- Email: [email protected]
Lipid Microbubbles as Ultrasound-Stimulated Oxygen Carriers for Controllable Oxygen Release for Tumor Reoxygenation
ARTICLE IN PRESS Ultrasound in Med. & Biol., Vol. ■■, No. ■■, pp. ■■–■■, 2017 Copyright © 2017 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Printed in the USA. All rights reserved 0301-5629/$ - see front matter
https://doi.org/10.1016/j.ultrasmedbio.2017.08.1883
●
Original Contribution
LIPID MICROBUBBLES AS ULTRASOUND-STIMULATED OXYGEN CARRIERS FOR CONTROLLABLE OXYGEN RELEASE FOR TUMOR REOXYGENATION Chunjiang Yang,* Huan Xiao,* Yang Sun,† Lirong Zhu,* Yang Gao,* Sunny Kwok,‡ Zhigang Wang,† and Yi Tang* *
Department of Ultrasound, Children’s Hospital of Chongqing Medical University, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatrics, China International Science and Technology Cooperation base of Child Development and Critical Disorders, Chongqing, China; † Second Affiliated Hospital, Institute of Ultrasound Imaging, Chongqing Medical University, Chongqing, China; and ‡ Department of Biomedical Engineering, Ohio State University, Columbus, OH, USA (Received 25 March 2017; revised 22 August 2017; in final form 24 August 2017)
Abstract—Microbubbles are proposed as a potentially novel method for oxygen delivery in vivo in initial studies. The lack of commercial microbubbles for oxygen delivery in preclinical research prompted us to fabricate an oxygenloaded lipid microbubble. We aimed to extend the innovative strategy to modulate the tumor hypoxic microenvironment, using microbubbles intravenously as an oxygen carrier for the controllable tumor-specific delivery of oxygen by ultrasound (US). In our experiment, an oxygen-loaded lipid-coated microbubble (OLM) with mixed gas (O2/C3 F8, 5:1 v/v) was fabricated and exhibited a higher rate of oxygen release to a desaturated solution through burst by US than that in the absence of US. Although in in vivo studies, OLMs could be imaged and triggered by US to elevate the pO2 level in the breast VX2 tumor dramatically within a matter of minutes. The added presence of US-activated OLMs elicited a nearly six-fold increase in pO2 levels within 1 min compared with that of the pre-injection. Owing to the high oxygen payload, great acoustic stability and acoustic properties, OLMs may be proposed as an ideal radio-sensitizer. We conclude that oxygen release mediated by ultrasound-targeted microbubble destruction is feasible and shows potential in image-guided, site-specific cancer radiotherapy. (E-mail: [email protected]) © 2017 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound, Oxygen delivery, Lipid microbubble, Controllable release, Hypoxic tumor.
from achieving its full therapeutic potential (Ruan et al. 2009). The oxygenation level of the solid tumor plays a vital role in the radiation process. Some research has shown that the biological effects of ionizing radiation are largely mediated by reactive oxygen intermediates. Oxidative stress can lead to DNA damage and chromosomal instability, contributing to the cellular sensitivity to irradiation (Tulard et al. 2003). The oxygen enhancement ratio (OER) is often described as the degree of radiation sensitization and refers to the ratio of doses in the absence of oxygen to the doses in the presence of oxygen to provide the same cell survival. The OER for mammalian cells is 2.5–3.0; the seemingly low OER value of 2.6 corresponds with 50% of cells killed under hypoxic conditions as opposed to almost 99% of the cells killed under aerobic conditions with a radiation dose of 1000 cGy (Kwan et al. 2012). This information implies that if tumor cells with oxygen levels in the intermediate range can be reoxygenated, an increased sensitivity to radiation would result.
INTRODUCTION Hypoxia is a common phenomenon in most solid tumors because the oxygen supply is reduced by disturbed microcirculation and deteriorated diffusion. It is estimated that at least 50%–60% of advanced solid tumors contain hypoxic or anoxic tissue, typically because of irregularities in the tumor microcirculation (Ackerman and Simon 2014). When the transport of oxygen is unable to meet the demand, hypoxia occurs. Tumors are metabolically designed to thrive under hypoxic conditions, and the effectiveness of therapy is impaired by an inadequate oxygen supply in tumors because the hypoxia is associated with an increased risk of metastasis and therapeutic resistance (i.e., radiotherapy, chemotherapy and photodynamic therapy), which prevents cancer therapy
Address correspondence to: Yi Tang, 136#Zhongshanerlu, Chongqing 400014, China. E-mail: [email protected] 1
ARTICLE IN PRESS 2
Ultrasound in Medicine & Biology
It is crucial to improve the tumor oxygenation level during radiation therapy. Recently, various strategies have been proposed to oxygenate the tumor microenvironment to achieve better therapeutic outcomes. Traditional oxygen-delivering technologies fall into three categories: hyperbaric oxygen therapy (HBO2), oxygen-carrying materials and oxygen-generating materials. While various methods have been investigated to deliver O2 in vivo, no specific technology has provided an effective way for oxygen delivery into a targeted tissue. A recent study found that ultrasound (US) could favor nano-perfluorocarbon for tumor-specific delivery of oxygen and could dramatically enhance tumor oxygenation, thus improving therapeutic outcomes in radiation therapy treatment of tumors (Song et al. 2016). However, like most traditional technologies, it depends on the respiratory system and adsorbed oxygen in the lung to release oxygen into the circulation. Thus, such studies were used to construct more targeted specifications for oxygen-delivering biomaterials to avoid oxygen exchange through the respiratory system and determine suitable technologies for clinical translation. An emerging new technology that has extensive use clinically involves the microbubble as a US contrast agent and an oxygen carrier simultaneously. Because of the microbubble’s acoustic activity, US can be used to stimulate and image gas release in a targeted region. Another unique characteristic of microbubbles is that they can be burst locally using US in vivo, releasing an oxygen payload to a particular area of interest. These particles consist of gaseous cores surrounded by thin shells that are composed of lipids, polymers, surfactants, dextran and chitosan (Bisazza et al. 2008; Eisenbrey et al. 2015; Swanson et al. 2010). Among these coatings, results from many studies expressed an interest in lipid membranes; phospholipid monolayers present a relatively minor impediment to gas diffusion. Within the past decade, a few studies have focused on lipid-oxygen-loaded microbubbles. However, most studies are still performed in vitro and focus on fabrication. Few studies investigate the precise oxygen payload delivery to the hypoxic tumor. Although some results demonstrate that sonodynamic therapy may be significantly enhanced by incorporating an approach that involves US and microbubble-mediated delivery of oxygen to hypoxic tumor, the authors only compared the indirect parameter (i.e., the singlet oxygen generation) in cell-free systems before or after US exposure (McEwan et al. 2015). Furthermore, the ideal components of said gas cores have not been conclusively agreed upon. Pursuing this idea further we have developed oxygen-loaded lipid-coated microbubbles (OLM) and investigated its stability in terms of structural characterizations and ability to retain oxygen payload over time. To enlarge the range of OLM’s clinical application, OLMs are intravenously injected into tumor-bearing rabbits. With a low-power clinically adapted
Volume ■■, Number ■■, 2017
US transducer applied on their tumors, the elevation of oxygenation level in tumors was monitored. METHODS Implantation of VX2 breast tumors We anesthetized New Zealand white rabbits that weighed 2.0–2.5 kg (Laboratory Animal Center of Chongqing Medical University, Chongqing, China) with an intra-muscular injection of 3% pentobarbital solution (1 mL/kg). The abdomens of the rabbits were depilated with 8% Na2 S solution, then fixed in prone position and routinely disinfected. The VX2 tumor is a squamous cell carcinoma similar to human cancer in morphology and biologic characteristics, such as rapid growth, which makes the VX2 tumor a good model for experimental research. The tumor model could be built directly from obtained tumor blocks (from tumor-bearing rabbits) for transplantation. The experimental cost in using this method is lower than the cost of the cell culture method, and the technical operation is relatively simple and easy to promote. The VX2 tumor tissue was excised from the tumor-bearing rabbit and soaked in 20 mL Hanks’ balanced salt solution. Then the tumor was sheared to small masses with a size of approximately 0.5–1.0 mm in diameter. The final suspension was extracted into a 20-mL syringe, and a 1-mL tumor tissue suspension was injected into the mammary gland of the rabbits underneath the second left nipple. Each rabbit was given 500,000 units of streptomycin after the tumor implantation. Two weeks after implantation, the tumors reached 80–100 mm3 in volume. These tumorbearing rabbits were used for investigations of contrast imaging and oxygen release in vivo. All procedures involving the use and care of rabbits were approved ethically and scientifically by Chongqing Medical University in compliance with the Practice Guidelines for Laboratory Animals of China. Microbubble construction and preparation The compounds 1,2-distearoyl–sn-glycero-3phosphocholine (DSPC) and 1,2-dipalmitoyl-rac-glycero3-phospho (dimethylaminoethanol)–methoxy polyethylene glycol 5000 (DPPE-MPEG5000) (Avanti Polar Lipids, Inc., Alabaster, AL, USA) were synthesized in a 9:1 molar ratio and mixed with phosphate-buffered saline (PBS) and glycerin in a final volume of 0.5 mL in 1.5 mL vials. Parafilm was used to seal the vial. The mixtures were fully dissolved at 55°C under a thermostat water bath for 30 min. Then the headspace of each vial was filled with perfluoropropane (C3 F8) gas, highpurity oxygen (99.999%) and pure nitrogen (N2) with a gas volume ratio of O2:C3 F8 of 5:1 v/v to replace the air in the vials under a vacuum with a syringe. The dispersion in the vials was mechanically shaken in a dental
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
amalgamator (YJT, Medical Apparatus and Instrument Ltd., Shanghai, China), blending for 50 s at 4000 rpm, as described by Ren et al. (2009). After 1.5 mL of PBS was added to each vial, the suspension was diluted and stored at 4°C for further use. Morphology and size distribution The morphology of OLM was observed under an optical microscope (CX41, Olympus Corporation, Shinjuku, Tokyo, Japan). The concentration of OLM was calculated using a Hemocytometer (BLAUBRAND counting chamber, Wertheim, Germany) instrument during each observation, and the concentration of OLMs is expressed as follows (Hansen 2000):
Microbubbles concentration per milliliter = Microbubbles within 4 major squares × 2500 (1) × Dilution factor The diameter of OLMs was evaluated using the Mastersizer 3000, a laser particle-size analyzer (Malvern Instruments Ltd., Malvern, Worcestershire, UK), at 1, 3 and 7 d after fabrication, and the concentration was calculated 0.5 h, 1 d, 3 d and 7 d after fabrication (Li et al. 2015). To obtain accurate measurements, each syringe was inverted 5 times to be sure of heterogeneous emulsion before measurements, which were collected 3 times per test. The particle charge was qualified as zeta potential, using the Zetasizer 3000, a laser Doppler anemometry (Malvern Instruments Ltd.) 1 d after fabrication (Burke et al. 2014; Grenha et al. 2008). The pH values were measured using the Mettler Toledo Delta320 (Mettler-Toledo, Greifensee, Switzerland) at 0.5 h, 1 d, 3 d and 7 d after fabrication. In vitro oxygen release with or without US exposure Natural release of oxygen in hypoxic solutions. Normal saline solution (NaCl, NS) with an oxygen concentration of 0.9% was sealed in a vial, then the O2 concentration was reduced to 0.4 mg/L (severe hypoxia) (Cavalli et al. 2009; Rimoldi et al. 2012) by means of an N2 purge (Cavalli et al. 2009; Oh et al. 2009). A total of 3 mL of each type of OLMs were introduced into 20 mL of NS. All OLMs were stored at 4°C. Experiments were carried out after 0.5 h, 3 h and 1 d after OLM preparation (3 groups). The oxygen-release kinetic was monitored for 10 min using an oximeter (HQ30 d, HACH, Loveland, CO, USA) for each group, and the dissolved oxygen data of each solution was recorded every 20 s. The experiments were repeated in triplicate for each independent sample.
