Advertisement

Ultrasonic low-energy treatment

A novel approach to induce apoptosis in human leukemic cells

      Abstract

      Objective

      We evaluated the cytotoxic effect of ultrasonic irradiation at low energy on the viability of normal and leukemic cells and the potential mechanisms of action inducing this cytotoxicity.

      Materials and Methods

      Human leukemia cell lines (K562, HL-60, KG1a, and Nalm-6), primary leukemic cells, and normal mononuclear cells are treated by ultrasound at a frequency of 1.8 MHz during various exposure times (acoustical power of 7 mW/mL) and immediately tested for cell viability by the trypan blue exclusion assay. Apoptosis is evaluated by cell morphology, phosphatidylserine exposure, and DNA fragmentation. The mitochondrial potential, glutathione content, caspase-3 activation, PARP cleavage, and bcl-2/bax ratio are tested by flow cytometry. Cloning efficiency is evaluated by assays in methylcellulose.

      Results

      The technique we describe here, using minute amounts of energy and in the absence of any chemical synergy, specifically triggers apoptosis in leukemic cells while necrosis is significantly reduced. Ultrasonic treatment of 20 seconds' duration induces a series of successive phases showing the characteristic features of apoptosis: mitochondrial transmembrane potential disturbances, loss of phosphatidylserine asymmetry, morphological changes, and, finally, DNA fragmentation. In contrast to K562 cells, for which a significant reduction of cloning efficiency is observed, the growth of hematopoietic progenitors is totally unaffected. Ultrasound treatment is also associated with depletion of cellular glutathione content, suggesting a link with the oxidative stress. Moreover, the fact that active oxygen scavengers reduce ultrasonic-induced apoptosis suggests a sonochemical mechanism.

      Conclusion

      The cell damage observed after exposure of leukemic cells to ultrasound is associated with the apoptotic process and may be a promising tool for a smooth, specific, and effective ex vivo purging of leukemic cells.
      Cell-culture suspensions can be damaged by exposure to the therapeutic level of ultrasound [
      • Kaufman G.E
      • Miller M.W
      • Griffiths T.D
      • Ciaravino V
      • Carstensen E.L
      Lysis and viability of cultured mammalian cells exposed to 1 MHz ultrasound.
      ]. Ultrasound may cause irreversible cell damage and induce important cell membrane modifications [
      • Ellwart J.W
      • Brettel H
      • Kober L.O
      Cell membrane damage by ultrasound at different cell concentrations.
      ,
      • Fahnestock M
      • Rimer V.G
      • Yamawaki R.M
      • Ross P
      • Edmonds P.D
      Effects of ultrasound exposure in vitro on neuroblastoma cell membranes.
      ]. Several reports have suggested that cavitation resulting from the collapse of gas bubbles generated by acoustic pressure fields may be the cause for cell damage following ultrasonic irradiation [
      • Miller D.L
      • Williams A.R
      Bubble cycling as the explanation of the promotion of ultrasound in a rotating tube exposure system.
      ,
      • Thacker J
      An approach to the mechanism of killing of cells in suspension by ultrasound.
      ]. It has also been suggested that cavitation induces single-strand breaks in DNA by the action of residual hydrogen peroxide [
      • Miller D.L
      • Thomas R.M
      • Frazier M.E
      Ultrasonic cavitation indirectly induces single strand breaks in DNA of viable cells in vitro by the action of residual hydrogen peroxide.
      ].
      The use of ultrasound in cancer therapy has become an important issue [
      • Barnett S.B
      • ter Haar G.R
      • Ziskin M.C
      • Nyborg W.L
      • Maeda K
      • Bang J
      Current status of research on biophysical effects of ultrasound.
      ,

      Umemura S, Kawabata K, Sasaki K, Yumita N, Umemura K, Nishigaki R (1996) Recent advances in sonodynamic approach to cancer therapy. Ultrason Sonochem 3-S:187

      ,
      • Hill C.R
      • ter Haar G.R
      High intensity focused ultrasound—potential for cancer treatment.
      ]. Ultrasound has been used in conjunction with hyperthermia, and photo-, radio-, and chemotherapy [
      • ter Haar G.R
      • Loverock P
      Synergism between hyperthermia, ultrasound and γ irradiation.
      ]. Malignant cells are known to be more susceptible to these combined methods than their normal counterparts [
      • Lejbkowicz F
      • Salzberg S
      Distinct sensitivity of normal and malignant cells to ultrasound in vitro.
      ]. The effect of a direct irradiation (e.g., ultrasound, laser, light) on certain molecules (porphyrins, psoralenes, and anthracyclines) is to generate highly active oxygen species such as singlet oxygen, superoxyde radicals, or fatty acid radicals, which can play an important role in cancer treatment, acting selectively on malignant cells [
      • Kessel D
      • Jeffers R
      • Fowlkes J
      • Cain C
      Porphyrin-induced enhancement of ultrasound cytotoxicity.
      ,
      • Briviba K
      • Klotz L.O
      • Sies H
      Toxic and signaling effects of photochemically or chemically generated singlet oxygen in biological systems.
      ].
      According to the origin of the photons, the therapy is called PDT (photodynamic therapy) or, if by sonoluminescence, SDT (sonodynamic therapy). The effects generated by SDT and PDT are different in terms of cell viability; both SDT (specifically related to the ultrasonic cavitational activity) and PDT generate active oxygenated species and lead to a diminution of the intracellular thiol levels [
      • Kessel D
      • Jeffers R
      • Fowlkes J
      • Cain C
      Effects of sonodynamic and photodynamic treatment on cellular thiol levels.
      ]. In the case of PDT by ultraviolet-A (UVA), apoptosis of T helper cells is induced by the generation of singlet oxygen, but this effect depends essentially on the initial concentration in photosensitizers (PS) and on the local oxygen concentration [
      • Morita A
      • Werfel T
      • Stege H
      • et al.
      Evidence that singlet oxygen-induced human T helper cell apoptosis is the basic mechanism of ultraviolet-A radiation phototherapy.
      ]. For SDT, as a result of the high energies involved, the cell lysis is the major phenomenon, probably masking other effects on the surviving cells [
      • Miller D.L
      • Thomas R.M
      • Frazier M.E
      Single strand DNA breaks in CHO cells after exposure to ultrasound in vitro.
      ]. Classical SDT leading to apoptosis involves specific sensitizing molecules, and requires an electrical power of about 5 W/cm2 and irradiation time of several minutes. In the process described here, we used ultrasound at low energy to induce apoptosis specifically in leukemic cells, in the absence of any chemical agent synergy. Since singlet oxygen and hydroxyl radicals seem implicated in the induction of apoptosis, we have named this technique SLDT: Sonochemical Low-energy Dynamic Therapy.

