Due to the misuse and overuse of conventional antibiotics, resistant infections are on the rise [1
]. Those include well-known bacterial strains such as Staphylococcus aureus
], Streptococcus pneumoniae
] and Mycobacterium tuberculosis
]. It is predicted that approximately 10 million people will die from resistant infections by the year 2050 [5
], not considering the emergence of new resistant strains. This mortality prediction exceeds that of cancer and diabetes combined. In response to this crisis, new antibiotics are being developed [6
]; however, offering only a temporary solution and, rather, giving rise to multi-drug resistant infections [8
]. With mutated bacterial infections being the foundation for many large-scale health issues including M. tuberculosis
, MRSA and VRE (Vancomycin-resistant Enterococci) [9
], the crisis of antibiotic resistance, as well as the need in enhancing antibiotic efficacy becomes a global issue.
To date, very few routes were explored to address the antibiotic resistance including the inhibition of mutation [11
], change in the dosing regimen of existing antibiotics [12
] and delivery-assisted combination treatments [13
]. However, dosing strategies provide only a temporary solution that delays the formation of resistance, while inhibitory approaches have to be applied to bacteria prior to mutation that renders the antibiotics ineffective. With the few successful delivery-assisted attempts to address the existing antibiotic resistance, current studies still focus more on the development of new antibacterial strategies [16
]. In this work, an alternative multimodal approach to the issue at hand is explored. Although, microbial growth inhibition by carbon nanotubes platform was shown before [20
], in this work we propose novel non-covalent formulation of existing antibiotics with single-walled carbon nanotubes (SWCNTs) for delivery, imaging, and enhanced antibiotic efficacy.
For the context of this study, the SWCNT-antibiotic dispersions are tested against Staphylococcus epidermidis
. Because of the increased use of biomaterials in the hospital and clinical environment, S. epidermidis
has become one of the five most common bacteria to cause nosocomial infections on prosthetic parts, valves, surgical wounds, urinary tract or bone marrow transplants. While already causing nearly one million infections and many deaths per year, S. epidermidis
has become resistant to a wide scope of antibiotics [21
]. Strains of S. epidermidis
are resistant to methicillin, penicillin, penems, carbapanems, and cephalosporins [22
]. With these being the most commonly used antibiotics, an increase in S. epidermidis
infections becomes a big threat [23
Biocompatible lipid-based carriers are known to yield high encapsulation efficiency for guest drug molecules [24
]. On the other hand, SWCNTs offer great promise as antibiotic delivery vehicles due to their unique physical and optical properties. Known for their quasi-one-dimensional structure, SWCNTs have the dimensions suitable for cellular internalization [29
] show low cytotoxicity when formulated [31
] and accumulate partly in actin cytoskeleton but exhibit excretion over time [30
]. Additionally, a significant amount of SWCNTs can be loaded into a target cell [32
], making them suitable for the delivery of hydrophobic drugs and gene therapies sensitive to degradation in blood. So far, SWCNTs have successfully delivered siRNA to cancer cells [33
] and tissues [34
] and such anticancer drugs as cisplatin, methotrexate, and doxorubicin [36
]. SWCNTs also show a potential for antibiotic delivery as multiple antibiotics are known to adsorb well on SWCNT surface [40
] and with covalent attachment improve the efficacy of ciprofloxacin [43
]. The mechanism of SWCNT interaction with bacteria is so far unknown and can be further explored with molecular imaging. SWCNTs can be used for that purpose as efficient biomarkers since semiconducting species exhibit near-infrared fluorescence penetrating through the layers of biological tissue due to low tissue absorption/scattering in near-IR [44
]. Such image-guided delivery of antibiotics allowing to track their transport and elucidate SWCNT-mediated mechanism of action has not been explored to date as the covalent attachment of antibiotics quenches SWCNT emission.
In this work, we utilize non-covalent SWCNT antibiotic delivery to enhance drug efficacy and track the transport with intrinsic SWCNT fluorescence, while also aiming to circumvent the antibiotic resistance of the S. epidermis
showing low Methicillin sensitivity. Antibiotic resistance is based partly on enzymatic degradation of the existing antibiotics or a decreased membrane permeability to those. SWCNTs as delivery vehicles can be well-suited to address both of these issues as they are known to protect delivered gene therapeutics [33
] from enzymatic degradation and enhance internalization of other drug moieties [47
]. Non-covalent delivery also improves the possibility of antibiotic release within bacterial cells. Finally, antibacterial properties of SWCNTs known to disrupt the membrane and/or metabolic processes and morphology of bacteria [48
] may serve to the enhancement of antibacterial treatment efficacy. This all suggests that SWCNTs may be highly advantageous delivery vehicles for antibiotic treatment.