3
Oxygen release of OLMs triggered by US in hypoxic solutions The O2 concentration in a sealed vial containing NS solution was reduced to 4.0 mg/L by means of a N2 purge to mimic moderate hypoxic conditions. The solutions were divided into the following two groups: (a) OLM only and (b) OLM with US exposure. Also, using the same procedure described earlier, 3 mL of the OLM was introduced into 20 mL of NS 3 h after fabrication. The oxygenrelease kinetic was monitored for 10 min using an oximeter, and dissolved oxygen data of the solution were recorded every 20 s. For treatment of group b, continuous US pulses with a spatial average intensity of 0.5 W/cm2 were applied to the bottom of the vials for 10 min immediately after the OLMs were added to the solution. The US transducer was unfocused and consisted of a single element (UGT1025, Institute of US Imaging of Chongqing Medical University, Chongqing, China) with a diameter of 2 cm and was applied at 1 MHz, 3.5 W/cm2 and 50% duty cycle for 30 s (Zhou et al. 2015). The US was combined with an airtight apparatus consisting of organic glass and sealing rings. The oximeter was calibrated in the air before the experiment, and all experiments were performed at a stable room temperature (25°C) and humidity conditions. The experiments were repeated in triplicate for 3 independent samples and then averaged.
Quantitative evaluation of dissolved oxygen in blood samples Fresh healthy human venous blood was collected in heparin-containing vacuum tubes, then gathered in 6 pipettes, each having a volume of 3 mL, and stored at 4°C in a refrigerator (collection occurred immediately before the blood-gas analysis). The treatment groups were divided into 4 groups: (i) OLM, (ii) C3 F8 Ms microbubble, (iii) NS and (iv) the control group (just the blood sample). We then gently mixed 1 mL of the corresponding OLM or NS with 1 mL of the blood sample, collected by use of a 2.5 mL disposable syringe and sealed immediately, for approximately 3 min. The dissolved oxygen partial pressure (pO2) was immediately measured using a blood-gas analyzer (ABL510, Radiometer Medical, Copenhagen, Denmark). Each sample was taken using a disposable syringe at room temperature (25°C), and the needle was immediately plugged with a rubber stopper to avoid contamination from the air. The experiments were repeated in triplicate for 3 independent samples and then averaged. In addition, based on the hemoglobin saturation, the oxygen volume in the blood sample can be calculated as follows (Scholz et al. 2010):
CO2 = (a × pO2 + b × Hb ×Sat )×V
(2)
ARTICLE IN PRESS 4
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2017
where a = O2 solubility in blood (0.003 mL/mmHg); b = oxygen-carrying capacity of Hb (1.34 mL/g); Hb = concentration of hemoglobin in blood (15 g/dL); V = blood volume (3 L); and Sat = saturation of hemoglobin. In addition, the saturation of hemoglobin in blood at pH 7.4 and 37°C is expressed as follows (Dash and Bassingthwaighte 2010):
Sat = K O 2 × pO2 n
(1+ K O 2 × pO2 n )
(3)
where KO2 = Hill coefficient; n = Hill exponent (n = 2.7); KO2 = (P50)n; and P50 = 26.6 mmHg (Goutelle et al. 2008). Human venous blood was collected from healthy donors in our research team. Each participant provide informed consent. Contrast imaging in the solid tumor To evaluate the contrast enhancement of OLM in a solid tumor in vivo, three rabbits inoculated with the VX2 tumor line (squamous carcinoma) were fixed in a supine position after sedation with an intra-muscular injection of 3% pentobarbital solution (1 mL/kg). An LA522 linear transducer (Mylab Twice, Esaote, Genova, Italy) equipped with contrast pulse sequence technology (CnTI technology, Esaote) was performed on the tumor surface. The parameters of the time-gain compensation and focus range were adjusted to the optimum. Each rabbit received a bolus injection of 0.02 mL/kg of the OLMs suspension through a 20-gauge catheter in an ear vein. The tubing was flushed with 2 mL NS after injection of OLMs. Real-time scanning started immediately after injection of the microbubbles. The tumors’ images were then continually recorded for 5 min. The dynamic and static images were stored for further analysis. Quantitative analysis was detected, using the US Image Analyzer (DFY-II type, Institute of US Imaging of Chongqing Medical University, Chongqing, China) (Ren et al. 2009). Before and after contrast, the appropriate regions of interest in the tumor were selected. The echo intensity of the regions of interest and respective evaluations were all marked on the quantitative images. Determination of pO2 in VX2 tumor The area above the breast tumor of a total of three rabbits was carefully depilated with 8% Na2 S again after anesthesia. The tumor was located and measured by US before microbubble injection. The study was conducted with C3 F8 Ms as the control group compared with OLMs before injection and 5 s after injection without US exposure, or immediately after US triggered after injection. The working electrode (length: 50 mm, diameter: 0.8 mm, cable length: 500 mm, oxygen consumption rate: 10 ng/h) was introduced into the center of the tumors before agent injection. The calibration process was done before every
Fig. 1. Scheme of intravenous injection of OLMs triggered by image-guided US and monitoring pO2 level in tumor. OLMs = oxygen-loaded lipid-coated microbubbles; US = ultrasound.
measurement by bubbling 100% nitrogen gas and 20% oxygen gas, respectively, into the beaker. The OLM injection was performed after the total destruction of the C3 F8 Ms. In brief, the rabbits received an injection of 0.5 mL microbubbles, then 1.5 mL NS. The intra-tumoral oxygen concentration was recorded every 6 s for 60 s by an IMP211 apparatus (Inter Medical Co., Ltd., Shinagawaku, Tokyo, Japan) (Tokuhiro et al. 2010) before injection, similarly recorded 5 s after microbubble injection without US exposure and immediately after US-triggered bursting. To record pO2 after US-triggered bursting, every 5 s for 30 s after agent injection, the microbubbles present in the field of view containing the tumor were destroyed by temporarily (1 s) increasing the acoustic power (MI 1.9) (Marinelli et al. 2014) with an LA523 broadband linear array transducer (frequency range, 4–13 MHz). The B-mode and contrast-enhanced harmonic images mechanical index (MI, 0.06, derated pressure, 50 kPa) were concurrently obtained for each sample and all imaging parameters were kept constant for all imaging sessions (Mylab 90, Esaote). The process was guided by US imaging (Fig. 1).
Statistical analysis All data are shown as the mean ± standard deviation (SD). Statistical analysis was performed using SPSS17.0 software (IBM, New York, NY, USA) and GraphPad Prism 5.01 (GraphPad Software, La Jolla, CA, USA). Statistical significance among multiple groups was determined using a one-way analysis of variance with StudentNewman-Keuls test multiple comparison post-tests. A paired-samples t-test was used to analyze the pre-contrast and post-contrast echo intensities of tumors, and an unpaired t-test was used to analyze the oxygen-release of OLMs with and without US exposure. In all analyses, p < 0.05 was taken as indicative of statistical significance.
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
5
Fig. 2. Physical characteristics of OLMs. (a) Light microscopy images of OLMs (×200). (b) Size distribution plot of OLMs 1 d after preparation. OLMs = oxygen-loaded lipid-coated microbubbles.
RESULTS The characterizations of OLMs Some physical characteristics of OLMs are shown in Figure 2. The morphology of OLMs remained spherical under a microscope. The average diameter was 1033 ± 72 nm, concentration was (3.79 ± 0.32) × 109/mL, zeta potential of OLM was −12.9 ± 2.4 mV and the pH measurements of the OLM was approximately 7.120 ± 0.078 1 d after their fabrication.
The diameters, concentrations and pH values of OLM formulation did not change significantly 3 h, 1 d, 3 d and 7 d after fabrication (p > 0.05) (Table 1). The oxygen-release kinetic of OLMs with or without US exposure Oxygen-release kinetics in the presence of OLMs was observed 0.5 h, 3 h and 1 d after the microbubble
Table 1. Physicochemical characteristics of OLMs over time
Diameter ± SD (nm) Concentration ± SD (×109/mL) pH ± SD
0.5 h
1d
3d
7d
— 4.04 ± 0.87 7.12 ± 0.08
1033 ± 72 3.79 ± 0.32 7.07 ± 0.03
1069 ± 53 3.39 ± 0.31 7.07 ± 0.08
1055 ± 89 3.41 ± 0.34 7.010 ± 0.03
OLMs = oxygen-loaded lipid-coated microbubbles.