      Materials and methods

      Cell preparation

      Human leukemia cell lines (K562, Nalm-6, KG1a, and HL-60) obtained from the American Type Culture Collection (ATCC, Rockville, MD, USA) were grown in RPMI-1640 (BioWhittaker, Walkersville, MD, USA) supplemented with 10% fetal calf serum (Gibco, Grand Island, NY, USA) and 1% L-glutamine (Gibco). Leukemic cells were harvested, resuspended in phosphate-buffered saline (PBS, pH = 7.2, Gibco), and immediately used for the experiment at a concentration of 106 total. Heparinized venous blood was obtained from healthy volunteers and leukemic patients after informed consent was obtained. Mononuclear cells were separated by Ficoll-Hypaque gradient density centrifugation (International Medical Products, Brussels, Belgium).

      Ultrasonic treatment

      The ultrasonic treatment and its specific results are part of patent PCT/BE97/00078 (pending) under the name of Eric Cordemans de Meulenaer et al.
      A total of 2.5 mL of cell suspension in 13 × 100-mm disposable plastic tubes (Greiner, Frickenhausen, Germany) was treated in our system with a frequency of 1.8 MHz during various times of exposure. Unless indicated otherwise, the power supplied by the generator to the ceramic disk is 0.22 W/cm2, which represents an acoustical power transmitted to the interior of the test cell of 0.007 W/mL. This calorimetric measurement was made at the University of Coventry, England, under Professor Tim Mason. The irradiation (ON/OFF) cycles are 5.5 ms/3 ms.

      Cell viability

      The cell viability was assessed by the trypan blue exclusion test immediately after ultrasonic treatment and after 18 hours culture in the incubator (37°C and 5% CO2).

      Morphological studies

      Cytocentrifuge preparations were made from the cell suspension and after air drying, cells were stained with May Grünwald Giemsa and analyzed by light microscopy.

      Annexin V binding assay

      Flow cytometric analysis of annexin-V–fluorescein isothiocyanate (FITC)- and propidium iodide (PI)-stained cells was performed using the kit purchased from Biosource International (Camarillo, CA, USA) as recommended by the manufacturer. Data are presented as dot plots showing the change in mean fluorescence intensity of annexin-V–FITC/propidium iodide.

      DNA fragmentation

      The level of DNA fragmentation of apoptotic cells was determined using the Apotarget Quick DNA Ladder Detection Kit (Biosource). Cell pellets (106 cells) were resuspended in 20 μL of lysis buffer and DNA was extracted according to manufacturer's instructions. DNA was analyzed after separation by gel electrophoresis (1% agarose). As positive control, cells were irradiated with UV light by placing a plate directly under a UV transilluminator for 10 minutes (intensity of 5 mW/cm2). Cells were then incubated at 37°C for 5 and 18 hours before apoptosis was assessed.
      Quantification of cells with degraded DNA was also performed using a method described by Nicoletti et al. [
      • Nicoletti I
      • Migliorati G
      • Pagliacci M.C
      • Grignani F
      • Riccardi C
      A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry.
      ]. After permeabilization, cells were incubated with solution containing PI and RNAse (Coulter DNA-prep Reagent). The tubes were placed at 4°C in the dark overnight before analysis by flow cytometry to identify the sub-G0 peak corresponding to apoptosis.

      Mitochondrial damage

      Mitochondrial potential was estimated by incorporation of the cationic fluorochrome DiOC6 immediately after cell treatment according to a published protocol [
      • Macho A
      • Castedo M
      • Marchetti P
      • et al.
      Mitochondrial dysfunctions in circulating T lymphocytes from human immunodefiency virus-1 carriers.
      ]. Briefly, K562 cells (106/mL) were incubated with 2.5 nmol/L 3,3′-dihexyloxacarbocyanine (DiOC6; Molecular Probes, Eugene, OR) for 15 minutes at 37°C, followed by flow cytometric analysis.

      Glutathione determination

      Cell Tracker green CMFDA (5-chloromethyl fluorescein diacetate; Molecular Probes) was used for determining levels of intracellular glutathione as previously described [
      • Hedley D.W
      • Chow S
      Evaluation of methods for measuring cellular glutathione content using flow cytometry.
      ]. Glutathione level was evaluated in viable cells (PI cells).

