| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Second Department of Internal Medicine, Kagawa University School of Medicine, 1750-1, Ikenobe, Miki-cho, Kita-gun, Kagawa 761-0793, Japan. Received August 1, 2002; revision requested September 23; final revision received May 27, 2003; accepted June 10. Supported by grant for Hypertensive Arteriosclerosis Research Award, Tokyo, Japan. Address correspondence to K.O. (e-mail: komori@kms.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: BR14 and human albumin MBs (2 x 107/mL) were incubated with activated human neutrophils (2 x 106/mL) to allow phagocytosis. Deflation rate of the phagocytosed MBs after pulsed insonation (one burst per second for 5 seconds) at 1.8 MHz with peak negative pressure of -540 kPa or -1,340 kPa, lactate dehydrogenase (LDH) leakage, and terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling stain–positive cell count after insonation were compared between the two agents.
RESULTS: At -540 kPa, phagocytosed MBs remained nearly unchanged for both agents after insonation. At -1,340 kPa, although human albumin MBs were disrupted on the first or second burst, BR14 MBs remained undisrupted. After -540-kPa insonation, a similar number of apoptotic cells appeared in neutrophils containing human albumin and BR14 MBs. At -540 kPa, LDH leakage was limited in human albumin MBs and BR14 MBs. At -1,340 kPa, LDH leakage was significantly increased in human albumin MBs and BR14 MBs (P < .01, both vs -540 kPa). Apoptotic cells were significantly decreased in human albumin MBs and BR14 MBs (P < .01, both vs -540 kPa). LDH leakage was lower and apoptotic cell count was greater in BR14 MB–containing neutrophils than in human albumin MB–containing neutrophils (both P < .01).
CONCLUSION: Compared with human albumin MBs, BR14 MBs were more stable after phagocytosis with insonation. This stability is associated with less disruption and greater induction of apoptosis in leukocytes after relatively high-pressure insonation in the range for diagnostic use.
© RSNA, 2004
Index terms: Leukocytes • Microbubbles • Ultrasound (US), contrast media • Ultrasound (US), experimental studies
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
BR14 (Bracco Research, Geneva, Switzerland) is a phospholipid-stabilized third-generation US contrast agent (5) that produces persistent contrast enhancement of tissue perfusion. This persistent contrast enhancement has been attributed to its transient retention in the tissue microcirculation (6). However, the feasibility of delivering this agent into leukocytes is unknown. Therefore, the purpose of our in vitro study was to confirm that BR14 MBs can be phagocytosed by activated leukocytes, to determine the persistence of this agent inside neutrophils, and to evaluate how such characteristics influence the fate of neutrophils containing MBs after insonation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Human albumin microspheres (Optison; Tyco Healthcare/Mallinckrodt, St Louis, Mo) that consisted of octafluoropropane gas–filled MBs with shells composed of albumin, which was supplied by the manufacturer as a suspension of 5 x 108 MBs per milliliter in saline, were used. The mean MB diameter was 4 µm ± 1 [SD]. BR14 consisted of perfluorobutane-containing MBs stabilized by a phospholipid monolayer. For MBs, the mean diameter was 2.5–3.0 µm and the mean concentration was (2–5) x 108/mL (5,6). Human albumin microspheres and BR14 were diluted with saline to obtain a final concentration of 2 x 107/mL (2). The neutrophils (2 x 106/mL), pholbor-12-myristate-13-acetate (10 nmol/L), normal human serum (50 µL), and MBs were mixed and suspended in the medium for a total volume of 10 mL.
Electron Microscopic Evaluation of MB Phagocytosis
At 15 and 30 minutes after incubation, a sample suspension was morphologically evaluated with transmission electron microscopy. Each milliliter of the sample was combined with an equivalent volume of 0.1 mol/L sodium cacodylate buffer (pH 7.5) containing 2% osmium tetroxide for 30 minutes. The sample was centrifuged, washed in phosphate-buffered saline, fixed in 2% glutaraldehyde-paraformaldehyde, dehydrated in a graded series of acetone baths, and embedded in epoxy resin. Thin (1-µm-thick) sections were stained with saturated uranyl acetate and lead citrate. Observations were made by one of the coauthors (I.K.) on three photomicrographs obtained with a transmission electron microscope (JEM-1200EX; Front-End-of-the-Line Cluster Technologies, Tokyo, Japan) at a maximal magnification of x6,000; these photomicrographs showed one to two neutrophils. When the inclusion of a vacuolar structure in the cytoplasm was observed, the phagocytosis of MBs by the neutrophils was confirmed. The visually determined sizes of the vacuoles at 15 and 30 minutes of incubation were qualitatively compared.