ARTICLE IN PRESS 6
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2017
Fig. 3. Dissolved oxygen–time plots for 10 min immediately after OLMs were introduced into the various hypoxic solutions. (a) The severe hypoxic solution (0.4 mg/L) 0.5 h, 3 h and 1 d stored at 4°C after OLM fabrication. (b) The moderate hypoxic solution (4.0 mg/L) with or without US exposure. OLMs = oxygen-loaded lipid-coated microbubbles; US = ultrasound.
preparation. OLMs showed the capacity for oxygen release throughout their preparation over time. The 10-min oxygen increase by OLMs 0.5 h, 3 h and 1 d after their fabrication are reported in Figure 3a. A high amount of oxygen was released in severe hypoxic conditions (0.4 mg/L), producing an increase in the oxygen concentration of about 6- to 12-fold within 10 min. The oxygen concentration of hypoxic solutions increased markedly 1 min after the injection of OLMs individually and then decreased slowly to steady-state within 10 min. The oxygen release profiles of OLMs 0.5 h, 3 h and 1 d after fabrication were very similar and did not significantly change throughout time, showing the oxygen payload stability of the OLMs stored at 4°C. A greater amount of oxygen was released in severe hypoxic conditions (3.80 ± 0.90 mg/L) than in moderate hypoxic conditions (1.86 ± 0.66 mg/L) in the absence of
US 3 h after fabrication, producing an increase in the oxygen concentration of about two-fold in 10 min. In the moderately hypoxic solution, the OLMs can release oxygen and increase the dissolved oxygen of the solution by approximately 0.5-fold, about 1.86 ± 0.66 mg/L without US; however, the oxygen concentration increased by approximately 1.2-fold, about 3.88 ± 0.83 mg/L in the presence of US (Fig. 3b). Blood oxygenation by use of OLMs Dissolved oxygen concentration plots for OLMs introduced into the blood sample for 10 min are shown in Figure 4. The average pO2 in blood diluted with OLM was 19.53 ± 1.39 kPa, the value of average pO2 in control blood (without dilution) was 4.63 ± 0.42 kPa, and the values in blood diluted with pure C3 F8 microbubbles (C3 F8 Ms)
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
7
Fig. 4. Dissolved oxygen–time plots for OLM introduced into the blood sample. Comparison of pO2 increases in blood with OLMs in the blood sample without sonication (*p < 0.05). OLMs = oxygenloaded lipid-coated microbubbles.
and in NS were (7.93 ± 0.31 kPa) and (6.43 ± 0.80 kPa), respectively. OLMs caused an obvious increase in the pO2 of the blood sample in vitro (p < 0.05). However, the other groups exhibited no obvious effect on the pO2 of the blood sample (p > 0.05). In this study, the pO2 of blood in the OLMs group was improved from 4.2 kPa to 18.1~20.9 kPa (135.8~156.8 mmHg). Given this data at 37°C, with pH 7.4, the saturation value of hemoglobin was calculated to be between 98.8% ~ 99.2%, and the value of the total Co2 for blood diluted with OLMs (Sat) was 60.1 ~ 61.1 mL. For the NS group, the Sat was calculated to be 77.4% ~ 87.1%, and the value of total Co2 was 47.1 ~ 53.0 mL. For the C3 F8 Ms group, the Sat was calculated to be 88.7% ~ 90.5%, and the value of total Co2 was 53.9 ~ 55.2 mL. For the control group, the Sat was calculated to be 60.9% ~ 71.8%, and the value of total Co2 was 37.0 ~ 43.6 mL. The value of total Co2 for blood mixed with OLMs was higher than that with the NS, the C3 F8 Ms and nothing. The results show that OLMs could significantly increase the pO2 value of venous blood. Contrast imaging in vivo and quantitative analysis This study included a total of three rabbits, which received OLM and C3 F8 Ms successively by injections into the ear vein. The tumor US images pre- and post-injection of OLMs, using DFY (Institute of US Imaging of Chongqing Medical University) analysis are shown in Figure 5. The echo intensity of the tumor was enhanced and significant enhancements were observed for 5 min. A contrast enhancement of the tumor appeared quickly in 20 s. After injecting OLMs, the tumor was initially enhanced (approximately 5 s after injection). The tumor parenchyma then exhibited uniformly quick enhancement, the imaging reached a peak within 15–25 s and the echo intensity slightly decreased. The contrast-enhanced process lasted for more than 5 min. In a single imaging plane, 20 separate regions of interest were selected within the tumor
Fig. 5. Ultrasound contrast imaging and quantitative analysis image of the echo intensity in VX2 breast tumor. (a) Pre-contrast echo intensity. (b) Post-contrast echo intensity. Tumor location (indicated by lines).
of each rabbit to compute the mean dB. The average intensity in the tumor before administration was 14.03 ± 3.00 dB. The intensity in the tumor at 20 s after administration was 115.85 ± 34.83 dB. In vivo experimentation showed that OLMs could strongly enhance the echo intensity of the solid tumor within 5 min (p < 0.05). Elevation of pO2 in VX2 tumors triggered by US We further assessed the pO2-increasing ability of OLMs as the carrier for the oxygen delivery using rabbits bearing VX2 tumor. After intravenous injection of OLMs, using C3 F8 Ms as the control group, the tumor was burst by US and the pO2 value within the tumors were monitored at each point during this process. As shown in Figure 6, pO2 values have no obvious change before and after the injection of OLMs (p = 0.777) or C3 F8 Ms microbubbles (p = 0.999). However, after US irradiation, significant enhancement of pO2 in the breast tumor tissue in the OLMs group was generated at 1 min in comparison with pre-injection, post-injection and triggered C3 F8 Ms microbubbles (p < 0.05). After triggering by US, the pO2 of the tumors in OLMs group were recorded to be 364.37 ± 80.48 mmHg compared with 54.49 ± 12.05 mmHg
ARTICLE IN PRESS 8
Ultrasound in Medicine & Biology
Fig. 6. Determination of pO2 within VX2 tumor in rabbits. (a) The measurements of pO2 in tumors treated with OLMs and ultrasound activating. Tumors with triggered OLMs showed a highest pO2 levels (*p < 0.01). (b) The measurements of pO2 in tumors treated with C3 F8 Ms and ultrasound activating. No difference was found among three statuses (p > 0.05). OLMs = oxygen-loaded lipidcoated microbubbles.
before US triggering, while the C3 F8 Ms group did not significantly increase after being triggered by US (p = 0.974). The added presence of US-activated OLMs elicited a nearly six-fold increase in pO2 levels at 1 min compared with that of pre-injection, while the C3 F8 Ms microbubbles did not cause such an effect. DISCUSSION Microbubbles have attractive prospects for local oxygen delivery in reversing hypoxic organs or tumors by using ultrasound-targeted microbubble destruction (UTMD), which may provide life-sustaining doses of oxygen in the event of airway distress and help elevate the local oxygenation level within a short period of time. However, no commercial microbubbles are available for oxygen delivery in preclinical research. Among current studies, few provided the quantitative evaluation of pO2 in vivo. In exploring an oxygen-loaded microbubble suitable for this preclinical study, we fabricated OLMs with
Volume ■■, Number ■■, 2017
DSPC as the main coating component and a mixed-gas core (O2:C3 F8, 5:1 v/v). DSPC has been recognized as the optimum shell for oxygen containment and exhibits high biocompatibility (Rawat et al. 2008; Swanson and Borden 2010). The compression elasticity of DSPC is relatively stronger, and DSPC could resist the size and shape deformation of microbubbles during the gas exchange to carry oxygen more stably. Moreover, DSPC retained its gas for a far longer period and still contained more than 30% gas of volume at the end of testing. These results favor DSPC as the main lipid for US molecular imaging and oxygen carriage. Another critical factor in the design of oxygenloaded microbubbles is the gas core, which is strongly associated with the bubble size. Bubble size needs to be controlled both at the time of injection and throughout the circulation lifetime; bubble growth or aggregation in the circulation must be avoided. Most initial studies introduced pure oxygen into microbubbles. However, early observations showed that, while pure oxygen microbubbles could be generated, they resulted in low yields and were generally unstable. Our earlier findings showed that, with the lowest concentration, such pure oxygen microbubbles were unstable for oxygen delivery and contrast imaging. Oxygen as the gas phase is easily soluble in water and dissolves rapidly in blood under the mutual influence of Laplace pressure and arterial pressure. Fluorocarbon gases could hinder microbubble dissolution significantly because of their low aqueous solubility and diffusivity. Thus, they could increase the half-life of the bubbles from a few s to several min (Riess 2005). However, the optimum volume ratio of the two gasses is still unclear. In our earlier results, the ratio 5:1 (O2:C3 F8 v/v) was verified as the optimal ratio, which could support more stable microbubbles to deliver the maximum oxygen payloads. In this study, OLMs (O2:C3 F8, 5:1 v/v) exhibited relativity strong stability in morphology and in its oxygen payload over time. A greater amount of oxygen was released in severe hypoxic conditions than in moderate hypoxic conditions in the absence of US 3 h after OLMs fabrication, producing an increase in the oxygen concentration of about 2-fold in 10 min, demonstrating that OLMs exhibit a varying degree of permeability under various hypoxic conditions. In fact, UTMD has been demonstrated as a controllable avenue for gene and drug delivery. According to the natural release of oxygen gas from OLMs, we believe that UTMD provides a new critical technology to push the clinical application of this novel particle. In general, UTMD has been strongly demonstrated to burst microbubbles to release genes or drugs directly at the targeted site (Fan et al. 2016). The use of US in our study as a mechanism for local microbubble fragmentation makes it possible for the oxygen to be released from the microbubbles directly. In our in vitro study, under a US probe, oxygen release was higher than that in the absence of US. These results indicate that
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
