Detection and quantification of bacterial biofilms combining high-frequency acoustic microscopy and targeted lipid microparticles

Abstract

Background: Immuno-compromised patients such as those undergoing cancer chemotherapy are susceptible to bacterial infections leading to biofilm matrix formation. This surrounding biofilm matrix acts as a diffusion barrier that binds up antibiotics and antibodies, promoting resistance to treatment. Developing non-invasive imaging methods that detect biofilm matrix in the clinic are needed. The use of ultrasound in conjunction with targeted ultrasound contrast agents (UCAs) may provide detection of early stage biofilm matrix formation and facilitate optimal treatment.

Results: Ligand-targeted UCAs were investigated as a novel method for pre-clinical non-invasive molecular imaging of early and late stage biofilms. These agents were used to target, image and detect Staphylococcus aureus biofilm matrix in vitro. Binding efficacy was assessed on biofilm matrices with respect to their increasing biomass ranging from 3.126 × 103 ± 427 UCAs per mm(2) of biofilm surface area within 12 h to 21.985 × 103 ± 855 per mm(2) of biofilm matrix surface area at 96 h. High-frequency acoustic microscopy was used to ultrasonically detect targeted UCAs bound to a biofilm matrix and to assess biofilm matrix mechanoelastic physical properties. Acoustic impedance data demonstrated that biofilm matrices exhibit impedance values (1.9 MRayl) close to human tissue (1.35 - 1.85 MRayl for soft tissues). Moreover, the acoustic signature of mature biofilm matrices were evaluated in terms of integrated backscatter (0.0278 - 0.0848 mm(-1) × sr(-1)) and acoustic attenuation (3.9 Np/mm for bound UCAs; 6.58 Np/mm for biofilm alone).

Conclusions: Early diagnosis of biofilm matrix formation is a challenge in treating cancer patients with infection-associated biofilms. We report for the first time a combined optical and acoustic evaluation of infectious biofilm matrices. We demonstrate that acoustic impedance of biofilms is similar to the impedance of human tissues, making in vivo imaging and detection of biofilm matrices difficult. The combination of ultrasound and targeted UCAs can be used to enhance biofilm imaging and early detection. Our findings suggest that the combination of targeted UCAs and ultrasound is a novel molecular imaging technique for the detection of biofilms. We show that high-frequency acoustic microscopy provides sufficient spatial resolution for quantification of biofilm mechanoelastic properties.

Figure 1

Biofilm matrix formation. Individual bacterial cells gain entrance into the bloodstream and attach at favorable sites. As they continue growing, they form a protective biofilm matrix against hostile agents, the immune system or fluid turbulences caused by hemodynamic forces. As the biofilm matrix matures, individual cells are dispersed into the bloodstream where they travel to distant sites in the body forming colonies. Figure adapted from [6].

Figure 2

Targeted ultrasound contrast agents bind to biofilm matrix in a time-dependent fashion. Targeted UCAs bind to the biofilm mass. As the biofilm matrix grows, an increased surface area is accompanied by an increase in the number of bound UCAs. (A-D) Epifluorescence microscopy imaging of the biofilm matrix for 24 h, 48 h, 72 h and 96 respectively (scale bar = 50 μm; scale bar of insets = 15 μm). Bacterial cells are stained with DAPI (blue; arrows), targeted UCAs are microbubbles conjugated with streptavidin (red; open arrowheads) and biofilm matrix is detected by staining for FITC-conjugated lectins (green; filled arrowheads). (E) Biofilm mass surface area over time (24 h, 48 h, 72 h and 96 h). (F) Number of targeted UCAs bound to the biofilm matrix over the same time course (24 h, 48 h, 72 h and 96 h).

Figure 3

Time-resolved high-frequency scanning acoustic microscopy for imaging and quantification of biofilm matrix. (A) A piezoelectric transducer transmits highly focused ultrasound beams at the sample under investigation. The ultrasound waves travel through the sample and the surrounding medium; reflected echoes are returned and recorded by the transducer. The echoes are resolved on the time axis. (B) The scanning acoustic microscopy unit is mounted onto an inverted light microscope allowing for a direct correlation of the same spatial regions in terms of optical (or fluorescence) and acoustic information assessment.

Figure 4

Ultrasound and optical/fluorescent images of biofilm matrix. A 1 × 1 mm area of the biofilm sample was scanned at a center frequency of 100 MHz with a time-resolved high-frequency scanning acoustic microscope. Inset shows a smaller region of the same sample imaged by fluorescence microscopy. Due to the fact that the acoustic lens is mounted onto a piezoelectric scanner, it provides greater flexibility in imaging larger regions than a microscope objective alone, which is limited by its stationary positioning.

Figure 5

Targeted UCAs bind specifically to the biofilm matrix and are detected by ultrasound. (A) Ultrasound insonification of a region where no biofilm matrix is present. The acoustic signature originating from the unbound UCAs will be different than the one from bound UCAs. (B) Biofilm matrix is detected by ultrasound using targeted UCAs that specifically bind to the biofilm matrix.

Figure 6

Ultrasound with targeted UCAs is a viable method to quantify biofilm matrix. The biofilm matrix was quantified by time-resolved high-frequency scanning acoustic microscopy. (A) Integrated backscatter shown for a range of different frequencies from 97 MHz to 104 MHz. Regions where targeted UCAs are bound to the biofilm exhibit stronger backscatter and thus allow for the detection of biofilm matrix. (B) Attenuation of sound over the same frequency range as in 6A. Regions with bound targeted UCAs show a significantly higher sound attenuation compared to regions with unbound agents. The different sound signature from bound vs. unbound regions may potentiate an earlier and more effective diagnosis of infected regions.