      Analysis of caspase-3 activity

      Caspase-3 was detected by flow cytometric analysis using the phycoerythrin (PE)-conjugated polyclonal rabbit antibody anti-active caspase-3 monoclonal antibody (BD-Pharmingen, San Diego, CA, USA). Cells were fixed and permeabilized using Fix and Perm kit (Caltag, Burlingame, CA) for 15 minutes at room temperature. Cells were then stained with anti-caspase-3 Ab and incubated for 15 minutes. Cells were washed and analyzed by flow cytometry. The enzymatic activity of caspase-3 was determined using the Apotarget caspase-3/cpp32/colorimetric protease assay kit (Biosource), as suggested by the manufacturer. Caspase-3 activation was also indirectly evaluated by PARP cleavage using a rabbit anti-PARP cleavage site AB, FITC conjugate (Biosource).

      bcl-2 and bax expression

      After permeabilization, cells were incubated with isotype-matched negative control, FITC-labeled mouse anti-human bcl-2 (Dako, Glostrup, Denmark), and polyclonal rabbit antibodies to bax. Subsequently, a FITC-labeled secondary antibody (Dako) was added to bax. To quantify bcl-2 and bax expression, the cytometer was calibrated using a mixture of beads labeled with known amounts of fluorochrome (Dako). The values of mean fluorescent intensity (MFI) were then converted to molecules of equivalent soluble fluorochrome (MESF) using a calibration curve.

      Clonogenic assay for K562 cell line

      Colony-forming unit-leukemic (CFU-L) cells were assayed as previously described [
      • Lagneaux L
      • Marie J.P
      • Delforge A
      • et al.
      Comparison of in vitro inhibition of etoposide (VP16) on leukemic and normal myeloid, erythroid clonogenic cells.
      ].
      Briefly, the culture medium consisted of IMDM supplemented with 20% FCS and methylcellulose at a final concentration of 4%. Cultures were incubated at 37°C in 5% CO2 air, and colonies (> 20 cells) were scored after 5 days. The clonogenic efficiency of K562 cell line was 16%.

      Assay for hematopoietic colony-forming cells

      Hematopoietic colony-forming cells (CFU-GEMM, CFU-GM, and BFU-E) were assayed using fetal bovine serum (FBS)-free methylcellulose medium supplemented by granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-3 (IL-3), IL-6, GCSF, and 3 U/mL EPO (erythropoietin; Methocult H4536, StemCell Technologies, Vancouver, BC, Canada). Cultures were plated in 0.25-mL volumes in 4-well tissue culture plates (Nunclon, Life Technologies, Merelbeke, Belgium). Mononuclear cells were plated at 104 cells/well. After 14 days at 37°C in 5% CO2 in air and 100% humidity, the colonies were counted.

      Statistical analysis

      The Wilcoxon Signed Ranks test was used to analyze the statistical significance of experimental results.

      Results

      Loss of phosphatidylserine asymmetry during ultrasonic treatment

      During apoptosis, phosphatidylserine residues flip from the inside to the outside of the plasma membrane and this change can be detected using annexin-FITC, which binds to the PS residues [
      • Vermes I
      • Haanen C
      • Steffens-Nakken H
      • Reutelingsperger C
      A novel assay for apoptosis flow cytometric detection of phophatidylserine expression on early apoptotic cells using fluorescein labeled annexin V.
      ]. Figure 1A indicates the changes of phosphatidylserine distribution according to time. Ultrasonic treatment provokes plasma membrane injury in a low percentage of cells, demonstrating that, in our conditions, the necrotic action of ultrasound is very weak. Interestingly, 2 hours posttreatment, an increase in apoptotic cells is observed and 5 hours after the treatment, 35% of cells are annexin-V-positive, demonstrating the ultrasonic induction of apoptosis in K562 cells.
      Figure thumbnail gr1
      Figure 1(A) Flow cytometric analysis of apoptosis in K562 cells. The cytograms represent the evolution of green (annexin-V) and red (PI) fluorescences for various times post–ultrasonic (US) treatment. In each panel, the upper right quadrants represent the percentage of necrotic cells or late-apoptotic cells (positive for annexin-V binding and for PI uptake). The lower right quadrant contains the apoptotic cells (annexin-V+ and PI, demonstrating cytoplasmic membrane integrity). (B) The DiOC6 staining of K562 cells treated or not by ultrasound. The histograms represent cell number (counts) vs green fluorescence intensity (FL-1): K562 cells were incubated with DiOC6 30 minutes (A), 2 hours (B), and 5 hours (C) posttreatment followed by immediate cytofluorometric analysis. Results from one representative experiment are shown. After US treatment, a population of cells appeared displaying a reduced DiOC6 incorporation.

      Ultrasonic treatment affects the mitochondrial membrane potential (Δϒm)

      The early disruption of mitochondrial transmembrane potential (Δϒm), preceding advanced DNA fragmentation, has been observed in several models of cell apoptosis [
      • Castedo M
      • Macho A
      • Zamzami N.N
      • et al.
      Mitochondrial perturbations define lymphocytes undergoing apoptotic depletion in vivo.
      ]. Ultrasonic treatment is accompanied by an increase of cell population displaying a low Δϒm (Fig. 1B). Already, 30 minutes posttreatment, a population of cells displaying a reduced DiOC6 incorporation is evidenced. The drop of mitochondrial potential is very clear 5 hours after the ultrasonic treatment with more than 50% of Δϒmlow cells.

      DNA fragmentation after ultrasonic treatment

      DNA fragmentation is assessed by agarose gel electrophoresis and by measuring signals in the hypodiploid region after PI labeling. Classic nucleosomal DNA ladder patterns are observed in DNA samples from cells treated by ultraviolet (positive control) and by ultrasound. Internucleosomal DNA cleavage is barely noticeable at 5 hours post–ultrasonic treatment but becomes evident at 18 hours posttreatment (Fig. 2). Five hours posttreatment, an increase in the number of nuclei with fragmented DNA is observed after PI staining (respectively 2% and 15% for untreated and treated cells) (data not shown).
      Figure thumbnail gr2
      Figure 2Analysis of ultrasound-induced DNA fragmentation in K562 cells. DNA was extracted from K562 cells treated or not by ultrasound (US) at 5 hours and 18 hours posttreatment. The samples were analyzed by electrophoresis on a 1% agarose gel and stained with ethidium bromide. Lane M contains a 50-bp DNA ladder used as a marker. As positive controls, K562 were treated by ultraviolet radiation.