2. Materials and Methods
2.1. Dispersion of SWCNT in Antibiotic Solutions
Concentrated and supersaturated aqueous Doxycycline (20 mg/mL) (purchased from Alfa Aesar) and Methicillin (25 mg/mL) (purchased from Sigma Aldrich) antibiotic suspensions were prepared for SWCNT complexation and further dilution to concentrations used in the antibacterial efficacy studies. Each antibiotic in aqueous suspension was complexed with 500 µg of raw HiPco (Nanointegris batch # HR27-075A) non-covalently via 30 min of ultrasonic bath treatment followed by 20 min ultrasonic tip treatment at 16.5 W of power. Resulting suspensions containing antibiotic-suspended SWCNTs were characterized via absorption spectroscopy and stored at 4 °C with further exposure to 2 min ultrasonic treatment prior to use.
For control experiments, a solution of SWCNTs/DSPE-PEG 5000 was prepared—0.5 mg of SWCNT was added to a 1600 µM solution of DSPE-PEG 5000 (NanoCS) and subjected to the aforementioned ultrasonic dispersion and filtration procedures to yield final SWCNTs/DSPE-PEG-5000 (1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-5000]) suspensions.
2.2. Characterization of SWCNT-Antibiotic Dispersions
The concentration of all SWCNT suspensions was characterized via absorption spectroscopy. Using standard calibration curve constructed from absorptions of unfiltered SWCNT/antibiotic fractions with known SWCNT amounts we have experimentally derived extinction coefficients at 632 nm for SWCNT dispersed with both drugs (0.015 (µg/mL)−1 for SWCNTs/doxycycline and 0.0134 (µg/mL)−1 for SWCNTs/methicillin). We further used those to assess the concentration of SWCNTs in centrifuged suspensions.
The concentration of antibiotics in the suspensions of antibiotic/SWCNT hybrids was assessed via deconvoluting absorption spectra of those into components for SWCNTs and antibiotics. SWCNTs/DSPE-PEG 5000 spectra were used as an assessment for SWCNT component and antibiotic standards at known concentrations were used as reference component for antibiotics. This calculation showed w/w ratios of 1:4 for SWCNT/methicillin and 1:5 for SWCNT/doxycycline in stock SWCNT suspensions that were further used throughout this work.
Near-infrared fluorescence of antibiotic/SWCNT suspensions was collected via Nanofluorescence NS2 Nanospecralyzer spectrometer with 637 nm laser excitation for SWCNT antibiotic suspensions after preparation and after a 24 h treatment period. Minimal agitation is applied to ensure no loose aggregation.
2.3. Disk Diffusion Assay
(VWR 470176-542) broth of McFarland 0.5 standard (absorption of 0.08 to 0.1) was created with stationary phase culture using Mueller Hilton Broth. This standard stabilized the cell count at an approximate 1 × 108
]. Dilution was plated within 15 min of standardization. Following the proper aseptic techniques, 0.2 mL of bacterial broth was placed in the center of prepared agar dish. A sterile bacteria spreader was used to evenly spread the bacteria throughout the plate to create a lawn.
We tested two different dosages of the antibacterial solutions to increase the breadth and reliability of data. Blank sensitivity discs were loaded with 10 µL and 20 µL (based on respective dosage) of stock suspensions and placed onto the surface of the agar using sterile forceps. Discs were impregnated with the test solution dropwise. Five discs were evenly placed equidistant from one another. Before tilting over the Petri dishes, discs were left to dry and gently pressed down to ensure attachment to agar. Once all Petri dishes are prepared, they were turned upside down to prevent surface condensation. Petri dishes are incubated for 24 h at 37 °C, then the zones of inhibition were measured with the inclusion of disk diameter in the measurements. The 24 h time point is chosen based on previous work [50
] suggesting a possibility of bacterial growth in antibiotic-resistant strains beyond 20 h and employing this time interval for the assessment of bacterial growth.
2.4. Colony Count Assay
Using S. epidermidis broth (McFarland 0.5 standard), 100 µL was placed in the center of the agar plate. 100 µL of the respective antibacterial stock solution is added to the center. Bacteria were spread through the Petri dish and the plates were further incubated for 24 h at 37 °C. Pictures of the plates were uploaded onto the OpenCFU software to count the number of colonies grown on the plate. Two plates were prepared for each antibacterial treatment with corresponding controls.