Real-time Observation of Response of Phagocytosed MBs to Ultrasound
For the real-time observation of temporal alterations of phagocytosed MBs and their responses to ultrasound, the samples were transferred into microwells on a microwell plate (Multidish; Nunc, Rochester, NY) after a 15-minute incubation period for light microscopic observation. The plate was set in a water bath on the microscope stage (IX-FLA; Olympus, Tokyo, Japan) such that one of the wells was observed at high-power magnification. Then, a transducer (S4; Philips Medical Systems, Andover, Mass) connected to a diagnostic US scanner (SONOS5500; Philips Medical Systems) was placed to apply US to the well from the lateral side of the bath. The distance from the transducer to the center of the well (4-mm diameter, 2-mm depth) was fixed at 4 cm by using an acoustic coupler (degassed agar) attached to the lateral wall of the water bath so that the well was placed in the focal zone of US. The microwells and the water bath were composed of polystyrene and were nearly acoustically transparent in the condition employed in the present study. Temporal alterations, as discussed later, of phagocytosed MBs were observed without insonation for a 30-minute period. For the assessment of the effects of insonation, the well was exposed to five US bursts by using the two-dimensional mode at 1 frame per second. A frame comprised 90 scan lines in the imaging angle of 85°. A scan line was a single four-wave-long compression-first pulse at 1.8 MHz with a peak negative acoustic pressure of -540 kPa (mechanical index, 0.4) or -1,340 kPa (mechanical index, 1.0). The well was covered with approximately five scan lines per frame. Recordings were made with a video camera (MotionCorder 1000; Kodak, Rochester, NY) interfaced with a videotape recorder (ST-120 S-VHS; Fujifilm, Tokyo, Japan).
Three to five neutrophils that contained a MB that appeared circular were chosen from a well in each sample, and the diameters of the phagocytosed MBs were measured off-line with video calipers by one of the coauthors (I.K.). First, the temporal change of the phagocytosed MB size without insonation was examined at 2.5-minute intervals. Then, the effect of ultrasound was also examined by measuring the diameter of the MBs at approximately 
second after each US burst in a well that was in the focal zone for each sample. The MB volume was calculated from the MB diameter, with the assumption of a spherical geometric shape. Then, the rate of reduction in MB size was determined for each of the three to five MBs in each condition by fitting a monoexponential function, y = e-ßx + C, to the relationship between the US burst number or time (x) and the normalized MB volume (y), where ß is a constant of decay and C is a constant (2). We repeated the assessment of alteration in MB size three times to obtain nine to 15 ß values for each condition for statistical analysis.
Assessment of Insonation Effects on Fate of Neutrophils Containing MBs
One milliliter of the suspension was pipetted into polypropylene microtubes (4 mm in diameter) by one of the coauthors (K.S.). These samples were sorted among four different conditions on the basis of the absence or presence of human albumin MBs or BR14 MBs and the acoustic pressure of ultrasound (peak negative pressure of -540 kPa or -1,340 kPa). For the exposure to US bursts, the microtube was placed into housing built in degassed agar, where five scans were obtained with the two-dimensional mode at 1.8 MHz and 1 burst per second with the same settings of the same US scanner as employed in the real-time microscopic observation. The distance from the transducer to the center of the tube was fixed at 4 cm. The transducer was positioned for visualization of the long-axis image of the tube, which confirmed that the entire volume of the leukocyte suspension in the tube was exposed to US. The elevation of the ultrasound beam was approximately 4 mm at the focal zone. US was applied at the end of a 15-minute incubation period at 37°C. We prepared two sets of leukocyte suspension for each condition: one for the terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling (TUNEL) stain and the other for assay of enzyme activity. Both TUNEL stain and enzyme assay were triplicated.
Apoptosis.—TUNEL staining (8) with diaminobenzidine colorization was performed 1 hour after insonation by using an in situ kit (Apoptosis Detection Kit; TaKaRa Bio, Otsu, Japan) for semiquantitative assessment of apoptosis of the neutrophils. The sediment was air dried and fixed in 4% buffered paraformaldehyde–phosphate-buffered saline (pH 7.4) for 30 minutes at room temperature. Endogenous peroxidase was inactivated with 0.3% H2O2 methanol for 15 minutes at room temperature. The plates were then rinsed with phosphate-buffered saline, and after processing with Permeabilization Buffer (contained in the kit), Labeling Safe Buffer containing terminal deoxynucleotidyl transferase and fluorescein isothiocyanate–deoxyuridine 5-triphosphate (contained in the kit) was added to the plate. The plate was incubated in a humid atmosphere at 37°C for 60 minutes. The reaction was terminated by washing the plate with phosphate-buffered saline. Each plate was covered with anti–fluorescein isothiocyanate horseradish peroxidase conjugate and incubated for 30 minutes at 37°C. After background staining with 3% methyl green, the sample was processed for light microscopy. One of the coauthors (A.O.) examined five high-power fields to count the number of TUNEL stain–positive cells per high-power field for each condition in each session, which was repeated three times to obtain 15 values for each condition.