      Effect of successive ultrasonic treatments

      Figure 3A shows the results obtained by a flow cytometry follow-up of K562 cells cultured during 0.5, 2, and 5 hours post–ultrasonic treatment. After one treatment, the level of apoptotic cells observed is thrice that of the control (respectively 26% and 8% after 5 hours of culture for treated and untreated cells). A necrotic effect of 5 to 10% is observed, which is well below that found when using drugs or photodynamic treatment (PDT) treatment. With successive irradiations, under the same conditions (7 mW/mL, 20 seconds) and at different intervals, apoptosis of K562 cells increases to 37 ± 3% (p < 0.02) and 49 ± 5% (p < 0.02) after respectively 1 and 3 successive treatments (Fig. 3B). Morphological variations (e.g., cell shrinkage, membrane blebbing, chromatin condensation) observed after successive treatments are shown in Figure 3C.
      Figure thumbnail gr3
      Figure 3Effect of successive ultrasonic (US) treatments on the viability of K562 cells. (A) Changes of phosphatidylserine distribution according to time and successive US treatments. Results from one representative experiment are expressed as percentage of cells stained with annexin-V-FITC. (B) Percentage of apoptotic K562 cells 5 hours after 1 or 3 ultrasonic treatments. The rate of apoptosis is determined by flow cytometry after annexin-V staining and the results are expressed as mean ± SEM of 7 independent experiments. (C) Morphological features of May-Grünwald Giemsa-stained K562 cells. (a) untreated cells; (b) 1 treatment; (c) 2 treatments; (d) 3 treatments. Arrows indicate cells with morphological changes (pyknosis, chromatin condensation and fragmentation). Magnification ×1000.

      Caspase-3 activity and PARP cleavage during ultrasound-induced apoptosis

      Caspase-3 has been shown to play an important role in chemotherapy-induced apoptosis [
      • Ibrado A.M
      • Huang Y
      • Fang G
      • Liu L
      • Bhalla K
      Overexpression of bcl-2 or bcl-xL inhibits Ara-C-induced CPP32/Yama protease activity and apoptosis of human acute myelogenous leukemia HL-60 cells.
      ]. To directly address the involvement of caspase-3 in ultrasound-induced apoptosis, caspase activity was determined using flow cytometry and colorimetric assay. As shown in Figure 4 (a and b), ultrasonic treatment leads to activation of caspase-3. Moreover, this protease activity was maximal at 1 hour posttreatment. Activation of caspases leads to cell demise via cleavage of cellular substrates such as actin, gelsolin, or PARP [
      • Lazebnik Y.A
      • Kaufmann S.H
      • Desnoyers S
      • Poirier G.G
      • Earnshaw W.C
      Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE.
      ]. As shown in Figure 4c, cleavage of PARP was apparent 2 hours after treatment, with 40% of cells stained by the rabbit anti-PARP FITC vs 5% for untreated cells.
      Figure thumbnail gr4
      Figure 4(a) Flow cytometric analysis of caspase-3 activity. K562 cells were treated or not by ultrasound and after 1 hour of culture, cells were fixed, permeabilized, and stained with anti-active caspase-3. A minimum of 10,000 events was analyzed and the dotted line represents the isotypic control. (b) Caspase-3 activity was also determined for times indicated by the colorimetric assay described in Materials and Methods. Results of a representative experiment are shown. (c) PARP cleavage evidenced by flow cytometry in K562 cells treated or not by ultrasound. Two hours posttreatment, cells were washed, permeabilized, and stained by anti-PARP cleavage site AB, FITC conjugate. (— untreated cells US-treated cells.)

      Ultrasound bcl-2/bax ratio modulation

      Different proteins of the bcl-2 family have been implicated in triggering or preventing apoptosis. Therefore, we have evaluated whether bcl-2 and bax proteins, the two major members of the bcl-2 family, are involved in the induction of apoptosis by ultrasound. As shown in Figure 5, untreated cells express high levels of bcl-2 protein (47 ± 4 × 103 MESF) and this expression appears as a unimodal peak of fluorescence. One hour post–ultrasonic treatment, the expression of bcl-2 protein is already downregulated (respectively 40 ± 0.9 and 32 ± 0.9 × 103 MESF in K562 cells treated by 1 or 3 ultrasonic treatments). Two hours posttreatment, bcl-2 expression appears clearly bimodal, the cells displaying either a bcl-2high (comparable to untreated cells) or bcl-2low phenotype (11 ± 2 × 103 MESF). In contrast, bax protein level was higher after ultrasonic treatment as compared with the bax level in untreated cells (respectively 85 ± 0.5 and 48 ± 5 × 103 MESF for treated and untreated K562). The ratio of bcl-2/bax is thus significantly reduced during ultrasonic treatment (0.98 for control cells vs 0.38 for ultrasound-treated cells).
      Figure thumbnail gr5
      Figure 5The effect of ultrasonic treatment on bcl-2 and bax protein expression in K562 cells. Histograms showing mean fluorescent intensity for bcl-2 and bax expression in untreated cells and in ultrasound (US)-treated cells (2 hours post–3-US treatment). Black line, expression in untreated K562 cells; gray line, expression in US-treated cells; light gray line, isotype control. MFI = mean fluorescence intensity.