one property of OLMs is that they easily burst with US, which is the premise of oxygen release. Moreover, in this manner, US-favored oxygen delivery from OLMs and the release kinetics were decidedly quicker than those in the absence of US. Therefore, UTMD controllably induced the majority release of oxygen from OLMs at a fixed area, which offers a new opportunity for delivering oxygen directly to tumors or organs in vivo. The extent of the elevation of saturated oxygen in blood samples induced by OLMs was investigated in this study as well. The value of total CO2 for blood mixed with OLM was 60.1–61.1 mL, which was higher than the values of total CO2 of others (NS group: 47.1–53.0 mL, C3 F8 Ms group: 53.9–55.2 mL, control group: 37.0–43.6 mL). The results showed that OLMs could also release more oxygen payloads to the normal blood sample in the absence of US in 3 min, compared with other groups. In addition, the value of total CO2 in blood mixed with OLM was 23.1 mL higher than that of the control blood. Therefore, the OLMs described here can be injected into blood circulation in vivo, which provides the potential for blood substitutes. However, some results indicated that too much oxygen can cause oxygen toxicity (Aggarwal and Brower 2014). More in vivo studies are needed to screen the precise quantity of injectable microbubbles and evaluate their effects on the tissue and organism as a whole. Additionally, more studies are required to measure its sufficiency for general infusion in vivo. Of importance, higher quality images can improve the determination of the positions of tumors and can play a decisive role in the image-guided UTMD. A higher contrast in the image indicates a higher concentration of microbubbles at the targeted tissues and consequently, a higher oxygen payload delivered. Contrast imaging in vivo indicated a sufficient circulation time and homogenous contrast enhancement in tumors. The effective residence time lasted for about 5 min. The intensity in the tumor at 20 s after administration was nearly 8-fold the intensity before administration. Although the strongest enhancement time is limited to about 5 min, this is enough time for UTMD to deliver oxygen payloads. Owing to the quick enhancement and peak imaging reached within 15–25 s after injection, an optimal time for UTMD would be selected to be either 1 min or every 2–5 min for interrupted destruction. This imaging time range meets clinical radiotherapy requirements. It has long been known that hypoxic cells are more resistant to cancer therapy, and the means to overcome tumor hypoxia have been the subject of clinical investigations for many years. In our study, the application of oxygen-carried lipid microbubbles combined with US as a platform for image-guided oxygen delivery in vivo was further investigated, using rabbits bearing VX2 tumors. After intravenous injection of OLMs, pO2 levels in the
9
tumor tissue were highly elevated when OLMs were triggered by US locally. The added presence of US-activated OLM elicited a nearly 6-fold increase in pO2 levels of tumor at 1 min compared with that of pre-injection. On the other hand, non-triggered OLMs and C3 F8 Ms microbubbles did not cause such an effect. Although our results show that non-triggered OLMs could also carry more oxygen and can significantly increase the pO2 levels of venous blood in vitro, the tumor treated with non-triggered OLMs only underwent a small initial change in pO2 measurements at 1 min. Thus, as expected, when combined with US, this could become a strong technique used to elevate the oxygenation levels locally in tumors as soon as possible. If high oxygenation levels are sustained in the tumor when radiation occurs, the tumor radio-sensitivity would be largely enhanced. These results encourage further study with US-activated oxygen release for the radiotherapy sensitization in vivo. CONCLUSIONS In this study, the OLMs (O2:C3 F8, 5:1 v/v) exhibited excellent biomedical gas delivery capabilities to allow intra-venous administration, which is a leap toward clinical translation. This oxygen release mediated by UTMD strategy in vivo is a preliminarily attempt to identify an acoustic and oxygen-delivering biomaterial that would enable precise image-guided oxygen release in the local tissue. This study provided us with the foundation to further our investigations into in vivo cancer radiotherapysensitivity research. Acknowledgments—The authors thank Yang Liu, Yuping Gong, Li Cheng and Jingyu Chen for their help in the oxygen releasing and US imaging experiments. This work was supported by the Natural Science Foundation of Chongqing (No. cstc2015 jcyjA10090) and National Science Foundation (No. 81425014, No. 81630047 and No. 81571688).
REFERENCES Ackerman D, Simon MC. Hypoxia, lipids, and cancer: Surviving the harsh tumor microenvironment. Trends Cell Biol 2014;24:472–478. Aggarwal NR, Brower RG. Targeting normoxemia in acute respiratory distress syndrome may cause worse short-term outcomes because of oxygen toxicity. Ann Am Thorac Soc 2014;11:1449–1453. Bisazza A, Giustetto P, Rolfo A, Caniggia I, Balbis S, Guiot C, Cavalli R. Microbubble-mediated oxygen delivery to hypoxic tissues as a new therapeutic device. Conf Proc IEEE Eng Med Biol Soc 2008; 2008:2067–2070. Burke CW, Alexander E, 4th, Timbie K, Kilbanov AL, Price RJ. Ultrasound-activated agents comprised of 5 FU-bearing nanoparticles bonded to microbubbles inhibit solid tumor growth and improve survival. Mol Ther 2014;22:321–328. Cavalli R, Bisazza A, Giustetto P, Civra A, Lembo D, Trotta G, Guiot C, Trotta M. Preparation and characterization of dextran nanobubbles for oxygen delivery. Int J Pharm 2009;381:160–165. Dash RK, Bassingthwaighte JB. Erratum to: Blood HbO2 and HbCO2 dissociation curves at varied O2, CO2, pH, 2,3-DPG and temperature levels. Ann Biomed Eng 2010;38:1683–1701.
ARTICLE IN PRESS 10
Ultrasound in Medicine & Biology
Eisenbrey JR, Albala L, Kramer MR, Daroshefski N, Brown D, Liu JB, Stanczak M, O’Kane P, Forsberg F, Wheatley MA. Development of an ultrasound sensitive oxygen carrier for oxygen delivery to hypoxic tissue. Int J Pharm 2015;478:361–367. Fan CH, Chang EL, Ting CY, Lin YC, Liao EC, Huang CY, Chang YC, Chan HL, Wei KC, Yeh CK. Folate-conjugated gene-carrying microbubbles with focused ultrasound for concurrent blood-brain barrier opening and local gene delivery. Biomaterials 2016;106:46– 57. Goutelle S, Maurin M, Rougier F, Barbaut X, Bourguignon L, Ducher M, Maire P. The Hill equation: A review of its capabilities in pharmacological modeling. Fundam Clin Pharmacol 2008;22:633–648. Grenha A, Remuñán-López C, Carvalho EL, Seijo B. Microspheres containing lipid/chitosan nanoparticles complexes for pulmonary delivery of therapeutic proteins. Eur J Pharm Biopharm 2008;69:83–93. Hansen PJ. Use of a hemocytometer. Department of Animal Science, University of Florida; 2000. Kwan JJ, Kaya M, Borden MA, Dayton PA. Theranostic oxygen delivery using ultrasound and microbubbles. Theranostics 2012;2:1174– 1184. Li P, Song Y, Wang S, Tao Z, Yu S, Liu Y. Enhanced decolorization of methyl orange using zero-valent copper nanoparticles under assistance of hydrodynamic cavitation. Ultrason Sonochem 2015;22:132–138. Marinelli S, Salvatore V, Baron Toaldo M, Milazzo M, Croci L, Venerandi L, Pecorelli A, Palamà C, Diana A, Bolondi L, Piscaglia F. Evaluation of the impact of transient interruption of antiangiogenic treatment using ultrasound-based techniques in a murine model of hepatocellular carcinoma. BMC Cancer 2014;14:403. McEwan C, Owen J, Stride E, Fowley C, Nesbitt H, Cochrane D, Coussios CC, Borden M, Nomikou N, McHale AP, Callan JF. Oxygen carrying microbubbles for enhanced sonodynamic therapy of hypoxic tumours. J Control Release 2015;203:51–56. Oh SH, Ward CL, Atala A, Yoo JJ, Harrison S. Oxygen generating scaffolds for enhancing engineered tissue survival. Biomaterials 2009; 30:757–762. Rawat M, Singh D, Saraf S. Lipid carrier: A versatile delivery vehicle for proteins and peptides. Yakugaku Zasshi 2008;128:269– 280.
Volume ■■, Number ■■, 2017 Ren J, Xu C, Zhou Z, Zhang Y, Li X, Zheng Y, Ran H, Wang Z. A novel ultrasound microbubble carrying gene and Tat peptide: Preparation and characterization. Acad Radiol 2009;16:1457–1465. Riess JG. Understanding the fundamentals of perfluorocarbons and perfluorocarbon emulsions relevant to in vivo oxygen delivery. Artif Cells Blood Substit Immobil Biotechnol 2005;33:47–63. Rimoldi S, Terova G, Ceccuzzi P, Marelli S, Antonini M, Saroglia M. HIF-1α mRNA levels in Eurasian perch (Perca fluviatilis) exposed to acute and chronic hypoxia. Mol Biol Rep 2012;39:4009–4015. Ruan K, Song G, Ouyang G. Role of hypoxia in the hallmarks of human cancer. J Cell Biochem 2009;107:1053–1062. Scholz AW, Eberle B, Heussel CP, David M, Schmittner MD, Quintel M, Schreiber LM, Weiler N. Ventilation-perfusion ratio in perflubron during partial liquid ventilation. Anesth Analg 2010;110:1661– 1668. Song X, Feng L, Liang C, Yang K, Liu Z. Ultrasound triggered tumor oxygenation with oxygen-shuttle nanoperfluorocarbon to overcome hypoxia-associated resistance in cancer therapies. Nano Lett 2016;16:6145–6153. Swanson EJ, Borden MA. Injectable oxygen delivery based on proteinshelled microbubbles. Nano Life 2010;1:215–218. Swanson EJ, Mohan V, Kheir J, Borden MA. Phospholipid-stabilized microbubble foam for injectable oxygen delivery. Langmuir 2010; 26:15726–15729. Tokuhiro S, Ogawa Y, Tsuzuki K, Akima R, Ue H, Kariya S, Nishioka A. Development of a novel enzyme-targeting radiosensitizer (KORTUC) containing hydrogenperoxide for intratumoral injection for patients with low linear energy transfer-radioresistant neoplasms. Oncol Lett 2010;1:1025–1028. Tulard A, Hoffschir F, de Boisferon FH, Luccioni C, Bravard A. Persistent oxidative stress after ionizing radiation is involved in inherited radiosensitivity. Free Radic Biol Med 2003;35:68–77. Zhou Y, Gu H, Xu Y, Li F, Kuang S, Wang Z, Zhou X, Ma H, Li P, Zheng Y, Ran H, Jian J, Zhao Y, Song W, Wang Q, Wang D. Targeted antiangiogenesis gene therapy using targeted cationic microbubbles conjugated with CD105 antibody compared with untargeted cationic and neutral microbubbles. Theranostics 2015;5: 399–417.
