Biophysical studies of the interaction of squalamine and other cationic amphiphilic molecules with bacterial and eukaryotic membranes: importance of the distribution coefficient in membrane selectivity

https://doi.org/10.1016/j.chemphyslip.2009.10.006Get rights and content

Abstract

The interaction of squalamine (SQ) with eukaryotic and prokaryotic membranes was studied and compared with the interaction of two other cationic amphipathic antimicrobials (CAAs), i.e. the antibiotic polymyxin B (PMB) and the detergent hexadecyltrimethylammonium bromide (CTAB). Whole cell experiments showed that the three CAA have in common the ability to interact with lipopolysaccharide-containing membranes through a divalent cation sensitive process. Differences were found regarding their kinetics of membrane permeabilisation and their selectivity for bacteria, with a preferential permeabilisation of bacteria by PMB > SQ and no selectivity for CTAB. Experiments with lipid monolayers and bilayers showed that this selectivity did not correlate with a preferential interaction of the CAAs with lipids but rather relies on differences in their ability to penetrate lipid bilayers and to cause electrically active lesions. Incidentally, our results also suggest that the distribution coefficient of CAAs could be used to predict their selectivity for bacteria.

Introduction

Emergence of pathogenic bacteria exhibiting resistance against multiple conventional antibiotics has led to the development of new antimicrobial molecules. Among such molecules, cationic amphiphilic (amphipathic) antimicrobials (CAAs) seem to be attractive candidates (Savage et al., 2002).

CAAs are structurally unrelated molecules that are positively charged and amphipathic. CAAs include cationic surfactants, cyclic lipopeptides, β-sheet and α-helix-forming peptides and cholesterol derivatives (Savage et al., 2002). The antibacterial activity of CAAs relies on their ability to interact with bacterial lipids, especially lipopolysaccharide (LPS) and phosphatidylglycerol (PG), resulting in: (i) the release of divalent cations bridging LPS molecules together, (ii) a change in the outer membrane permeability barrier, (iii) the release of membrane vesicle, (iv) membrane lipid exchange, which in turn causes membrane permeabilisation and bacteria killing (Katsu et al., 1984, Katsu et al., 2002, Savage et al., 2002, Clausell et al., 2004, Clausell et al., 2005, Clausell et al., 2007, McBroom and Kuehn, 2007). Binding of CAAs to bacterial membranes involves both electrostatic and hydrophobic interactions (Savage et al., 2002). The conserved cationic charge of CAAs allows them to bind to the anionic charge of lipopolysaccharide (LPS) and/or phosphatidylglycerol (PG) exposed at the surface of the bacterial membrane. Such electrostatic interactions, sufficient to cause the release of divalent cations, could explain the selectivity of CAAs for the bacterial membrane over the eukaryotic membrane mainly formed of neutral lipids at physiological pH (phosphatidylcholine and cholesterol) although negatively charged lipids and proteins are also represented (Salmi et al., 2008a, Savage et al., 2002).

Although required, the cationic charge is not sufficient per se to cause either the permeabilisation of bacterial membranes or the bacteria killing (Katsu et al., 2002, Savage et al., 2002). In fact, the hydrophobic moiety of CAAs is responsible for their insertion into the hydrophobic core of the bacterial lipid bilayer, leading to membrane permeabilisation and eventually to CAAs-mediated bacterial killing as convincingly indicated by the lack of membrane permeabilisation of a deacylated polymyxin analogue (Vaara and Vaara, 1983, Tsubery et al., 2000, Tsubery et al., 2001, Tsubery et al., 2002, Katz et al., 2003, Clausell et al., 2005, Clausell et al., 2007).

Based on similarities with their molecular target (i.e. LPS) and of their physicochemical properties (i.e. cationic detergent), one can roughly extrapolate that all CAAs interact in a similar way with biological membranes. However, this assumption is unlikely to be correct due to the structural diversity of their hydrophobic domain, the number of cationic groups and/or the variations in their polar:apolar ratio. Surprisingly, studies focusing on the comparison of the activities of various CAAs in the same membrane system are scarce (Ding et al., 2001, Xiong et al., 2005) although such studies are critical to understand the mechanism of action of CAAs on biological membranes and to develop more efficient analogues.

