Thermal and Acoustic Stabilization Of Volatile Phase-Change Contrast Agents Via Layer-By-Layer Assembly

Abstract

Objective: Phase-change contrast agents (PCCAs) are perfluorocarbon nanodroplets (NDs) that have been widely studied for ultrasound imaging in vitro, pre-clinical studies, and most recently incorporated a variant of PCCAs, namely a microbubble-conjugated microdroplet emulsion, into the first clinical studies. Their properties also make them attractive candidates for a variety of diagnostic and therapeutic applications including drug-delivery, diagnosis and treatment of cancerous and inflammatory diseases, as well as tumor-growth tracking. However, control over the thermal and acoustic stability of PCCAs both in vivo and in vitro has remained a challenge for expanding the potential utility of these agents in novel clinical applications. As such, our objective was to determine the stabilizing effects of layer-by-layer assemblies and its effect on both thermal and acoustic stability.

Methods: We utilized layer-by-layer (LBL) assemblies to coat the outer PCCA membrane and characterized layering by measuring zeta potential and particle size. Stability studies were conducted by; 1) incubating the LBL-PCCAs at atmospheric pressure at 37^∘C and 45^∘C followed by; 2) ultrasound-mediated activation at 7.24 MHz and peak-negative pressures ranging from 0.71 - 5.48 MPa to ascertain nanodroplet activation and resultant microbubble persistence. The thermal and acoustic properties of decafluorobutane gas-condensed nanodroplets (DFB-NDs) layered with 6 and 10 layers of charge-alternating biopolymers, (LBL₆NDs and LBL₁₀NDs) respectively, were studied and compared to non-layered DFB-NDs. Half-life determinations were conducted at both 37^∘C and 45^∘C with acoustic droplet vaporization (ADV) measurements occurring at 23^∘C.

Discussion: Successful application of up to 10 layers of alternating positive and negatively charged biopolymers onto the surface membrane of DFB-NDs was demonstrated. Two major claims were substantiated in this study; namely, (1) biopolymeric layering of DFB-NDs imparts a thermal stability up to an extent; and, (2) both LBL₆NDs and LBL₁₀NDs did not appear to alter particle acoustic vaporization thresholds, suggesting that the thermal stability of the particle may not necessarily be coupled with particle acoustic vaporization thresholds.

Conclusion: Results demonstrate that the layered PCCAs had higher thermal stability, where the half-lifes of the LBL_xNDs are significantly increased after incubation at 37^∘C and 45^∘C. Furthermore, the acoustic vaporization profiles the DFB-NDs, LBL₆NDs, and LBL₁₀NDs show that there is no statistically significant difference between the acoustic vaporization energy required to initiate acoustic droplet vaporization.

Keywords: acoustic activation; biopolymers; decafluorobutane; layer-by-layer; lipids; microbubbles; nanodroplets; perfluorocarbon; phase-change contrast agents; thermal activation.

Figure 1:

Schematic of an LBL_xND. Cationic decafluorobutane nanodroplets (DFB-NDs) were generated via microbubble condensation followed by deposition of alternating layers of charged biopolymers onto the nanodroplet surface. (TAP = cationic lipid).

Figure 2:

Schematic of the US imaging system. A phantom chamber is coupled to the Vantage64 LE imaging system via the L11-4 transducer. The transducer is driven/controlled via commercially available software (Verasonics) executed in MATLAB. Raw pulse echo data is processed via an image processing routine to count particles and remove back-scatter/reflections.

Figure 3:

Image Processing Procedure. (A) Frames from US transducer - note the back-scatter and side-wall reflections. (B) Scatter plot contain the frame-frame cross correlation across 1600 frames. Note the markers highlighted in red demonstrate frames with a low degree of correlation - denoting potential ADV events. (C) Side by side comparison of an unprocessed and processed frame containing ADV events mediated via high intensity US probe. Image processing routine removes back-scatter and counts the number of ADV and cavitation events.

Figure 4:

Droplet size and zeta potential. (A) Box plot with droplet size as a function of number of layers. (B) Box plot with zeta potential as a function of number of layers. (C) Gaussian normalized particle size distributions of DFB-NDs, LBL₆NDs, and LBL₁₀NDs. (D) Gaussian normalized normalized particle zeta distributions of DFB-NDs, LBL₆NDs, and LBL₁₀NDs. (E) Gaussian normalized particle zeta distributions of DFB-NDs and LBL_nNDs. All data points and distributions consisted of N = 11 samples.

Figure 5:

ADV Response Curves. Denotes number of particles vaporized as a function of MI. The response (points) where fitted to the sigmoidal activation function (dashed curve). Error bars denote the standard error of each response with N = 8 samples per point.

Figure 6:

ADV response for (A) DFB-NDs, (B) LBL₆NDs, and (C) LBL₁₀NDs as a function of mechanical index (MI) and time when incubated at 37 °C. Similarly, (D-F) present the ADV response for (D) DFB-NDs, (E) LBL₆NDs, and (F) LBL₁₀NDs as a function of MI and time when incubated at 45 °C. The response (points) where fitted to the sigmoidal activation function (dashed curve). Error bars denote the standard error of each response with N = 8 samples per point.

Figure 7:

ADV Processed Data for (A-C) DFB-NDs, (D-F) LBL₆NDs, and (G-I) LBL₁₀NDs incubated at 45°C and exposed to a high intensity pulse, MI of 1.36 (P_n = 3.70 MPa), at times 0, 24, 48 hrs., columns 1, 2, and 3, respectively. Note particles highlighted with (+) indicate acoustically induced vaporization (ADV), and (−) indicate acoustically induced cavitation.

Figure 8:

ADV response at MI = 1.36 (P_n = 3.70 MPa) as a function of incubation time for (A) 37°C and (B) 45°C. Data points were fitted to exponential decay curves. Error bars denote the standard error of each response with N = 8 samples per point.