      Depletion of cellular glutathione content by ultrasonic treatment

      As shown in Figure 6, directly post–ultrasonic treatment, a subpopulation appears with lower GSH level than that observed in untreated cells (>50% of cells displaying a low level of GSH). Successive treatments indicate a larger GSH depletion after 5 hours. The results, expressed as a percentage of cells displaying a GSH level comparable to untreated cells, demonstrate clearly that ultrasonic treatment is associated with GSH depletion.
      Figure thumbnail gr6
      Figure 6Effect of ultrasonic treatment on cellular glutathione levels. Intracellular GSH content was evaluated by the method described in Materials and Methods. Dead cells that had lost the capacity to exclude propidium iodide were gated out from glutathione analysis. Data are expressed in percentage of cells displaying glutathione level comparable to untreated cells. Values are mean ± SEM of the data from three independent experiments.

      Susceptibility of normal and malignant cells to ultrasonic treatment

      We evaluated the effect of ultrasonic treatment on such other cell lines as KG1a (immature minimally differentiated acute myeloid leukemia blasts), HL-60 (promyelocytic leukemia), and Nalm-6 (ALL cell line). The results presented in Figure 7 demonstrate that the sensitivity to ultrasound seems to depend on cell type, but successive treatments lead to a significant increase in the number of apoptotic cells for all cell lines evaluated. Mononuclear cells from 5 patients (1 refractory anemia with excess of blast cells [RAEB], 1 secondary acute myelogenous leukemia [AML], and 3 cases of AML French-American-British [FAB] classification M3, M4, and M4Eo) are also treated by ultrasound, and blast cells are discriminated from contaminating normal cells on the basis of their CD45 expression as previously described [
      • Lacombe F
      • Durrieu F
      • Briais A
      • et al.
      Flow cytometry CD45 gating for immunophenotyping of acute myeloid leukemia.
      ]. These cells had also been labeled for their phosphatidylserine exposure by FITC-annexin. It is possible with this method to compare the respective apoptotic behavior of leukemic blast cells and normal cells treated by ultrasound. The results presented in Figure 7 demonstrate that primary leukemic cells are sensitive to ultrasonic treatment with more than 37 ± 18% of apoptotic cells observed 5 hours post–3 treatments. On the contrary, ultrasonic treatment has no effect on normal mononuclear cells isolated from normal subjects and leukemic patients (after CD45 gating).
      Figure thumbnail gr7
      Figure 7Effect of ultrasonic treatment on the apoptosis of normal mononuclear cells (MNC) and leukemic cells (Nalm-6, KG1a, HL-60, and primary leukemic cells obtained from 5 patients). Apoptosis was evaluated by annexin/PI assay 5 hours posttreatment directly in the case of cell lines but after a CD45 gating strategy from primary blast cells. Results are mean ± SEM of 5 independent expseriments.

      Effect of ultrasonic irradiation on cloning efficiency

      The ultimate proof for an effect on cell viability is the inability of a cell to multiply and form a colony. As shown in Figure 8, a significant reduction in cloning efficiency of K562 cells is observed after 1 and 3 treatments (respectively 25% and 42% of inhibition), confirming the sensitivity of leukemic cells to ultrasound. In the case of normal hematopoietic progenitors (CFU-GM, BFU-E, and CFU-GEMM), the cloning efficiency is totally unaffected by ultrasound even after 3 treatments (data not shown).
      Figure thumbnail gr8
      Figure 8Effect of irradiation on cloning efficiency of K562 cells. Cells were treated or not by ultrasound, washed, and immediately seeded in methylcellulose medium at a concentration of 1000 cells per plate. After 5 days of culture, the colonies were counted. Results are expressed as mean ± SEM of 3 independent experiments.

      Effect of active oxygen scavengers on the induction of apoptosis by ultrasound

      K562 cells are incubated with L-histidine (10 mM) and/or mannitol (100 mM) and treated or not by ultrasound. Cell apoptosis is detected by annexin-V/PI assay. Our results demonstrate that the ultrasonically induced cell damage is significantly reduced in the presence of histidine and mannitol (respectively 43% and 47% of inhibition of apoptosis induced by 3 successive treatments) (Table 1). The association of mannitol and histidine leads to more than 60% of inhibition. The effectiveness of these agents on reducing the cell apoptosis induced by ultrasonic treatment may imply that ultrasonically generated singlet oxygen and hydroxyl radicals are important mediators to induce apoptosis. These observations thus suggest a sonochemical mechanism.
      Table 1Effects of active oxygen scavengers on the ultrasonically induced cell apoptosis
      No US treatment1 US treatment3 US treatments
      No agent18 ± 642 ± 863 ± 5
      Histidine15 ± 524 ± 836 ± 11
      Mannitol11 ± 421 ± 530 ± 3
      Histidine + Mannitol11 ± 216 ± 325 ± 5
      Results are expressed as percentage of cells displaying phosphatidylserine externalization 5 hours posttreatment (mean ± SEM from 4 independent experiments).