https://doi.org/10.1016/j.ultrasmedbio.2017.08.1883
●
Original Contribution
LIPID MICROBUBBLES AS ULTRASOUND-STIMULATED OXYGEN CARRIERS FOR CONTROLLABLE OXYGEN RELEASE FOR TUMOR REOXYGENATION Chunjiang Yang,* Huan Xiao,* Yang Sun,† Lirong Zhu,* Yang Gao,* Sunny Kwok,‡ Zhigang Wang,† and Yi Tang* *
Department of Ultrasound, Children’s Hospital of Chongqing Medical University, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing Key Laboratory of Pediatrics, China International Science and Technology Cooperation base of Child Development and Critical Disorders, Chongqing, China; † Second Affiliated Hospital, Institute of Ultrasound Imaging, Chongqing Medical University, Chongqing, China; and ‡ Department of Biomedical Engineering, Ohio State University, Columbus, OH, USA (Received 25 March 2017; revised 22 August 2017; in final form 24 August 2017)
Abstract—Microbubbles are proposed as a potentially novel method for oxygen delivery in vivo in initial studies. The lack of commercial microbubbles for oxygen delivery in preclinical research prompted us to fabricate an oxygenloaded lipid microbubble. We aimed to extend the innovative strategy to modulate the tumor hypoxic microenvironment, using microbubbles intravenously as an oxygen carrier for the controllable tumor-specific delivery of oxygen by ultrasound (US). In our experiment, an oxygen-loaded lipid-coated microbubble (OLM) with mixed gas (O2/C3 F8, 5:1 v/v) was fabricated and exhibited a higher rate of oxygen release to a desaturated solution through burst by US than that in the absence of US. Although in in vivo studies, OLMs could be imaged and triggered by US to elevate the pO2 level in the breast VX2 tumor dramatically within a matter of minutes. The added presence of US-activated OLMs elicited a nearly six-fold increase in pO2 levels within 1 min compared with that of the pre-injection. Owing to the high oxygen payload, great acoustic stability and acoustic properties, OLMs may be proposed as an ideal radio-sensitizer. We conclude that oxygen release mediated by ultrasound-targeted microbubble destruction is feasible and shows potential in image-guided, site-specific cancer radiotherapy. (E-mail: [email protected]) © 2017 Published by Elsevier Inc. on behalf of World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound, Oxygen delivery, Lipid microbubble, Controllable release, Hypoxic tumor.
from achieving its full therapeutic potential (Ruan et al. 2009). The oxygenation level of the solid tumor plays a vital role in the radiation process. Some research has shown that the biological effects of ionizing radiation are largely mediated by reactive oxygen intermediates. Oxidative stress can lead to DNA damage and chromosomal instability, contributing to the cellular sensitivity to irradiation (Tulard et al. 2003). The oxygen enhancement ratio (OER) is often described as the degree of radiation sensitization and refers to the ratio of doses in the absence of oxygen to the doses in the presence of oxygen to provide the same cell survival. The OER for mammalian cells is 2.5–3.0; the seemingly low OER value of 2.6 corresponds with 50% of cells killed under hypoxic conditions as opposed to almost 99% of the cells killed under aerobic conditions with a radiation dose of 1000 cGy (Kwan et al. 2012). This information implies that if tumor cells with oxygen levels in the intermediate range can be reoxygenated, an increased sensitivity to radiation would result.
INTRODUCTION Hypoxia is a common phenomenon in most solid tumors because the oxygen supply is reduced by disturbed microcirculation and deteriorated diffusion. It is estimated that at least 50%–60% of advanced solid tumors contain hypoxic or anoxic tissue, typically because of irregularities in the tumor microcirculation (Ackerman and Simon 2014). When the transport of oxygen is unable to meet the demand, hypoxia occurs. Tumors are metabolically designed to thrive under hypoxic conditions, and the effectiveness of therapy is impaired by an inadequate oxygen supply in tumors because the hypoxia is associated with an increased risk of metastasis and therapeutic resistance (i.e., radiotherapy, chemotherapy and photodynamic therapy), which prevents cancer therapy
Address correspondence to: Yi Tang, 136#Zhongshanerlu, Chongqing 400014, China. E-mail: [email protected] 1
ARTICLE IN PRESS 2
Ultrasound in Medicine & Biology
It is crucial to improve the tumor oxygenation level during radiation therapy. Recently, various strategies have been proposed to oxygenate the tumor microenvironment to achieve better therapeutic outcomes. Traditional oxygen-delivering technologies fall into three categories: hyperbaric oxygen therapy (HBO2), oxygen-carrying materials and oxygen-generating materials. While various methods have been investigated to deliver O2 in vivo, no specific technology has provided an effective way for oxygen delivery into a targeted tissue. A recent study found that ultrasound (US) could favor nano-perfluorocarbon for tumor-specific delivery of oxygen and could dramatically enhance tumor oxygenation, thus improving therapeutic outcomes in radiation therapy treatment of tumors (Song et al. 2016). However, like most traditional technologies, it depends on the respiratory system and adsorbed oxygen in the lung to release oxygen into the circulation. Thus, such studies were used to construct more targeted specifications for oxygen-delivering biomaterials to avoid oxygen exchange through the respiratory system and determine suitable technologies for clinical translation. An emerging new technology that has extensive use clinically involves the microbubble as a US contrast agent and an oxygen carrier simultaneously. Because of the microbubble’s acoustic activity, US can be used to stimulate and image gas release in a targeted region. Another unique characteristic of microbubbles is that they can be burst locally using US in vivo, releasing an oxygen payload to a particular area of interest. These particles consist of gaseous cores surrounded by thin shells that are composed of lipids, polymers, surfactants, dextran and chitosan (Bisazza et al. 2008; Eisenbrey et al. 2015; Swanson et al. 2010). Among these coatings, results from many studies expressed an interest in lipid membranes; phospholipid monolayers present a relatively minor impediment to gas diffusion. Within the past decade, a few studies have focused on lipid-oxygen-loaded microbubbles. However, most studies are still performed in vitro and focus on fabrication. Few studies investigate the precise oxygen payload delivery to the hypoxic tumor. Although some results demonstrate that sonodynamic therapy may be significantly enhanced by incorporating an approach that involves US and microbubble-mediated delivery of oxygen to hypoxic tumor, the authors only compared the indirect parameter (i.e., the singlet oxygen generation) in cell-free systems before or after US exposure (McEwan et al. 2015). Furthermore, the ideal components of said gas cores have not been conclusively agreed upon. Pursuing this idea further we have developed oxygen-loaded lipid-coated microbubbles (OLM) and investigated its stability in terms of structural characterizations and ability to retain oxygen payload over time. To enlarge the range of OLM’s clinical application, OLMs are intravenously injected into tumor-bearing rabbits. With a low-power clinically adapted
Volume ■■, Number ■■, 2017
US transducer applied on their tumors, the elevation of oxygenation level in tumors was monitored. METHODS Implantation of VX2 breast tumors We anesthetized New Zealand white rabbits that weighed 2.0–2.5 kg (Laboratory Animal Center of Chongqing Medical University, Chongqing, China) with an intra-muscular injection of 3% pentobarbital solution (1 mL/kg). The abdomens of the rabbits were depilated with 8% Na2 S solution, then fixed in prone position and routinely disinfected. The VX2 tumor is a squamous cell carcinoma similar to human cancer in morphology and biologic characteristics, such as rapid growth, which makes the VX2 tumor a good model for experimental research. The tumor model could be built directly from obtained tumor blocks (from tumor-bearing rabbits) for transplantation. The experimental cost in using this method is lower than the cost of the cell culture method, and the technical operation is relatively simple and easy to promote. The VX2 tumor tissue was excised from the tumor-bearing rabbit and soaked in 20 mL Hanks’ balanced salt solution. Then the tumor was sheared to small masses with a size of approximately 0.5–1.0 mm in diameter. The final suspension was extracted into a 20-mL syringe, and a 1-mL tumor tissue suspension was injected into the mammary gland of the rabbits underneath the second left nipple. Each rabbit was given 500,000 units of streptomycin after the tumor implantation. Two weeks after implantation, the tumors reached 80–100 mm3 in volume. These tumorbearing rabbits were used for investigations of contrast imaging and oxygen release in vivo. All procedures involving the use and care of rabbits were approved ethically and scientifically by Chongqing Medical University in compliance with the Practice Guidelines for Laboratory Animals of China. Microbubble construction and preparation The compounds 1,2-distearoyl–sn-glycero-3phosphocholine (DSPC) and 1,2-dipalmitoyl-rac-glycero3-phospho (dimethylaminoethanol)–methoxy polyethylene glycol 5000 (DPPE-MPEG5000) (Avanti Polar Lipids, Inc., Alabaster, AL, USA) were synthesized in a 9:1 molar ratio and mixed with phosphate-buffered saline (PBS) and glycerin in a final volume of 0.5 mL in 1.5 mL vials. Parafilm was used to seal the vial. The mixtures were fully dissolved at 55°C under a thermostat water bath for 30 min. Then the headspace of each vial was filled with perfluoropropane (C3 F8) gas, highpurity oxygen (99.999%) and pure nitrogen (N2) with a gas volume ratio of O2:C3 F8 of 5:1 v/v to replace the air in the vials under a vacuum with a syringe. The dispersion in the vials was mechanically shaken in a dental
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
amalgamator (YJT, Medical Apparatus and Instrument Ltd., Shanghai, China), blending for 50 s at 4000 rpm, as described by Ren et al. (2009). After 1.5 mL of PBS was added to each vial, the suspension was diluted and stored at 4°C for further use. Morphology and size distribution The morphology of OLM was observed under an optical microscope (CX41, Olympus Corporation, Shinjuku, Tokyo, Japan). The concentration of OLM was calculated using a Hemocytometer (BLAUBRAND counting chamber, Wertheim, Germany) instrument during each observation, and the concentration of OLMs is expressed as follows (Hansen 2000):
Microbubbles concentration per milliliter = Microbubbles within 4 major squares × 2500 (1) × Dilution factor The diameter of OLMs was evaluated using the Mastersizer 3000, a laser particle-size analyzer (Malvern Instruments Ltd., Malvern, Worcestershire, UK), at 1, 3 and 7 d after fabrication, and the concentration was calculated 0.5 h, 1 d, 3 d and 7 d after fabrication (Li et al. 2015). To obtain accurate measurements, each syringe was inverted 5 times to be sure of heterogeneous emulsion before measurements, which were collected 3 times per test. The particle charge was qualified as zeta potential, using the Zetasizer 3000, a laser Doppler anemometry (Malvern Instruments Ltd.) 1 d after fabrication (Burke et al. 2014; Grenha et al. 2008). The pH values were measured using the Mettler Toledo Delta320 (Mettler-Toledo, Greifensee, Switzerland) at 0.5 h, 1 d, 3 d and 7 d after fabrication. In vitro oxygen release with or without US exposure Natural release of oxygen in hypoxic solutions. Normal saline solution (NaCl, NS) with an oxygen concentration of 0.9% was sealed in a vial, then the O2 concentration was reduced to 0.4 mg/L (severe hypoxia) (Cavalli et al. 2009; Rimoldi et al. 2012) by means of an N2 purge (Cavalli et al. 2009; Oh et al. 2009). A total of 3 mL of each type of OLMs were introduced into 20 mL of NS. All OLMs were stored at 4°C. Experiments were carried out after 0.5 h, 3 h and 1 d after OLM preparation (3 groups). The oxygen-release kinetic was monitored for 10 min using an oximeter (HQ30 d, HACH, Loveland, CO, USA) for each group, and the dissolved oxygen data of each solution was recorded every 20 s. The experiments were repeated in triplicate for each independent sample.
