. Phenotypic and genomic comparisons of highly vancomycin-resistant Staphylococcus aureus strains developed from multiple clinical MRSA strains by in vitro mutagenesis. Sci. Rep. 5, 17092 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Qiao, Y. et al. Lipid II overproduction allows direct assay of transpeptidase inhibition by beta-lactams. Nat. Chem. Biol. 13, 793–798 (2017).
Article CAS PubMed PubMed Central Google Scholar
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012).
Article CAS PubMed PubMed Central Google Scholar
VanNieuwenhze, M. S. et al. The first total synthesis of lipid II: the final monomeric intermediate in bacterial cell wall biosynthesis. J. Am. Chem. Soc. 124, 3656–3660 (2002).
Article CAS PubMed Google Scholar
Rybak, M. J. The pharmacokinetic and pharmacodynamic properties of vancomycin. Clin. Infect. Dis. 42, S35–S39 (2006).
Article CAS PubMed Google Scholar
Tsuji, Y. et al. Population pharmacokinetics and pharmacodynamics of linezolid-induced thrombocytopenia in hospitalized patients. Br. J. Clin. Pharmacol. 83, 1758–1772 (2017).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

This work was supported by JSPS KAKENHI Grant Numbers 19K07140JP, 26102714 and 24689008, the Mochida Memorial Foundation for Medical and Pharmaceutical Research and the Takeda Science Foundation (to H.H.), and in part by JSPS KAKENHI Grant Numbers 20K16253, 21H02733 and 15H05783 and Drug Discovery Support Promotion Project from AMED: Japan Agency for Medical Research and Development to K.S., A.M. and H.H., and JP17F17421, TBRF and IFO fellowships to S.P. and K.S. We thank the Genome Pharmaceutical Institute Co., Ltd for preparing the lysocin E and silkworms.

Author information

These authors contributed equally: Hiroshi Hamamoto, Suresh Panthee.

Authors and Affiliations

Teikyo University Institute of Medical Mycology, Tokyo, Japan
Hiroshi Hamamoto & Atsushi Miyashita
Division of Sport and Health Science, Graduate School of Medical Care and Technology, Teikyo University, Tokyo, Japan
Hiroshi Hamamoto
Drug Discoveries by Silkworm Models, Faculty of Pharma-Science, Teikyo University, Tokyo, Japan
Suresh Panthee & Kazuhisa Sekimizu
International Institute for Zoonosis Control, Hokkaido University, Sapporo, Japan
Atmika Paudel
Department of Biological Sciences, Graduate School of Science, The University of Tokyo, Tokyo, Japan
Kenichi Ishii
Department of Biochemistry, Faculty of Pharmaceutical Sciences, Doshisha Women’s College of Liberal Arts, Kyoto, Japan
Jyunichiro Yasukawa
National Marine Environmental Monitoring Center, Dalian, China
Jie Su
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Tokyo, Japan
Hiroaki Itoh, Kotaro Tokumoto & Masayuki Inoue
Genome Pharmaceuticals Institute, Ltd, Tokyo, Japan
Kazuhisa Sekimizu

Authors

Hiroshi Hamamoto
View author publications
You can also search for this author in PubMed Google Scholar
Suresh Panthee
View author publications
You can also search for this author in PubMed Google Scholar
Atmika Paudel
View author publications
You can also search for this author in PubMed Google Scholar
Kenichi Ishii
View author publications
You can also search for this author in PubMed Google Scholar
Jyunichiro Yasukawa
View author publications
You can also search for this author in PubMed Google Scholar
Jie Su
View author publications
You can also search for this author in PubMed Google Scholar
Atsushi Miyashita
View author publications
You can also search for this author in PubMed Google Scholar
Hiroaki Itoh
View author publications
You can also search for this author in PubMed Google Scholar
Kotaro Tokumoto
View author publications
You can also search for this author in PubMed Google Scholar
Masayuki Inoue
View author publications
You can also search for this author in PubMed Google Scholar
Kazuhisa Sekimizu
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

H.H. conceived the idea. H.H and S.P. designed the experiments. S.P. analysed the interactions between ApoA-I, lipid II, and lysocin E. A.P., J.S., K.I., and J.Y. purified the ApoA-I and assayed the antimicrobial activity. H.H., S.P. and A.P. performed the membrane disruption assay. H.I. and K.T. synthesised the lysocin E derivatives and performed the structural analysis of lipid II. M.I. and K.S. provided critical discussion for writing the manuscript. A.M. purified lipid II and performed the liposome disruption assay. H.H., S.P., and A.P. wrote the manuscript. All authors contributed critically to the interpretation of the results and gave final approval for publication.

Corresponding author

Correspondence to Kazuhisa Sekimizu.

Ethics declarations

Competing interests

K.S. is a consultant for Genome Pharmaceutical Institute Co., Ltd. The remaining authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Martin McGavin, Paul Weers and the other, anonymous, reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Hamamoto, H., Panthee, S., Paudel, A. et al. Serum apolipoprotein A-I potentiates the therapeutic efficacy of lysocin E against Staphylococcus aureus. Nat Commun 12, 6364 (2021). https://doi.org/10.1038/s41467-021-26702-0

Download citation

Received: 12 November 2020
Accepted: 07 October 2021
Published: 04 November 2021
DOI: https://doi.org/10.1038/s41467-021-26702-0

This article is cited by

Peanut triacylglycerols activate innate immunity both in insects and mammals
- Wenyuan Li
- Atsushi Miyashita
- Kazuhisa Sekimizu
Scientific Reports (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.