Lipid-Functionalized Single-Walled Carbon Nanotubes as Probes for Screening Cell Wall Disruptors

Abstract

Membrane-active molecules are of great importance to drug delivery and antimicrobials applications. While the ability to prototype new membrane-active molecules has improved greatly with the advent of automated chemistries and rapid biomolecule expression techniques, testing methods are still limited by throughput, cost, and modularity. Existing methods suffer from feasibility constraints of working with pathogenic living cells and by intrinsic limitations of model systems. Herein, we demonstrate an abiotic sensor that uses semiconducting single-walled carbon nanotubes (SWCNTs) as near-infrared fluorescent transducers to report membrane interactions. This sensor is composed of SWCNTs aqueously suspended in lipid, creating a cylindrical, bilayer corona; these SWCNT probes are very sensitive to solvent access (changes in permittivity) and thus report morphological changes to the lipid corona by modulation of fluorescent signals, where binding and disruption are reported as brightening and attenuation, respectively. This mechanism is first demonstrated with chemical and physical membrane-disruptive agents, including ethanol and sodium dodecyl sulfate, and application of electrical pulses. Known cell-penetrating and antimicrobial peptides are then used to demonstrate how the dynamic response of these sensors can be deconvoluted to evaluate different parallel mechanisms of interaction. Last, SWCNTs functionalized in several different bacterial lipopolysaccharides (Pseudomonas aeruginosa, Klebsiella pneumoniae, and Escherichia coli) are used to evaluate a panel of known membrane-disrupting antimicrobials to demonstrate that drug selectivity can be assessed by suspension of SWCNTs with different membrane materials.

Keywords: cell proxy; disruption; membrane; nanosensor; screening techniques.

Figure 1.

(a) Scheme of transduction mechanism. Phospholipid functionalized SWCNT (i) are exposed to membrane-interacting molecules which can then (ii) bind to the surface of the membrane and subsequently (iii) diffuse through the membrane or (iv) self-assemble and form pores. (b) Sample sensor response (shaded region indicates one standard deviation, n = 4). Spike in signal seen during addition of disruptor occurs when the fluorimeter is uncovered and exposed to ambient light.

Figure 2.

(a) Temporal response of PC-SWCNT to addition of ethanol. (b) Temporal response of phosphatidylcholine-SWCNT to addition of sodium dodecyl sulfate. Shaded regions in (a) and (b) indicate a single standard deviation (n = 4). (c) Temporal response of LPS (E. coli serotype O127)-SWCNT to electrical fields, applied every 60 s.

Figure 3.

Temporal sensor responses of (a) phosphatidylcholine-SWCNT and (b) LPS-SWCNT to the TAT peptide. Shaded regions indicate a single standard deviation (n = 4). Disrupter concentrations correspond to values after addition. Responses with large error values were omitted for clarity. Final signal changes (signal at last timepoints) observed at different concentrations of TAT peptide for (c) phosphatidylcholine-SWCNT and (d) LPS-SWCNT. Error bars indicate a single standard deviation (n = 4).

Figure 4.

Temporal sensor responses of (a) phosphatidylcholine-SWCNT and (b) LPS-SWCNT to the crotamine-derived amphipathic peptide. Disrupter concentrations correspond to values after addition. Shaded regions indicate a single standard deviation (n = 4). Responses with large error values were omitted for clarity. Final signal changes (signal at last timepoints) observed at different concentrations of amphipathic peptide for (c) phosphatidylcholine-SWCNT and (d) LPS-SWCNT. Error bars indicate a single standard deviation (n = 4).

Figure 5.

(a) Sample sensor response separated into different response components: (b) fast-kinetics sensor response and (c) slow-kinetics sensor response. (d) Slow-kinetics sensor response plotted against a Langmuir isotherm, R-squared(adj) = 0.8373. X-axis term corresponds to fractional coverage of available membrane binding sites (K_d = 12 μg/mL). Error bars and shaded regions indicate a single standard deviation (n = 4).

Figure 6.

Heat maps indicating (a) total, (b) fast-kinetics, and (c) slow-kinetics sensor responses (% signal change) with respect to disruptor and disruptor concentrations for (i) Pseudomonas aeruginosa, (ii) Klebsiella pneumoniae, and (iii) Escherichia coli (O26) LPS-SWCNT.