3
Oxygen release of OLMs triggered by US in hypoxic solutions The O2 concentration in a sealed vial containing NS solution was reduced to 4.0 mg/L by means of a N2 purge to mimic moderate hypoxic conditions. The solutions were divided into the following two groups: (a) OLM only and (b) OLM with US exposure. Also, using the same procedure described earlier, 3 mL of the OLM was introduced into 20 mL of NS 3 h after fabrication. The oxygenrelease kinetic was monitored for 10 min using an oximeter, and dissolved oxygen data of the solution were recorded every 20 s. For treatment of group b, continuous US pulses with a spatial average intensity of 0.5 W/cm2 were applied to the bottom of the vials for 10 min immediately after the OLMs were added to the solution. The US transducer was unfocused and consisted of a single element (UGT1025, Institute of US Imaging of Chongqing Medical University, Chongqing, China) with a diameter of 2 cm and was applied at 1 MHz, 3.5 W/cm2 and 50% duty cycle for 30 s (Zhou et al. 2015). The US was combined with an airtight apparatus consisting of organic glass and sealing rings. The oximeter was calibrated in the air before the experiment, and all experiments were performed at a stable room temperature (25°C) and humidity conditions. The experiments were repeated in triplicate for 3 independent samples and then averaged.
Quantitative evaluation of dissolved oxygen in blood samples Fresh healthy human venous blood was collected in heparin-containing vacuum tubes, then gathered in 6 pipettes, each having a volume of 3 mL, and stored at 4°C in a refrigerator (collection occurred immediately before the blood-gas analysis). The treatment groups were divided into 4 groups: (i) OLM, (ii) C3 F8 Ms microbubble, (iii) NS and (iv) the control group (just the blood sample). We then gently mixed 1 mL of the corresponding OLM or NS with 1 mL of the blood sample, collected by use of a 2.5 mL disposable syringe and sealed immediately, for approximately 3 min. The dissolved oxygen partial pressure (pO2) was immediately measured using a blood-gas analyzer (ABL510, Radiometer Medical, Copenhagen, Denmark). Each sample was taken using a disposable syringe at room temperature (25°C), and the needle was immediately plugged with a rubber stopper to avoid contamination from the air. The experiments were repeated in triplicate for 3 independent samples and then averaged. In addition, based on the hemoglobin saturation, the oxygen volume in the blood sample can be calculated as follows (Scholz et al. 2010):
CO2 = (a × pO2 + b × Hb ×Sat )×V
(2)
ARTICLE IN PRESS 4
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2017
where a = O2 solubility in blood (0.003 mL/mmHg); b = oxygen-carrying capacity of Hb (1.34 mL/g); Hb = concentration of hemoglobin in blood (15 g/dL); V = blood volume (3 L); and Sat = saturation of hemoglobin. In addition, the saturation of hemoglobin in blood at pH 7.4 and 37°C is expressed as follows (Dash and Bassingthwaighte 2010):
Sat = K O 2 × pO2 n
(1+ K O 2 × pO2 n )
(3)
where KO2 = Hill coefficient; n = Hill exponent (n = 2.7); KO2 = (P50)n; and P50 = 26.6 mmHg (Goutelle et al. 2008). Human venous blood was collected from healthy donors in our research team. Each participant provide informed consent. Contrast imaging in the solid tumor To evaluate the contrast enhancement of OLM in a solid tumor in vivo, three rabbits inoculated with the VX2 tumor line (squamous carcinoma) were fixed in a supine position after sedation with an intra-muscular injection of 3% pentobarbital solution (1 mL/kg). An LA522 linear transducer (Mylab Twice, Esaote, Genova, Italy) equipped with contrast pulse sequence technology (CnTI technology, Esaote) was performed on the tumor surface. The parameters of the time-gain compensation and focus range were adjusted to the optimum. Each rabbit received a bolus injection of 0.02 mL/kg of the OLMs suspension through a 20-gauge catheter in an ear vein. The tubing was flushed with 2 mL NS after injection of OLMs. Real-time scanning started immediately after injection of the microbubbles. The tumors’ images were then continually recorded for 5 min. The dynamic and static images were stored for further analysis. Quantitative analysis was detected, using the US Image Analyzer (DFY-II type, Institute of US Imaging of Chongqing Medical University, Chongqing, China) (Ren et al. 2009). Before and after contrast, the appropriate regions of interest in the tumor were selected. The echo intensity of the regions of interest and respective evaluations were all marked on the quantitative images. Determination of pO2 in VX2 tumor The area above the breast tumor of a total of three rabbits was carefully depilated with 8% Na2 S again after anesthesia. The tumor was located and measured by US before microbubble injection. The study was conducted with C3 F8 Ms as the control group compared with OLMs before injection and 5 s after injection without US exposure, or immediately after US triggered after injection. The working electrode (length: 50 mm, diameter: 0.8 mm, cable length: 500 mm, oxygen consumption rate: 10 ng/h) was introduced into the center of the tumors before agent injection. The calibration process was done before every
Fig. 1. Scheme of intravenous injection of OLMs triggered by image-guided US and monitoring pO2 level in tumor. OLMs = oxygen-loaded lipid-coated microbubbles; US = ultrasound.
measurement by bubbling 100% nitrogen gas and 20% oxygen gas, respectively, into the beaker. The OLM injection was performed after the total destruction of the C3 F8 Ms. In brief, the rabbits received an injection of 0.5 mL microbubbles, then 1.5 mL NS. The intra-tumoral oxygen concentration was recorded every 6 s for 60 s by an IMP211 apparatus (Inter Medical Co., Ltd., Shinagawaku, Tokyo, Japan) (Tokuhiro et al. 2010) before injection, similarly recorded 5 s after microbubble injection without US exposure and immediately after US-triggered bursting. To record pO2 after US-triggered bursting, every 5 s for 30 s after agent injection, the microbubbles present in the field of view containing the tumor were destroyed by temporarily (1 s) increasing the acoustic power (MI 1.9) (Marinelli et al. 2014) with an LA523 broadband linear array transducer (frequency range, 4–13 MHz). The B-mode and contrast-enhanced harmonic images mechanical index (MI, 0.06, derated pressure, 50 kPa) were concurrently obtained for each sample and all imaging parameters were kept constant for all imaging sessions (Mylab 90, Esaote). The process was guided by US imaging (Fig. 1).
Statistical analysis All data are shown as the mean ± standard deviation (SD). Statistical analysis was performed using SPSS17.0 software (IBM, New York, NY, USA) and GraphPad Prism 5.01 (GraphPad Software, La Jolla, CA, USA). Statistical significance among multiple groups was determined using a one-way analysis of variance with StudentNewman-Keuls test multiple comparison post-tests. A paired-samples t-test was used to analyze the pre-contrast and post-contrast echo intensities of tumors, and an unpaired t-test was used to analyze the oxygen-release of OLMs with and without US exposure. In all analyses, p < 0.05 was taken as indicative of statistical significance.
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
5
Fig. 2. Physical characteristics of OLMs. (a) Light microscopy images of OLMs (×200). (b) Size distribution plot of OLMs 1 d after preparation. OLMs = oxygen-loaded lipid-coated microbubbles.
RESULTS The characterizations of OLMs Some physical characteristics of OLMs are shown in Figure 2. The morphology of OLMs remained spherical under a microscope. The average diameter was 1033 ± 72 nm, concentration was (3.79 ± 0.32) × 109/mL, zeta potential of OLM was −12.9 ± 2.4 mV and the pH measurements of the OLM was approximately 7.120 ± 0.078 1 d after their fabrication.
The diameters, concentrations and pH values of OLM formulation did not change significantly 3 h, 1 d, 3 d and 7 d after fabrication (p > 0.05) (Table 1). The oxygen-release kinetic of OLMs with or without US exposure Oxygen-release kinetics in the presence of OLMs was observed 0.5 h, 3 h and 1 d after the microbubble
Table 1. Physicochemical characteristics of OLMs over time
Diameter ± SD (nm) Concentration ± SD (×109/mL) pH ± SD
0.5 h
1d
3d
7d
— 4.04 ± 0.87 7.12 ± 0.08
1033 ± 72 3.79 ± 0.32 7.07 ± 0.03
1069 ± 53 3.39 ± 0.31 7.07 ± 0.08
1055 ± 89 3.41 ± 0.34 7.010 ± 0.03
OLMs = oxygen-loaded lipid-coated microbubbles.
