Carbon-Infiltrated Carbon Nanotube Topography Reduces the Growth of Staphylococcus aureus Biofilms

Abstract

Orthopedic implant-associated infections are a growing problem. These infections are often associated with bacterial biofilms, such as those formed by Staphylococcus aureus. Nanotextured surfaces can reduce or prevent the development of bacterial biofilms and could help reduce infection rates and severity. Previous work has shown that a carbon-infiltrated carbon nanotube (CICNT) surface reduces the growth of S. aureus biofilms. This work expands on previous experiments, showing that the topography of the CICNT, rather than its surface chemistry, is responsible for the reduction in biofilm growth. Additionally, the CICNT surface does not reduce biofilm growth by killing the bacteria or by preventing their attachment. Rather it likely slows cell growth, resulting in fewer cells and reduced biofilm formation.

Keywords: MRSA; Staphylococcus aureus; biofilm; carbon nanotubes; nanostructured surface.

Figure 1

SEM images of top-down SEM images of (A) bare Ti, (B) carbon control, and (C) CICNT at 10,000×. Side view (80° tilt) SEM images of (D) bare Ti, (E) carbon control, and (F) CICNT at 12,000×.

Figure 2

CFU/sample +/− standard error for adherent bacteria from four S. aureus isolates grown on either bare Ti, unstructured carbon, or CICNT surfaces. One overnight shaking culture was used to inoculate 2–3 separate and independent samples of each surface type. Each of these samples was then analyzed separately. This entire format was repeated 3–4 times to ensure reproducibility, resulting in at least 6 replicates for each isolate. Biofilms were grown for 36 h in RPMI. The difference in bacteria between Ti and carbon surfaces was not significant for any isolate. * p < 0.01.

Figure 3

(A) SEM images of Al₂O₃-coated Ti and Al₂O₃-coated CICNT surfaces before biofilm growth. (B) CFU/sample +/− standard error for adherent bacteria from S. aureus (JE2) grown on Al₂O₃-coated Ti or Al₂O₃-coated CICNT surfaces. One overnight shaking culture was used to inoculate 2–3 separate and independent samples of each surface type. Each of these samples was then analyzed separately. This entire format was repeated 3 times to ensure reproducibility, resulting in 8–9 replicates for each surface type. Biofilms were grown for 36 h in RPMI. * p < 0.01.

Figure 4

(A) Representative flow cytometry histograms for PI-positive cells from bacteria grown on Ti, unstructured carbon, or CICNT. The positive control consists of cells that were grown in a biofilm and then killed using 70% isopropyl alcohol. (B) Percent of dead cells after 36 h on various surfaces as measured by flow cytometry using PI to show membrane-permeable cells. Differences between groups are not significant (p = 0.31). Measurements are from combined adherent and non-adherent cells after 36 h of growth in RPMI (JE2 strain). One overnight shaking culture was used to inoculate 3 separate and independent samples of each surface type. Each of these samples was then analyzed separately. This entire format was repeated 2 times to ensure reproducibility, resulting in 6 replicates for each surface type.

Figure 5

CFU/sample +/− standard error for adherent S. aureus (JE2) grown on bare Ti or CICNT surfaces. Biofilms were grown for the given amount of time in RPMI. One overnight shaking culture was used to inoculate two individual samples of either CICNT or Ti. Each sample was analyzed separately. This process was repeated on new samples three times to ensure reproducibility, for a total of 6 samples of both CICNT and Ti at each time point. **** p < 1 × 10⁻⁹, ** p < 0.001.