Among CAAs, squalamine (SQ), a steroid derivative initially isolated from dogfish shark (Moore et al., 1993), has attracted high interest for a broad range of researchers. SQ is currently used for the treatment of various human pathologies, such as macular degeneration or cancer-associated angiogenesis (Brunel et al., 2005). Moreover, SQ has been proposed as a potential new class of antimicrobial agents against multidrug resistant Gram-negative and Gram-positive bacteria (Moore et al., 1993, Ding et al., 2001, Savage and Li, 2000, Savage et al., 2002, Brunel et al., 2005, Salmi et al., 2008a, Salmi et al., 2008b).

Whereas the interaction of other CAAs with biological membranes has been the subject of numerous studies, particularly with polymyxin B (PMB) (Chen and Feingold, 1972, Clausell et al., 2004, Clausell et al., 2005, Clausell et al., 2007, Hartmann et al., 1978, Katsu et al., 1984, Katsu et al., 2002, Katz et al., 2003, Sixl and Galla, 1981, Sixl and Galla, 1982, Tsubery et al., 2000, Tsubery et al., 2001, Tsubery et al., 2002, Wiese et al., 1998, Wiese et al., 2003), relatively little information is available regarding the interaction of SQ with cellular membranes (Chen et al., 2006, Salmi et al., 2008a, Selinsky et al., 1998, Selinsky et al., 2000).

In the present study, we have investigated the interaction of SQ with prokaryotic and eukaryotic membranes (whole cells, lipid monolayers and lipid bilayers) and compared it with two other CAAs, i.e. the antimicrobial polymyxin B (PMB) and the cationic detergent hexadecyltrimethylammonium bromide (CTAB).

Our data shed some light on the mechanism of interaction of SQ with biological membranes and demonstrate that, although CAAs share common biophysical properties, they possess some specificity of interaction that depends on the nature of the biological membrane studied. Most importantly, our results suggest that the distribution coefficient (log D) could be a valuable indicator of the CAAs selectivity for bacteria over eukaryotic cells, a required condition to guarantee the innocuousness of an antimicrobial used in human/animal medicine.

Section snippets

Cell culture and bacteria

Escherichia coli (ATCC No. 54127) were maintained on Luria–Bertani (LB) agar plate at 37 °C. The day before experiments, single colonies were collected and grown over-night in LB solution at 37 °C. Bacterial number was then evaluated by measuring the optical density (OD) of the LB suspension at 600 nm (OD600 nm of 1 corresponding to approximately to 109 bacteria/mL as determined by colony counting after bacteria plating). Murine B lymphoma (Wehi-231, ATCC No. CRL-1702) were grown as cell suspension

Structure and biophysical properties of SQ, CTAB and PMB

We compared the structure and the biophysical properties of SQ, CTAB and PMB (Fig. 1 and Table 1). The three molecules have in common a charged polar domain and an apolar domain. However they differ in both the number of charges and the chemical structure of the apolar moiety. The hydrophobic domain of SQ (Fig. 1A), corresponds to a cholestane structure whereas the polar part consists of three cationic charges, due to the polyamine spermidine, and an anionic charge linked to a sulphate group,

Discussion

The cholesterol derivative squalamine, one of the members of CAAs, has been proposed as new drug for the treatment of macular degeneration and cancer and as a new class of antibiotics (Brunel et al., 2005, Salmi et al., 2008a, Savage and Li, 2000, Savage et al., 2002). However, the mechanism of interaction of SQ with prokaryotic and eukaryotic membranes remains mostly unknown. In particular, no comparison has been made between the interaction of SQ and other CAAs with biological membranes. In

Acknowledgments

We acknowledge Pr M. Zasloff from Georgetown University, Washington (USA) providing us squalamine. We thank Henri Chahinian and Nicolas Garmy for helpful discussions and stimulating advice throughout this study. We also thank Sabine Quitard for her help with the manuscript. This work has been supported by academic funds from University Paul Cezanne, the Centre National de la Recherche Scientifique (CNRS) and a grant (FAN-BCL) from the Institut National du Cancer (INCA).