ARTICLE IN PRESS 6
Ultrasound in Medicine & Biology
Volume ■■, Number ■■, 2017
Fig. 3. Dissolved oxygen–time plots for 10 min immediately after OLMs were introduced into the various hypoxic solutions. (a) The severe hypoxic solution (0.4 mg/L) 0.5 h, 3 h and 1 d stored at 4°C after OLM fabrication. (b) The moderate hypoxic solution (4.0 mg/L) with or without US exposure. OLMs = oxygen-loaded lipid-coated microbubbles; US = ultrasound.
preparation. OLMs showed the capacity for oxygen release throughout their preparation over time. The 10-min oxygen increase by OLMs 0.5 h, 3 h and 1 d after their fabrication are reported in Figure 3a. A high amount of oxygen was released in severe hypoxic conditions (0.4 mg/L), producing an increase in the oxygen concentration of about 6- to 12-fold within 10 min. The oxygen concentration of hypoxic solutions increased markedly 1 min after the injection of OLMs individually and then decreased slowly to steady-state within 10 min. The oxygen release profiles of OLMs 0.5 h, 3 h and 1 d after fabrication were very similar and did not significantly change throughout time, showing the oxygen payload stability of the OLMs stored at 4°C. A greater amount of oxygen was released in severe hypoxic conditions (3.80 ± 0.90 mg/L) than in moderate hypoxic conditions (1.86 ± 0.66 mg/L) in the absence of
US 3 h after fabrication, producing an increase in the oxygen concentration of about two-fold in 10 min. In the moderately hypoxic solution, the OLMs can release oxygen and increase the dissolved oxygen of the solution by approximately 0.5-fold, about 1.86 ± 0.66 mg/L without US; however, the oxygen concentration increased by approximately 1.2-fold, about 3.88 ± 0.83 mg/L in the presence of US (Fig. 3b). Blood oxygenation by use of OLMs Dissolved oxygen concentration plots for OLMs introduced into the blood sample for 10 min are shown in Figure 4. The average pO2 in blood diluted with OLM was 19.53 ± 1.39 kPa, the value of average pO2 in control blood (without dilution) was 4.63 ± 0.42 kPa, and the values in blood diluted with pure C3 F8 microbubbles (C3 F8 Ms)
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
7
Fig. 4. Dissolved oxygen–time plots for OLM introduced into the blood sample. Comparison of pO2 increases in blood with OLMs in the blood sample without sonication (*p < 0.05). OLMs = oxygenloaded lipid-coated microbubbles.
and in NS were (7.93 ± 0.31 kPa) and (6.43 ± 0.80 kPa), respectively. OLMs caused an obvious increase in the pO2 of the blood sample in vitro (p < 0.05). However, the other groups exhibited no obvious effect on the pO2 of the blood sample (p > 0.05). In this study, the pO2 of blood in the OLMs group was improved from 4.2 kPa to 18.1~20.9 kPa (135.8~156.8 mmHg). Given this data at 37°C, with pH 7.4, the saturation value of hemoglobin was calculated to be between 98.8% ~ 99.2%, and the value of the total Co2 for blood diluted with OLMs (Sat) was 60.1 ~ 61.1 mL. For the NS group, the Sat was calculated to be 77.4% ~ 87.1%, and the value of total Co2 was 47.1 ~ 53.0 mL. For the C3 F8 Ms group, the Sat was calculated to be 88.7% ~ 90.5%, and the value of total Co2 was 53.9 ~ 55.2 mL. For the control group, the Sat was calculated to be 60.9% ~ 71.8%, and the value of total Co2 was 37.0 ~ 43.6 mL. The value of total Co2 for blood mixed with OLMs was higher than that with the NS, the C3 F8 Ms and nothing. The results show that OLMs could significantly increase the pO2 value of venous blood. Contrast imaging in vivo and quantitative analysis This study included a total of three rabbits, which received OLM and C3 F8 Ms successively by injections into the ear vein. The tumor US images pre- and post-injection of OLMs, using DFY (Institute of US Imaging of Chongqing Medical University) analysis are shown in Figure 5. The echo intensity of the tumor was enhanced and significant enhancements were observed for 5 min. A contrast enhancement of the tumor appeared quickly in 20 s. After injecting OLMs, the tumor was initially enhanced (approximately 5 s after injection). The tumor parenchyma then exhibited uniformly quick enhancement, the imaging reached a peak within 15–25 s and the echo intensity slightly decreased. The contrast-enhanced process lasted for more than 5 min. In a single imaging plane, 20 separate regions of interest were selected within the tumor
Fig. 5. Ultrasound contrast imaging and quantitative analysis image of the echo intensity in VX2 breast tumor. (a) Pre-contrast echo intensity. (b) Post-contrast echo intensity. Tumor location (indicated by lines).
of each rabbit to compute the mean dB. The average intensity in the tumor before administration was 14.03 ± 3.00 dB. The intensity in the tumor at 20 s after administration was 115.85 ± 34.83 dB. In vivo experimentation showed that OLMs could strongly enhance the echo intensity of the solid tumor within 5 min (p < 0.05). Elevation of pO2 in VX2 tumors triggered by US We further assessed the pO2-increasing ability of OLMs as the carrier for the oxygen delivery using rabbits bearing VX2 tumor. After intravenous injection of OLMs, using C3 F8 Ms as the control group, the tumor was burst by US and the pO2 value within the tumors were monitored at each point during this process. As shown in Figure 6, pO2 values have no obvious change before and after the injection of OLMs (p = 0.777) or C3 F8 Ms microbubbles (p = 0.999). However, after US irradiation, significant enhancement of pO2 in the breast tumor tissue in the OLMs group was generated at 1 min in comparison with pre-injection, post-injection and triggered C3 F8 Ms microbubbles (p < 0.05). After triggering by US, the pO2 of the tumors in OLMs group were recorded to be 364.37 ± 80.48 mmHg compared with 54.49 ± 12.05 mmHg
ARTICLE IN PRESS 8
Ultrasound in Medicine & Biology
Fig. 6. Determination of pO2 within VX2 tumor in rabbits. (a) The measurements of pO2 in tumors treated with OLMs and ultrasound activating. Tumors with triggered OLMs showed a highest pO2 levels (*p < 0.01). (b) The measurements of pO2 in tumors treated with C3 F8 Ms and ultrasound activating. No difference was found among three statuses (p > 0.05). OLMs = oxygen-loaded lipidcoated microbubbles.
before US triggering, while the C3 F8 Ms group did not significantly increase after being triggered by US (p = 0.974). The added presence of US-activated OLMs elicited a nearly six-fold increase in pO2 levels at 1 min compared with that of pre-injection, while the C3 F8 Ms microbubbles did not cause such an effect. DISCUSSION Microbubbles have attractive prospects for local oxygen delivery in reversing hypoxic organs or tumors by using ultrasound-targeted microbubble destruction (UTMD), which may provide life-sustaining doses of oxygen in the event of airway distress and help elevate the local oxygenation level within a short period of time. However, no commercial microbubbles are available for oxygen delivery in preclinical research. Among current studies, few provided the quantitative evaluation of pO2 in vivo. In exploring an oxygen-loaded microbubble suitable for this preclinical study, we fabricated OLMs with
Volume ■■, Number ■■, 2017
DSPC as the main coating component and a mixed-gas core (O2:C3 F8, 5:1 v/v). DSPC has been recognized as the optimum shell for oxygen containment and exhibits high biocompatibility (Rawat et al. 2008; Swanson and Borden 2010). The compression elasticity of DSPC is relatively stronger, and DSPC could resist the size and shape deformation of microbubbles during the gas exchange to carry oxygen more stably. Moreover, DSPC retained its gas for a far longer period and still contained more than 30% gas of volume at the end of testing. These results favor DSPC as the main lipid for US molecular imaging and oxygen carriage. Another critical factor in the design of oxygenloaded microbubbles is the gas core, which is strongly associated with the bubble size. Bubble size needs to be controlled both at the time of injection and throughout the circulation lifetime; bubble growth or aggregation in the circulation must be avoided. Most initial studies introduced pure oxygen into microbubbles. However, early observations showed that, while pure oxygen microbubbles could be generated, they resulted in low yields and were generally unstable. Our earlier findings showed that, with the lowest concentration, such pure oxygen microbubbles were unstable for oxygen delivery and contrast imaging. Oxygen as the gas phase is easily soluble in water and dissolves rapidly in blood under the mutual influence of Laplace pressure and arterial pressure. Fluorocarbon gases could hinder microbubble dissolution significantly because of their low aqueous solubility and diffusivity. Thus, they could increase the half-life of the bubbles from a few s to several min (Riess 2005). However, the optimum volume ratio of the two gasses is still unclear. In our earlier results, the ratio 5:1 (O2:C3 F8 v/v) was verified as the optimal ratio, which could support more stable microbubbles to deliver the maximum oxygen payloads. In this study, OLMs (O2:C3 F8, 5:1 v/v) exhibited relativity strong stability in morphology and in its oxygen payload over time. A greater amount of oxygen was released in severe hypoxic conditions than in moderate hypoxic conditions in the absence of US 3 h after OLMs fabrication, producing an increase in the oxygen concentration of about 2-fold in 10 min, demonstrating that OLMs exhibit a varying degree of permeability under various hypoxic conditions. In fact, UTMD has been demonstrated as a controllable avenue for gene and drug delivery. According to the natural release of oxygen gas from OLMs, we believe that UTMD provides a new critical technology to push the clinical application of this novel particle. In general, UTMD has been strongly demonstrated to burst microbubbles to release genes or drugs directly at the targeted site (Fan et al. 2016). The use of US in our study as a mechanism for local microbubble fragmentation makes it possible for the oxygen to be released from the microbubbles directly. In our in vitro study, under a US probe, oxygen release was higher than that in the absence of US. These results indicate that
ARTICLE IN PRESS Ultrasound-guided oxygen delivery for tumor reoxygenation ● C. Yang et al.