      Discussion

      Apoptosis ensures the homeostasis of tissues during development, host defense, and aging and occurs in response to a large variety of signals including γ-irradiation and ultraviolet exposure [
      • Kulms D
      • Schwarz T
      Molecular mechanisms of UV-induced apoptosis.
      ,
      • Belyaev I.Y
      • Czene S
      • Harms-Ringdahl M
      Changes in chromatin conformation during radiation-induced apoptosis in human lymphocytes.
      ,
      • Carloni M
      • Meschini R
      • Ovidi L
      • Palitti F
      PHA-induced cell proliferation rescues human peripheral blood lymphocytes from X-ray-induced apoptosis.
      ]. Apoptotic cell death is characterized by early changes in the nuclear membrane and by chromatin condensation, followed by DNA fragmentation [
      • Arends M.J
      • Morris R.G
      • Wyllie A.H
      Apoptosis the role of the endonuclease.
      ]. Ultrasound at high power also induces structural and functional changes in sonicated cells [
      • Alter A
      • Roznszajn L.A
      • Miller H.I
      • Rosenschein U
      Ultrasound inhibits the adhesion and migration of smooth muscle cells in vitro.
      ]. Moreover, ultrasound beams have the potential to treat malignant tumors when combined with sensitizers such as porphyrins [
      • Yumita N
      • Nishigaki R
      • Umemura K
      • et al.
      Sonochemical activation of hematoporphyrin An ESR study.
      ]. Recently, in vitro studies demonstrated therapeutic ultrasound-induced apoptosis in cultured cells via the process of cavitation [
      • Ashush H
      • Leon A
      • Rozenszajn
      • et al.
      Apoptosis induction of human myeloid leukemic cells by ultrasound exposure.
      ]. This therapeutic ultrasound (750 kHz), characterized by a “high-intensity” delivery to generate cavitation leading to apoptotic cell death, was observed in the surviving cells besides a large amount of necrosis (more than 40%). In contrast to the mechanism reported for γ-irradiation, apoptosis induced by ultrasound seems independent of cell-cycle modifications [
      • Yu Y.Q
      • Giocanti N
      • Averbeck D
      • Megnin-Chanet F
      • Favaudon V
      Radiation-induced arrest of cells in G2 phase elicits hypersensitivity to DNA double-strand break inducers and an altered pattern of DNA cleavage upon re-irradiation.
      ].
      In this study, the induction of apoptosis in leukemic cells by a “low-energy” ultrasonic treatment is demonstrated. This treatment leads to a sequence of events recognized as hallmarks of apoptosis: a drop in mitochondrial potential, loss of phosphatidylserine asymmetry, morphological variations, DNA fragmentation, and, finally, loss of plasma membrane. This “low-energy” ultrasound-induced apoptosis involves activation of caspase-3 and is accompanied by the proteolytic degradation of the caspase substrate PARP and by the modulation of bcl-2/bax ratio, demonstrated using flow cytometric analysis. A comparable mechanism has been demonstrated after photodynamic therapy [
      • Kessel D
      • Luo Y
      Photodynamic therapy a mitochondrial inducer of apoptosis.
      ,
      • He J
      • Whitacre C.M
      • Xue L.Y
      • Berger N.A
      • Oleinick N.L
      Protease activation and cleavage of poly(ADP-ribose) polymerase an integral part of apoptosis in response to photodynamic treatment.
      ].
      Soon after ultrasonic treatment, an important decrease of intracellular glutathione level is observed, suggesting that oxidative stress may play a role in ultrasound-induced apoptosis. Loss of glutathione and oxidative damage have been suggested to constitute early signaling events in apoptotic cell death [
      • Macho A
      • Hirsch T
      • Marzo I
      • et al.
      Glutathione depletion is an early and calcium elevation is a late event of thymocyte apoptosis.
      ].
      A mechanism coherent with all our observations involves ultrasound-induced sonochemical luminescence triggering photosensitized singlet oxygen production to initiate apoptosis as previously described for direct photoirradiation [
      • Sharman W.M
      • Allen C.M
      • Van Lier J.E
      Role of activated oxygen species in photodynamic therapy.
      ]. In classic ultrasonic irradiation conditions, the direct destructive cavitation effects dominate the sonoluminescence, which is weak in the absence of an air/liquid interface injected into the medium. The fact that there are no ultrasound effects in the absence of bubbles suggests the major role played by 1O2 in our system.
      Data obtained in the presence of histidine, a quencher of 1O2, suggest the importance of singlet oxygen in the induction of apoptosis by ultrasound under the “low energy” conditions. However, the fact that mannitol, an inhibitor of hydroxyl radicals, protects against ultrasound-induced apoptosis also implies the intervention of these radicals in our system.
      Evidence against singlet oxygen formation during sonodynamic therapy has been presented by Miyoshi et al. [
      • Miyoshi N
      • Igarashi T
      • Riesz P
      Evidence against singlet oxygen formation by sonolysis of aqueous oxygen-saturated solutions of Hematoporphyrin and Rose Bengal. The mechanism of sonodynamic therapy.
      ], but these data are only consistent with a long and “high-energy” ultrasound exposure, leading to an accumulation of sensitizer-derived free radicals either by direct pyrolysis or due to reactions with Ho or OHo radicals formed by pyrolysis of the water solvent. In the present study, the supernatant of cells submitted to one or several ultrasonic treatments is unable to induce apoptosis in K562 cells, demonstrating that the effects generated do not originate from chemical species produced by the solvent. At the chosen frequencies, the “low-energy” ultrasound does not directly generate free radicals such as those originating in the sonolysis of the solvent or those from the addition of molecules such as dimsethylsulfoxide (DMSO) or N, N-dimethylformamide (DMF), which generate a sonodynamic action on leukemia cells [
      • Misik V
      • Riesz P
      Peroxyl radical formation in aqueous solutions of N,N-dimethylformamide and N-methylformamide and dimethylsulfoxide by ultrasound implications for sonosensitized cell killing.
      ,
      • Jeffers R
      • Feng R
      • Fowlkes J
      • Hunt J
      • Kessel D
      • Cain C
      Dimethylformamide as enhancer of cavitation-induced cell lysis in vitro.
      ].
      Sonoluminescence being in our case the only possible inducing physical phenomenon due to the presence of topical cellular PS, it is normal that these results can be compared to photolytical damages suffered after an exposure to classic ultrasound. The effects obtained with our technique are achieved without the necessity of classical PS. The physiological effects obtained with techniques such as phototherapy depend at the same time on the radiation dose, on the nature of the PS used, on their concentration, and on their localization [
      • Kessel D
      • Luo Y
      • Deng Y
      • Chang C.K
      The role of subcellular localization in initiation of apoptosis by photodynamic therapy.
      ]. The major targets of PDT are cell membranes but, significantly, the technique discussed here is the only one to act directly within the interior of the cells. Within the cells, it is the endogenous photoabsorbing molecules such as the porphyrins and flavoproteins that play the photosensitizing role [
      • Giese A
      Photosensitization of organisms with special reference to natural photosensitizers.
      ]. Indeed, an implication of endogenous porphyrins in photodynamic DNA damage has been proposed [
      • Duez P
      • Hanocq M
      • Dubois J
      Photodynamic DNA damage mediated by δ-aminolevulinic acid-induced porphyrins.
      ].
      In our technique, the net effects of the ultrasonic action suggest that endogenous PS may be implicated in the structure where their local concentration is high. Endogenous PS are localized mainly in the membrane structures such as lysosomes, mitochondria, nuclear membranes, and the microsomes of the endoplasmic reticulum, of which the relative surface represents nearly 50% of the cell membrane surface [
      • Moan J
      • Berg K
      • Kvam E
      • et al.
      Intracellular localization of photosensitizers.
      ].
      Under these soft conditions to which they are subjected, healthy cells are much less sensitive to ultrasound than leukemic cells. This difference in behavior between the healthy and leukemic cells cannot be related to a difference in the localization of the endogenous PS but is probably due to a modification of the fundamental cell mechanisms (e.g., p53 status, signaling pathways, resistance to oxidative stress).
      In conclusion, this “low-energy” ultrasonic treatment (SLDT) induces a similar sequence of events as that induced by PDT: a rapid formation of 1O2, having on the one hand an effect on mitochondrial membranes (drop of mitochondrial potential) and provoking on the other hand a lipidic oxidation of the membrane (decrease of cellular GSH level). Specific tests have confirmed the very rapid induction of apoptosis in the absence of necrosis. This ultrasonic treatment could be a promising tool for the ex vivo elimination of leukemic cells by means of apoptosis.