Neutrophil cell membrane integrity.—One hour after insonation, lactate dehydrogenase (LDH) activity was measured by a coauthor (K.S.) to assess the extent of neutrophil cell membrane injury (9) caused by the interventions. The sample was divided into three portions; in each portion, a commercial kit (Sigma Chemical) was used to measure the LDH activity as the rate of decrease in absorbance at 344, 340, and 365 nm caused by the formation of nicotinamide adenine dinucleotide from pyruvate and nicotinamide adenine dinucleotide (reduced form). We repeated the LDH assessment three times to obtain nine values for each condition. The activity was presented as the percentage of the predetermined total activity obtained from the same number of neutrophils (2 x 106/mL) incubated with 1 mL of surfactant (1% Triton X-100; Sigma Chemical) for 30 minutes.
Statistical Analysis
All experiments were triplicated, and all the values of the multiple measurements for each parameter were collected for statistical analyses. The data were expressed as medians and 10th–90th percentiles. For comparison of the TUNEL stain–positive cell count and LDH activity among multiple conditions, the Kruskal-Wallis test with the Student–Newman-Keuls test, which allows all pairwise comparisons as the post hoc test, was used. For comparison of ß values (decay rate of MB size) between the two conditions, the Mann-Whitney U test was used. Differences with P < .05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
Fate of Leukocytes after Insonation with Phagocytosed MBs
In the absence of MBs, 60 minutes after insonation few TUNEL stain–positive cells were observed after insonation at low pressure (-540 kPa) (Fig 5, A) or high pressure (-1,340 kPa) (Fig 5, B). At low-pressure insonation, a substantial number of TUNEL stain–positive neutrophils were observed in human albumin MBs (Fig 5, C). However, at high acoustic pressure, the number of TUNEL stain–positive cells was reduced (Fig 5, D). A substantial number of TUNEL stain–positive cells were observed also for BR14 MBs at -540 kPa (Fig 5, E), and this number was reduced at -1,340 kPa (Fig 5, F) but to a lesser extent compared with human albumin MBs (Fig 5, D). Thus, as summarized in Figure 6, which shows a comparison of the number of TUNEL stain–positive cells, insonation with phagocytosed MBs resulted in neutrophil apoptosis in 1 hour regardless of the composition of MBs. However, at high acoustic pressure, the number of apoptotic cells was significantly reduced in the presence of human albumin MBs (median, 6 cells per high-power field; 10th–90th percentiles, 3–9 cells per high-power field) versus these values at low pressure (median, 12 cells per high-power field; 10th–90th percentiles, 9–18 cells per high-power field) (P < .01). In contrast, in the presence of BR14 MBs, a substantial degree of apoptosis was induced in neutrophils, even with high-pressure insonation (median, 8 cells per high-power field; 10th–90th percentiles, 6–14 cells per high-power field; P < .01 vs human albumin MBs at high acoustic pressure).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In the present study, BR14 has been demonstrated to be phagocytosed by activated neutrophils and to be durable inside the neutrophils, with a minor gradual deflation during pulsed insonation at the energies used. Although the albumin-shell agent human albumin microspheres has been shown in a previous study (1) and also in the present study to be phagocytosed by neutrophils, the high susceptibility of this agent to ultrasound caused substantial neutrophil disruption at a relatively high acoustic pressure in the range of diagnostic use. Such disruption of neutrophils resulted in a significant increase in the leakage of cytosolic enzyme, which suggests an excessive release of proinflammatory contents that potentially aggravates inflammation in in vivo settings.
In this regard, the induction of apoptosis in neutrophils may be a solution. Apoptotic neutrophils are processed by monocytes in vivo; this activity causes no marked release of proinflammatory contents and leads to a natural subsidence of inflammation (10). In the present study, although human albumin MBs caused marked apoptosis of neutrophils only at low-pressure insonation, BR14 MBs exhibited preserved apoptosis and less LDH leakage, even at the high-pressure insonation, compared with human albumin MBs. Thus, the phospholipid-stabilized MB agent BR14 is stable within the neutrophils and allows a wider range of acoustic pressure that selectively induces apoptosis and preserves the cell membrane integrity of neutrophils, which may be an advantage in US leukocyte imaging, in terms of safety.