References (39)

  • W.H. Chen et al.

    A bioconjugate approach toward squalamine mimics: insight into the mechanism of biological action

    Bioconjug. Chem.

    (2006)
  • A. Clausell et al.

    Polymyxin B–lipid interactions in Langmuir–Blodgett monolayers of Escherichia coli lipids: a thermodynamic and atomic force microscopy study

    Biopolymers

    (2004)
  • A. Clausell et al.

    Synthesis and membrane action of polymyxin B analogues

    Luminescence

    (2005)
  • A. Clausell et al.

    Gram-negative outer and inner membrane models: insertion of cyclic cationic lipopeptides

    J. Phys. Chem. B

    (2007)
  • B. Ding et al.

    Correlation of the antibacterial activities of cationic peptide antibiotics and cationic steroid antibiotics

    J. Med. Chem.

    (2001)
  • E. Di Pasquale et al.

    The insertion and transport of anandamide in synthetic lipid membranes are both cholesterol-dependent

    PLoS ONE

    (2009)
  • J. Fantini et al.

    Infection of colonic epithelial cell lines by type 1 human immunodeficiency virus is associated with cell surface expression of galactosylceramide, a potential alternative gp120 receptor

    Proc. Natl. Acad. Sci. U.S.A.

    (1993)
  • T. Katsu et al.

    Temperature dependence of action of polymyxin B on Escherichia coli

    J. Biochem.

    (1984)
  • T. Katsu et al.

    Interaction between polyamines and bacterial outer membranes as investigated with ion-selective electrodes

    Antimicrob. Agents Chemother.

    (2002)
  • Cited by (32)

    • Aurone derivatives as promising antibacterial agents against resistant Gram-positive pathogens

      2019, European Journal of Medicinal Chemistry
      Citation Excerpt :

      t-Test and two way ANOVA analyses were used to address the significant differences between mean values with significance set at p < 0.05. Bacterial membrane permeabilization was evaluated using the cell-impermeable DNA/RNA probe propidium iodide as previously explained [24,29]. Logarithmic growing bacterial suspension was prepared from overnight bacterial suspension by 1 in 10 dilution.

    • Negatively charged lipids as a potential target for new amphiphilic aminoglycoside antibiotics: A biophysical study

      2016, Journal of Biological Chemistry
      Citation Excerpt :

      The fluorescence intensity was measured with a SpectraMax® M3 microplate reader at 25 °C after 15 min of stabilization (Molecular Devices, Sunnyvale, CA) for excitation and emission wavelengths of 540 and 610 nm, respectively. The data were normalized based on the fluorescence intensity measured in the presence of either 150 μm quaternary ammonium hexadecyltrimethylammonium bromide (positive control, 100%)(38) or imipenem (negative control, 0%). We investigated the depolarization of bacterial membranes and LUVs by using the membrane potential-sensitive fluorescent probe DiSC3(5).

    • A <sup>2</sup>H solid-state NMR study of the effect of antimicrobial agents on intact Escherichia coli without mutating

      2013, Biochimica et Biophysica Acta - Biomembranes
      Citation Excerpt :

      New antimicrobial agents are required to act against the emergence of pathogenic bacteria that are resistant to the commonly used antibiotics. Principally designed from less than ten types of molecular scaffolds [1,2], their action mechanism targets few cellular processes such as cell wall and protein biosynthesis as well as DNA replication and repair. Antibiotic resistance is due to their excessive consumption by human patients but could also be related to the widespread routine feeding of antibiotics to livestock [3].

    View all citing articles on Scopus
    View full text