      References

        • Kaufman G.E
        • Miller M.W
        • Griffiths T.D
        • Ciaravino V
        • Carstensen E.L
        Lysis and viability of cultured mammalian cells exposed to 1 MHz ultrasound.
        Ultrasound Med Biol. 1977; 3: 21
        • Ellwart J.W
        • Brettel H
        • Kober L.O
        Cell membrane damage by ultrasound at different cell concentrations.
        Ultrasound Med Biol. 1988; 14: 43
        • Fahnestock M
        • Rimer V.G
        • Yamawaki R.M
        • Ross P
        • Edmonds P.D
        Effects of ultrasound exposure in vitro on neuroblastoma cell membranes.
        Ultrasound Med Biol. 1989; 15: 133
        • Miller D.L
        • Williams A.R
        Bubble cycling as the explanation of the promotion of ultrasound in a rotating tube exposure system.
        Ultrasound Med Biol. 1989; 15: 641
        • Thacker J
        An approach to the mechanism of killing of cells in suspension by ultrasound.
        Biochim Biophys Acta. 1973; 304: 240
        • Miller D.L
        • Thomas R.M
        • Frazier M.E
        Ultrasonic cavitation indirectly induces single strand breaks in DNA of viable cells in vitro by the action of residual hydrogen peroxide.
        Ultrasound Med Biol. 1991; 17: 607
        • Barnett S.B
        • ter Haar G.R
        • Ziskin M.C
        • Nyborg W.L
        • Maeda K
        • Bang J
        Current status of research on biophysical effects of ultrasound.
        Ultrasound Med Biol. 1994; 20: 205
      1. Umemura S, Kawabata K, Sasaki K, Yumita N, Umemura K, Nishigaki R (1996) Recent advances in sonodynamic approach to cancer therapy. Ultrason Sonochem 3-S:187