one property of OLMs is that they easily burst with US, which is the premise of oxygen release. Moreover, in this manner, US-favored oxygen delivery from OLMs and the release kinetics were decidedly quicker than those in the absence of US. Therefore, UTMD controllably induced the majority release of oxygen from OLMs at a fixed area, which offers a new opportunity for delivering oxygen directly to tumors or organs in vivo. The extent of the elevation of saturated oxygen in blood samples induced by OLMs was investigated in this study as well. The value of total CO2 for blood mixed with OLM was 60.1–61.1 mL, which was higher than the values of total CO2 of others (NS group: 47.1–53.0 mL, C3 F8 Ms group: 53.9–55.2 mL, control group: 37.0–43.6 mL). The results showed that OLMs could also release more oxygen payloads to the normal blood sample in the absence of US in 3 min, compared with other groups. In addition, the value of total CO2 in blood mixed with OLM was 23.1 mL higher than that of the control blood. Therefore, the OLMs described here can be injected into blood circulation in vivo, which provides the potential for blood substitutes. However, some results indicated that too much oxygen can cause oxygen toxicity (Aggarwal and Brower 2014). More in vivo studies are needed to screen the precise quantity of injectable microbubbles and evaluate their effects on the tissue and organism as a whole. Additionally, more studies are required to measure its sufficiency for general infusion in vivo. Of importance, higher quality images can improve the determination of the positions of tumors and can play a decisive role in the image-guided UTMD. A higher contrast in the image indicates a higher concentration of microbubbles at the targeted tissues and consequently, a higher oxygen payload delivered. Contrast imaging in vivo indicated a sufficient circulation time and homogenous contrast enhancement in tumors. The effective residence time lasted for about 5 min. The intensity in the tumor at 20 s after administration was nearly 8-fold the intensity before administration. Although the strongest enhancement time is limited to about 5 min, this is enough time for UTMD to deliver oxygen payloads. Owing to the quick enhancement and peak imaging reached within 15–25 s after injection, an optimal time for UTMD would be selected to be either 1 min or every 2–5 min for interrupted destruction. This imaging time range meets clinical radiotherapy requirements. It has long been known that hypoxic cells are more resistant to cancer therapy, and the means to overcome tumor hypoxia have been the subject of clinical investigations for many years. In our study, the application of oxygen-carried lipid microbubbles combined with US as a platform for image-guided oxygen delivery in vivo was further investigated, using rabbits bearing VX2 tumors. After intravenous injection of OLMs, pO2 levels in the
9
tumor tissue were highly elevated when OLMs were triggered by US locally. The added presence of US-activated OLM elicited a nearly 6-fold increase in pO2 levels of tumor at 1 min compared with that of pre-injection. On the other hand, non-triggered OLMs and C3 F8 Ms microbubbles did not cause such an effect. Although our results show that non-triggered OLMs could also carry more oxygen and can significantly increase the pO2 levels of venous blood in vitro, the tumor treated with non-triggered OLMs only underwent a small initial change in pO2 measurements at 1 min. Thus, as expected, when combined with US, this could become a strong technique used to elevate the oxygenation levels locally in tumors as soon as possible. If high oxygenation levels are sustained in the tumor when radiation occurs, the tumor radio-sensitivity would be largely enhanced. These results encourage further study with US-activated oxygen release for the radiotherapy sensitization in vivo. CONCLUSIONS In this study, the OLMs (O2:C3 F8, 5:1 v/v) exhibited excellent biomedical gas delivery capabilities to allow intra-venous administration, which is a leap toward clinical translation. This oxygen release mediated by UTMD strategy in vivo is a preliminarily attempt to identify an acoustic and oxygen-delivering biomaterial that would enable precise image-guided oxygen release in the local tissue. This study provided us with the foundation to further our investigations into in vivo cancer radiotherapysensitivity research. Acknowledgments—The authors thank Yang Liu, Yuping Gong, Li Cheng and Jingyu Chen for their help in the oxygen releasing and US imaging experiments. This work was supported by the Natural Science Foundation of Chongqing (No. cstc2015 jcyjA10090) and National Science Foundation (No. 81425014, No. 81630047 and No. 81571688).
REFERENCES Ackerman D, Simon MC. Hypoxia, lipids, and cancer: Surviving the harsh tumor microenvironment. Trends Cell Biol 2014;24:472–478. Aggarwal NR, Brower RG. Targeting normoxemia in acute respiratory distress syndrome may cause worse short-term outcomes because of oxygen toxicity. Ann Am Thorac Soc 2014;11:1449–1453. Bisazza A, Giustetto P, Rolfo A, Caniggia I, Balbis S, Guiot C, Cavalli R. Microbubble-mediated oxygen delivery to hypoxic tissues as a new therapeutic device. Conf Proc IEEE Eng Med Biol Soc 2008; 2008:2067–2070. Burke CW, Alexander E, 4th, Timbie K, Kilbanov AL, Price RJ. Ultrasound-activated agents comprised of 5 FU-bearing nanoparticles bonded to microbubbles inhibit solid tumor growth and improve survival. Mol Ther 2014;22:321–328. Cavalli R, Bisazza A, Giustetto P, Civra A, Lembo D, Trotta G, Guiot C, Trotta M. Preparation and characterization of dextran nanobubbles for oxygen delivery. Int J Pharm 2009;381:160–165. Dash RK, Bassingthwaighte JB. Erratum to: Blood HbO2 and HbCO2 dissociation curves at varied O2, CO2, pH, 2,3-DPG and temperature levels. Ann Biomed Eng 2010;38:1683–1701.
ARTICLE IN PRESS 10
Ultrasound in Medicine & Biology
Eisenbrey JR, Albala L, Kramer MR, Daroshefski N, Brown D, Liu JB, Stanczak M, O’Kane P, Forsberg F, Wheatley MA. Development of an ultrasound sensitive oxygen carrier for oxygen delivery to hypoxic tissue. Int J Pharm 2015;478:361–367. Fan CH, Chang EL, Ting CY, Lin YC, Liao EC, Huang CY, Chang YC, Chan HL, Wei KC, Yeh CK. Folate-conjugated gene-carrying microbubbles with focused ultrasound for concurrent blood-brain barrier opening and local gene delivery. Biomaterials 2016;106:46– 57. Goutelle S, Maurin M, Rougier F, Barbaut X, Bourguignon L, Ducher M, Maire P. The Hill equation: A review of its capabilities in pharmacological modeling. Fundam Clin Pharmacol 2008;22:633–648. Grenha A, Remuñán-López C, Carvalho EL, Seijo B. Microspheres containing lipid/chitosan nanoparticles complexes for pulmonary delivery of therapeutic proteins. Eur J Pharm Biopharm 2008;69:83–93. Hansen PJ. Use of a hemocytometer. Department of Animal Science, University of Florida; 2000. Kwan JJ, Kaya M, Borden MA, Dayton PA. Theranostic oxygen delivery using ultrasound and microbubbles. Theranostics 2012;2:1174– 1184. Li P, Song Y, Wang S, Tao Z, Yu S, Liu Y. Enhanced decolorization of methyl orange using zero-valent copper nanoparticles under assistance of hydrodynamic cavitation. Ultrason Sonochem 2015;22:132–138. Marinelli S, Salvatore V, Baron Toaldo M, Milazzo M, Croci L, Venerandi L, Pecorelli A, Palamà C, Diana A, Bolondi L, Piscaglia F. Evaluation of the impact of transient interruption of antiangiogenic treatment using ultrasound-based techniques in a murine model of hepatocellular carcinoma. BMC Cancer 2014;14:403. McEwan C, Owen J, Stride E, Fowley C, Nesbitt H, Cochrane D, Coussios CC, Borden M, Nomikou N, McHale AP, Callan JF. Oxygen carrying microbubbles for enhanced sonodynamic therapy of hypoxic tumours. J Control Release 2015;203:51–56. Oh SH, Ward CL, Atala A, Yoo JJ, Harrison S. Oxygen generating scaffolds for enhancing engineered tissue survival. Biomaterials 2009; 30:757–762. Rawat M, Singh D, Saraf S. Lipid carrier: A versatile delivery vehicle for proteins and peptides. Yakugaku Zasshi 2008;128:269– 280.
Volume ■■, Number ■■, 2017 Ren J, Xu C, Zhou Z, Zhang Y, Li X, Zheng Y, Ran H, Wang Z. A novel ultrasound microbubble carrying gene and Tat peptide: Preparation and characterization. Acad Radiol 2009;16:1457–1465. Riess JG. Understanding the fundamentals of perfluorocarbons and perfluorocarbon emulsions relevant to in vivo oxygen delivery. Artif Cells Blood Substit Immobil Biotechnol 2005;33:47–63. Rimoldi S, Terova G, Ceccuzzi P, Marelli S, Antonini M, Saroglia M. HIF-1α mRNA levels in Eurasian perch (Perca fluviatilis) exposed to acute and chronic hypoxia. Mol Biol Rep 2012;39:4009–4015. Ruan K, Song G, Ouyang G. Role of hypoxia in the hallmarks of human cancer. J Cell Biochem 2009;107:1053–1062. Scholz AW, Eberle B, Heussel CP, David M, Schmittner MD, Quintel M, Schreiber LM, Weiler N. Ventilation-perfusion ratio in perflubron during partial liquid ventilation. Anesth Analg 2010;110:1661– 1668. Song X, Feng L, Liang C, Yang K, Liu Z. Ultrasound triggered tumor oxygenation with oxygen-shuttle nanoperfluorocarbon to overcome hypoxia-associated resistance in cancer therapies. Nano Lett 2016;16:6145–6153. Swanson EJ, Borden MA. Injectable oxygen delivery based on proteinshelled microbubbles. Nano Life 2010;1:215–218. Swanson EJ, Mohan V, Kheir J, Borden MA. Phospholipid-stabilized microbubble foam for injectable oxygen delivery. Langmuir 2010; 26:15726–15729. Tokuhiro S, Ogawa Y, Tsuzuki K, Akima R, Ue H, Kariya S, Nishioka A. Development of a novel enzyme-targeting radiosensitizer (KORTUC) containing hydrogenperoxide for intratumoral injection for patients with low linear energy transfer-radioresistant neoplasms. Oncol Lett 2010;1:1025–1028. Tulard A, Hoffschir F, de Boisferon FH, Luccioni C, Bravard A. Persistent oxidative stress after ionizing radiation is involved in inherited radiosensitivity. Free Radic Biol Med 2003;35:68–77. Zhou Y, Gu H, Xu Y, Li F, Kuang S, Wang Z, Zhou X, Ma H, Li P, Zheng Y, Ran H, Jian J, Zhao Y, Song W, Wang Q, Wang D. Targeted antiangiogenesis gene therapy using targeted cationic microbubbles conjugated with CD105 antibody compared with untargeted cationic and neutral microbubbles. Theranostics 2015;5: 399–417.