Phagocytosis of BR14 MBs with Activated Leukocytes
It has been demonstrated that interactions between activated leukocytes and albumin-shell and phospholipid-stabilized MBs include adhesion and phagocytosis (1,2). Phagocytosis of human albumin MBs is mediated by the CD18 subunit of the ß2 integrin for Mac-1 (1), whereas lipid MBs require the C3 component of serum complement (1) to be phagocytosed. Neutrophils and other phagocytotic cells bind liposomes in a complement-dependent process (11,12). In the present study, BR14 MBs were phagocytosed by activated neutrophils to a similar extent as human albumin MBs in the presence of normal human serum that contains complement. Thus, the recognition by leukocytes of BR14 MBs with a phospholipid layer may require opsonization by complement, which can be easily provided in in vivo settings.
Stability of MBs in Cytoplasm of Leukocytes
BR14 MBs exhibited less deflation inside the activated leukocyte compared with human albumin MBs. A phospholipid monolayer with high static stability may provide a durable barrier against the biologic intracellular digestive process, and the high molecular weight and low diffusiveness of the gas content may prevent deflation (13). Light microscopic observations clearly demonstrated that intracellular BR14 MBs maintained the spherical structure at 30 minutes after the onset of the reaction, at which time human albumin MBs were substantially deflated.
As illustrated in previous studies (1,2), a single US pulse delivered at a high negative pressure greater than 1 MPa can destroy albumin-shell MBs regardless of whether they are phagocytosed or nonphagocytosed. At an intermediate pressure (600 kPa), intracellular MBs were increased to more than 200% of the initial diameter at the rarefactional phase, which stretched the cell membrane but did not result in its rupture (4). In the present study, either phagocytosed or free BR14 MBs were not destroyed by diagnostic ultrasound at a high pressure but exhibited a progressive deflation during each of the subsequent ultrasound bursts. The higher deflation rate of the intracellular MBs in the presence of insonation compared with the deflation rate of MBs in its absence suggests that volume oscillation during a pulse may facilitate the leakage of gas from the MBs. The viscoelasticity of the surrounding cellular milieu absorbs the acoustic energy and results in smaller fluctuations in size and less shell damage of intracellular MBs compared with those of free MBs (4). Since the viscoelastic properties of the leukocytes that phagocytosed human albumin MBs and BR14 MBs should be similar, the difference in the results with pulsed insonation can be ascribed to the difference in MB composition. Thus, the shell-and-gas component of BR14 MBs that was resistant to both biologic and physiologic insults was also resistant to acoustic interventions. Thereby, ultrasound energy was effectively transmitted intracellularly to leukocytes in the presence of BR14 MBs. The delivered energy altered the intracellular conditions that ultimately triggered the process of apoptosis.
Mechanism of Induction of Apoptosis
Several hypothetical mechanisms are possible among various scenarios of apoptosis induction (14). Therapeutic ultrasound of low frequency and high output that causes cavitation induced apoptosis in human myeloid leukemic cells in vitro (15). However, the absence of free radical production excluded the free radical–dependent induction of apoptosis (15). Mechanical stretching of the cytoplasmic membrane caused by intracellular MB oscillation may activate the Fas antigen, a transmembrane protein, to trigger the process (14,16). The mechanical impact on the mitochondria may directly facilitate the cytochrome c leakage (17) in the cytoplasm, and subsequent activation of the caspase cascade (14,17) also may be possible.
Implications of MB Stability in US Leukocyte Imaging
The stability of MBs in the leukocytes with insonation is important in leukocyte imaging. The persistence of BR14 MBs in activated neutrophils should lead to a prolonged retention in the sites of inflammation compared with MBs that are easily digested after phagocytosis. This persistence of BR14 MBs may provide selective detection of signals from the leukocytes that retain BR14 MBs in the inflamed site after the freely circulating MBs are cleared from the blood pool (2,3).
Fragile MBs are sensitive to US energy and may produce strong ultrasound signals for better imaging caused by bubble destruction. However, BR14 MBs were disrupted much less frequently than were human albumin MBs when exposed to the same acoustic energy. Therefore, it is conceivable that the oscillation amplitude of BR14 MBs in the leukocytes was lower and hence returned weaker signals than did those from human albumin MBs. However, it was demonstrated that phagocytosed MBs produced a frequency shift caused by a viscous damping from the rheologic characteristics of leukocytes—only at a rarefaction-first pulse but not at a compression-first pulse—which resulted in a greater frequency shift compared with that produced by free MBs (4). Therefore, a pulse-inversion method that allows detection of the difference in responses to normal and phase-inverted pulses with low acoustic pressure (18,19) may be a promising modality for discrimination of phagocytosed MBs from free MBs, as well as from tissue (4). Nevertheless, the longer persistence of the phagocytosed MBs compared with that of free MBs may be the most likely factor for the differentiation.