        • Hill C.R
        • ter Haar G.R
        High intensity focused ultrasound—potential for cancer treatment.
        Br J Radiol. 1995; 68: 1296
        • ter Haar G.R
        • Loverock P
        Synergism between hyperthermia, ultrasound and γ irradiation.
        Ultrasound Med Biol. 1991; 17: 607
        • Lejbkowicz F
        • Salzberg S
        Distinct sensitivity of normal and malignant cells to ultrasound in vitro.
        Environ Health Perspect. 1997; 105: 1575
        • Kessel D
        • Jeffers R
        • Fowlkes J
        • Cain C
        Porphyrin-induced enhancement of ultrasound cytotoxicity.
        Int J Radiat Biol. 1994; 66: 221
        • Briviba K
        • Klotz L.O
        • Sies H
        Toxic and signaling effects of photochemically or chemically generated singlet oxygen in biological systems.
        Biol Chem. 1997; 378: 1259
        • Kessel D
        • Jeffers R
        • Fowlkes J
        • Cain C
        Effects of sonodynamic and photodynamic treatment on cellular thiol levels.
        J Photochem Photobiol B. 1996; 32: 103
        • Morita A
        • Werfel T
        • Stege H
        • et al.
        Evidence that singlet oxygen-induced human T helper cell apoptosis is the basic mechanism of ultraviolet-A radiation phototherapy.
        J Exp Med. 1997; 186: 1763
        • Miller D.L
        • Thomas R.M
        • Frazier M.E
        Single strand DNA breaks in CHO cells after exposure to ultrasound in vitro.
        J Ultrasound Med. 1990; 9S: 21
        • Nicoletti I
        • Migliorati G
        • Pagliacci M.C
        • Grignani F
        • Riccardi C
        A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry.
        J Immunol Methods. 1991; 139: 271
        • Macho A
        • Castedo M
        • Marchetti P
        • et al.
        Mitochondrial dysfunctions in circulating T lymphocytes from human immunodefiency virus-1 carriers.
        Blood. 1995; 86: 2481
        • Hedley D.W
        • Chow S
        Evaluation of methods for measuring cellular glutathione content using flow cytometry.
        Cytometry. 1994; 15: 349
        • Lagneaux L
        • Marie J.P
        • Delforge A
        • et al.
        Comparison of in vitro inhibition of etoposide (VP16) on leukemic and normal myeloid, erythroid clonogenic cells.
        Exp Hematol. 1989; 17: 843
        • Vermes I
        • Haanen C
        • Steffens-Nakken H
        • Reutelingsperger C
        A novel assay for apoptosis flow cytometric detection of phophatidylserine expression on early apoptotic cells using fluorescein labeled annexin V.
        J Immunol Methods. 1995; 184: 39
        • Castedo M
        • Macho A
        • Zamzami N.N
        • et al.
        Mitochondrial perturbations define lymphocytes undergoing apoptotic depletion in vivo.
        Eur J Immunol. 1995; 25: 3277
        • Ibrado A.M
        • Huang Y
        • Fang G
        • Liu L
        • Bhalla K
        Overexpression of bcl-2 or bcl-xL inhibits Ara-C-induced CPP32/Yama protease activity and apoptosis of human acute myelogenous leukemia HL-60 cells.
        Cancer Res. 1996; 56: 4743
        • Lazebnik Y.A
        • Kaufmann S.H
        • Desnoyers S
        • Poirier G.G
        • Earnshaw W.C
        Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE.
        Nature. 1994; 371: 346
        • Lacombe F
        • Durrieu F
        • Briais A
        • et al.
        Flow cytometry CD45 gating for immunophenotyping of acute myeloid leukemia.
        Leukemia. 1997; 11: 1878
        • Kulms D
        • Schwarz T
        Molecular mechanisms of UV-induced apoptosis.
        Photodermatol Photoimmunol Photomed. 2000; 16: 195
        • Belyaev I.Y
        • Czene S
        • Harms-Ringdahl M
        Changes in chromatin conformation during radiation-induced apoptosis in human lymphocytes.
        Radiat Res. 2001; 156: 355
        • Carloni M
        • Meschini R
        • Ovidi L
        • Palitti F
        PHA-induced cell proliferation rescues human peripheral blood lymphocytes from X-ray-induced apoptosis.
        Mutagenesis. 2001; 16: 115
        • Arends M.J
        • Morris R.G
        • Wyllie A.H
        Apoptosis.
        Am J Pathol. 1990; 136: 593
        • Alter A
        • Roznszajn L.A
        • Miller H.I
        • Rosenschein U
        Ultrasound inhibits the adhesion and migration of smooth muscle cells in vitro.
        Ultrasound Med Biol. 1998; 24: 711
        • Yumita N
        • Nishigaki R
        • Umemura K
        • et al.
        Sonochemical activation of hematoporphyrin.
        Radiation Res. 1994; 138: 171
        • Ashush H
        • Leon A
        • Rozenszajn
        • et al.
        Apoptosis induction of human myeloid leukemic cells by ultrasound exposure.
        Cancer Res. 2000; 60: 1014
        • Yu Y.Q
        • Giocanti N
        • Averbeck D
        • Megnin-Chanet F
        • Favaudon V
        Radiation-induced arrest of cells in G2 phase elicits hypersensitivity to DNA double-strand break inducers and an altered pattern of DNA cleavage upon re-irradiation.
        Int J Radiat Biol. 2000; 76: 901
        • Kessel D
        • Luo Y
        Photodynamic therapy.
        Cell Death Differ. 1999; 6: 28
        • He J
        • Whitacre C.M
        • Xue L.Y
        • Berger N.A
        • Oleinick N.L
        Protease activation and cleavage of poly(ADP-ribose) polymerase.
        Cancer Res. 1998; 58: 940
        • Macho A
        • Hirsch T
        • Marzo I
        • et al.
        Glutathione depletion is an early and calcium elevation is a late event of thymocyte apoptosis.
        J Immunol. 1997; 158: 4612
        • Sharman W.M
        • Allen C.M
        • Van Lier J.E
        Role of activated oxygen species in photodynamic therapy.
        Methods Enzymol. 2000; 319: 376
        • Miyoshi N
        • Igarashi T
        • Riesz P
        Evidence against singlet oxygen formation by sonolysis of aqueous oxygen-saturated solutions of Hematoporphyrin and Rose Bengal. The mechanism of sonodynamic therapy.
        Ultrason Sonochem. 2000; 3S: 187
        • Misik V
        • Riesz P
        Peroxyl radical formation in aqueous solutions of N,N-dimethylformamide and N-methylformamide and dimethylsulfoxide by ultrasound.
        Free Radic Biol Med. 1996; 20: 129
        • Jeffers R
        • Feng R
        • Fowlkes J
        • Hunt J
        • Kessel D
        • Cain C
        Dimethylformamide as enhancer of cavitation-induced cell lysis in vitro.
        J Accoust Soc Am. 1995; 27: 669
        • Kessel D
        • Luo Y
        • Deng Y
        • Chang C.K
        The role of subcellular localization in initiation of apoptosis by photodynamic therapy.
        Photochem Photobiol. 1997; 65: 422
        • Giese A
        Photosensitization of organisms with special reference to natural photosensitizers.
        in: Hillenkampf F Patresi R Sacchi C Lasers in Biology and Medicine. Plenum, New York1980 (299p)
        • Duez P
        • Hanocq M
        • Dubois J
        Photodynamic DNA damage mediated by δ-aminolevulinic acid-induced porphyrins.
        Carcinogenesis. 2001; 22: 771
        • Moan J
        • Berg K
        • Kvam E
        • et al.
        Intracellular localization of photosensitizers.
        Ciba Found Symp. 1989; 146: 95