2.5. MIC (Minimal Inhibitory Concentration) Turbidity Assay
A serial dilution (using the factor of 2) of antibacterial solutions was conducted in 12-well plates starting with 200 µL of antibacterial solution placed in first well. 850 µL of broth and 50 µL of bacteria in broth were added to each well plate. The solubility of doxycycline and methicillin was 50 and 0.31 mg/mL, respectively according to the manufacturer (Methicillin: Sigma Aldrich; Doxycycline: Alfa Aesar) information. Thus, the antibiotic stock complexed with SWCNTs for all aqueous experiments was diluted to lower concentrations: for doxycycline: 1, 0.5, 0.25, 0.125 and 0 mg/mL and for methicillin are 1.25, 0.625, 0.313, 0.106 and 0 mg/mL. Plates were then incubated at 37 °C for 24 h. The solutions were transferred to cuvettes and their MIC turbidity was measured using a Cary 500 spectrophotometer with the broth used as a baseline. Two wells were prepared for each concentration.
2.6. Cytotoxicity Assay
An MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)-based cytotoxicity assay was conducted for 4 samples—doxycycline, SWCNTs/doxycycline, methicillin and SWCNTs/methicillin. Each sample was prepared via serial dilutions at the testing concentrations ranging from 0 to 3.5 µg/mL for doxycycline and SWCNTs dispersed with doxycycline and 0 to 0.25 µg/mL for methicillin and SWCNTs dispersed with methicillin. The absorbance was measured using the FLUOstar Omega microplate reader and was analyzed using Omega software to yield cell viability levels.
We utilized InGaAs near-IR (NIR) camera coupled to hyperspectral fluorescence filter (Photon etc.) to perform fluorescence microscopy of SWCNTs imaged in bacterial cells 24 h after introducing them to bacterial culture. The sample was excited with 637 nm diode laser excitation at 130 mW output power. SWCNTS showed up in the NIR broadband (900–1450 nm) images as bright fluorescent objects. Non-treatment control images were taken for each antibiotic target ensuring no emission in the near-IR. Scanning Electron Microscope (SEM; JEOL-JSM-7100F) was used at 5 kV to image bacterial cells. Samples were prepared by depositing bacteria from the culture onto conductive carbon tape via drop-casting of ~100 µL of SWCNT/antibiotic-treated bacteria in the media. SEM allowed imaging of the outer surface of bacterial cells and extracellular SWCNTs. Transmission Electron Microscopy (TEM; JEOL JEM-2100 TEM) was further utilized to assess the incorporation of SWCNTs into bacteria and the coating of SWCNTs with antibiotic only. Samples for TEM were prepared by drying ~10 µL of either SWCNT/antibiotic suspensions or SWCNT/antibiotic-treated bacterial culture on the carbon-coated 200-mesh copper grid under ambient conditions.
This work, for the first time, explores the joint delivery and imaging of antibiotics by single-walled carbon nanotubes. SWCNTs dispersed in water with doxycycline and methicillin non-covalently attached to their surface act as drug delivery vehicles facilitating the improved antibacterial effect in Staphylococcus epidermidis. In three different sensitivity assays performed in this work, the advantages of a SWCNT/antibiotic therapy are apparent. SWCNTs facilitate preferential bacterial accumulation and internalization enhancing antibacterial effect for methicillin with marginal improvement for doxycycline. SWCNT delivery yields a 40-fold (4000%) improvement in bacterial colony inhibition for the SWCNT complex with methicillin, to which S. epidermidis initially shows resistant behavior in our assays. These results confirmed by statistically significant findings from disc diffusion and a general trend provided by the MIC turbidity assays suggest that whereas for doxycycline positive variations in efficacy can be explained by potentially increased uptake facilitated by SWCNT delivery; SWCNT/methicillin complexes likely circumvent the antibiotic resistance of S. epidermidis. SWCNTs are expected to penetrate bacterial cell walls delivering methicillin and protecting it against degradation. Based on the reported interaction of SWCNTs with the cell wall and cellular membrane we consider that direct transport of the antibiotic by the SWCNTs may introduce cell wall disruption and/or facilitate enhanced delivery and increasing susceptibility of bacteria to methicillin.
These internalization-based hypotheses are supported by SWCNT fluorescence imaging within bacterial cell culture subject to SWCNT/antibiotic treatment indicating substantial SWCNT fluorescence signal originating from bacteria rather than extracellular environment. SEM images confirm the association of SWCNTs with the cell wall of bacteria, whereas TEM verifies successful cell wall internalization by SWCNTs. In this work, SWCNT acts as effective multifunctional antibiotic delivery/imaging agents with the potential to circumvent antibiotic resistance. As methicillin is one of the more widely known antibiotics for developing resistance, its activation through noncovalent hybridization with SWCNTs offers an alternative potential approach to the antibiotic resistance issue. It may further provide a chance to reduce the dose, reuse and recycle the existing antibiotics for the treatment of the new resistant bacterial epidemics.