Limitations of This Study
First, since it has been reported previously that both human albumin MBs and other phospholipid-stabilized perfluorocarbon MBs similar to BR14 MBs can produce sufficient signals for US with a commercial US scanner (2,3), we did not compare the two agents in terms of imaging. Human albumin MBs, which in this study were found to be more sensitive to ultrasound energy after phagocytosis, may better provide imaging. In addition, human albumin MBs also selectively induced apoptosis when low-pressure insonation was employed. Therefore, if an imaging method with a low acoustic pressure for detection of MBs is employed, both agents may provide safe imaging of inflammation, although the long persistence of BR14 MBs in leukocytes may be an additional advantage.
Second, we did not assess the effects of continuous insonation but employed intermittent insonation by using a diagnostic ultrasound system because investigators in studies of contrast-enhanced US have employed intermittent imaging (2,3,20) that allows more time for inertial effects, bubble oscillations, and replenishment of insonated volume. The behavior of phagocytosed MBs with continuous insonation may need to be studied, especially at the low acoustic pressure that has been introduced in contrast-enhanced US (18).
Third, although previous reports indicated that the apoptotic neutrophils are cleared by monocytes or macrophages—an activity that leads to the subsidence of inflammation (10)—the relevance of selective induction of apoptosis in neutrophils in vivo remains to be determined.
Finally, although the induction of apoptosis in leukocytes with insonation of phagocytosed MBs was shown in the present study, the mechanism for this phenomenon is unclear.
Practical application: In this study, we demonstrated that phospholipid-stabilized MBs of BR14 deflate less in the leukocytes after phagocytosis and are less susceptible to ultrasound than are human albumin MBs in the pressure range for diagnostic use. This stability is associated with less disruption of leukocytes and greater induction of apoptosis after insonation at relatively high pressure. Although the mechanisms remain to be clarified, switching the fate of neutrophils from cell disruption, which causes the release of excessive bioactive chemicals, to apoptosis, which is eventually cleared by a physiologic process that leads to subsidence of inflammation, would be an important advantage in leukocyte imaging. This agent characteristic may prove to be advantageous in leukocyte imaging, in terms of safety, especially for its application to critical pathologic conditions such as myocardial reperfusion injury (3). In addition, our results also may direct future studies toward optimization of factors related to ultrasound with MBs as a therapeutic means of controlling inflammation, besides its diagnostic use.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: LDH = lactate dehydrogenase, MB = microbubble, TUNEL = terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling
Author contributions: Guarantors of integrity of entire study, K.O., I.K., M.K.; study concepts, H.T., K.O., I.K., K.S., M.K., K.M.; study design, H.T., I.K., K.O., K.S., A.O.; literature research, H.T., K.O., I.K., K.M.; experimental studies, H.T., I.K., K.S.; data acquisition, H.T., I.K., K.S., J.Y., Y.T.; data analysis/interpretation, K.O., I.K., J.Y., Y.T., A.O.; statistical analysis, I.K., J.Y.; manuscript preparation, H.T., K.O., I.K., K.M., Y.T., J.Y.; manuscript definition of intellectual content, K.S., A.O., M.K.; manuscript editing, K.O., K.M.; manuscript revision/review, Y.T., M.K., K.M.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S Enomoto, M Yoshiyama, T Omura, R Matsumoto, T Kusuyama, D Nishiya, Y Izumi, K Akioka, H Iwao, K Takeuchi, et al. Microbubble destruction with ultrasound augments neovascularisation by bone marrow cell transplantation in rat hind limb ischaemia Heart, April 1, 2006; 92(4): 515 - 520. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yoshida, K. Ohmori, H. Takeuchi, K. Shinomiya, T. Namba, I. Kondo, H. Kiyomoto, and M. Kohno Treatment of Ischemic Limbs Based on Local Recruitment of Vascular Endothelial Growth Factor-Producing Inflammatory Cells With Ultrasonic Microbubble Destruction J. Am. Coll. Cardiol., September 6, 2005; 46(5): 899 - 905. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
= human albumin medians (median ß, 0.443; 10th-90th percentiles, 0.345-0.621, with P < .05 versus BR14 MBs [Mann-Whitney U test]), 
= BR14 medians (median ß, 0.020; 10th-90th percentiles, 0.018-0.022).
